A Peer Support Service Integrated Into the 988 Lifeline

A peer support option should be integrated into the 988 Suicide and Crisis Lifeline so that 988 service users can choose to connect with specialists based on a shared lived experience. As people and communities become more siloed, the risk of mortality and morbidity increases. Social connectedness is a critical protective factor. A peer support service would allow individuals to receive support built on a lived experience that is common to both the service user and the specialist. It should be free and easily accessible through phone call and text messaging. This service is especially timely, following the 2020 rollout of the 988 Suicide and Crisis Lifeline, as well as recent peer support initiatives at the federal and state levels.

While the efficacy of peer support is most known with mental illness, it has successfully helped a range of individuals including cancer patients, people experiencing homelessness, racial minorities, veterans, and formerly incarcerated people. The peer support service should provide support for all kinds of lived experience, including experiences with: disability resulting from poor physical or mental health, substance use, suicidal ideation, veteran-connected disability, financial insecurity, homelessness, domestic violence and family court, nonnuclear family structures or living alone, former incarceration, and belonging to racial or ethnic minority groups. The service should be a preventative intervention as well as complimentary assistance for those in recovery or treatment.

Challenge and Opportunity

The United States faces an “epidemic of loneliness and isolation.” While individuals from all backgrounds are afflicted, the most vulnerable and underserved members of society have suffered the most. A range of challenging circumstances cause this suffering, including poor physical or mental health, nontraditional living conditions, and historic inequality. Not having anyone who shares one’s lived experience is isolating. These challenging life circumstances take an emotional toll, and they become risk factors for a slew of physical and mental health conditions that increase morbidity and decrease life expectancy. 

In addition to the social costs, these outcomes have an increasingly devastating economic impact. Health care costs are projected to account for 20% of the U.S. economy by 2031. Psychiatric hospitalizations are steadily increasing, and a shortage of psychiatric inpatient beds is rampant.

Social connectedness alleviates physical and mental burdens. Peer support offers a cost-effective intervention for prevention and recovery. It reduces spending on both physical and mental illnesses, and it reduces psychiatric hospitalizations, saving an average of $4.76 for every $1 spent on peer support. In a New York City-based peer support program, service users saved about $2,000 per month in Medicaid costs and had 2.9 fewer hospitalizations per year. 

Telephone-based peer support has life-saving outcomes, too. Over telephone, peer support led to a 15% increase in women’s mammography screening rates, with the highest increase among women of low-to-middle income. Telephone-based peer support increased breastfeeding rates by 14% and reduced breastfeeding dissatisfaction by 10% among first-time mothers, and it led to a 10% change in diet among patients with heart disease. While peer support would not solve national crises of homelessness or rising healthcare costs, it would ease them by fostering community empowerment and self-reliance and reducing federal intervention.

In 2023, SAMHSA rolled out “National Model Standards for Peer Support Certification”. This guide provides recommendations for how each state can integrate its own “peer mental health workforce across all elements of the healthcare system.” In its current form, SAMHSA’s strategy targets lived experiences with substance use and mental health. A broader scope would assist and empower more underserved members of the community. Following the momentum of SAMHSA’s initiative, now is the most optimal time to integrate a peer support service into the 988 Lifeline.

Peer support exists in the U.S., but services are spread thin across private and nonprofit sectors. The Restoring Hope for Mental Health and Well-Being Act of 2022 authorized $1.7 trillion in funding until FY2027 for various health initiatives, including peer support mental health services. This funding relies on organizations and institutions to independently implement peer support services based on community needs. However, a fragmented system can result in underuse, limited accessibility, and a varied quality of service, with pockets of the United States lacking any service at all. It also poses privacy concerns about how individuals’ data are stored and used, as well as cybersecurity vulnerabilities with smaller organizations that may lack a robust security infrastructure.

A peer support service that is integrated into the 988 Lifeline would ensure that all Americans have equal access to a high-quality, confidential peer support network. The standardization of the 988 Lifeline is a prime example of successful implementation. Its transition from a 10-digit number to a three-digit dial led to a 33% average increase in in-state call volume over four months. Standardization shifted funding from primarily private and nonprofit initiatives and donations to stable public sector support. As a result, call pickup rates rose from 70% to 93%, and wait times dropped from 2 minutes 20 seconds to just 35 seconds.

The 988 Lifeline also adheres to privacy and confidentiality protocols in line with the Health Insurance Portability and Accountability Act (HIPAA) Security Rule. Under these protocols, the 988 Lifeline retains minimal information about callers and texters, and this information stays private and securely stored. A peer support service with similar safeguards would help both service users and peer support workers (PSWs) feel safe when sharing sensitive information.

Peer support for mental health has gained traction recently. The National Alliance for Mental Illness (NAMI) offers peer-to-peer courses, although these are currently limited in their duration and available locations. Last year, Congress expressed an interest in peer support mental health services: a “Supporting All Students Act” was proposed in the Senate for “peer and school-based mental health support.” This endeavor especially targets the escalating suicide crisis among youths in the form of a peer-to-peer suicide prevention. The Wisconsin Department of Public Instruction also recently introduced Peer-to-Peer Suicide Prevention Grants. The federal and state-level governments clearly recognize the value of peer support work. Still, they underutilize its potential.

Loneliness and isolation are at an all-time high in the United States. However, they stem from a multitude of causes. An empathic connection with a peer who has lived the same experience has immense potential for healing and recovery in both the PSW and the service user. The 988 Lifeline should include an integrated peer support service for both mental health and a broad range of lived experiences. While there exist some peer support services for conditions not related to mental health, these are not standardized or easily accessible to the entire American population. Many peer-operated warmlines (i.e., support lines) exist. However, they are spread across dozens of phone numbers and websites with varied and limited sources of funding. A single peer support service integrated into the 988 Lifeline would make peer support universally available. This service would address the emotional toll that accompanies a wide range of challenging life circumstances.

In short, a peer support service integrated into the 988 Lifeline would make the efforts of existing peer support organizations even more effective as it would:

Action Plan

The peer support service should begin by covering only a few kinds of lived experiences. After this implementation, the service should be expanded to cover a broader range of experiences.

Recommendation 1. Create a Peer Support Task Force (PSTF).

A PSTF should be established within SAMHSA. The secretary of the U.S. Department of Health and Human Services (HHS) should work with SAMHSA’s assistant secretary for mental health and substance use to establish this temporary task force.

The PSTF should lead the implementation of the peer support service, acting as an interagency task force that coordinates with partners across the public, private, and nonprofit sectors. This partnership will ensure that different lived experiences are accounted for and that existing resources are used effectively. The PSTF should collaborate with federal agencies, including the Department of Veterans Affairs (VA) and the Indian Health Service (IHS), as well as with advocacy organizations like NAMI and the American Cancer Society that champion the needs of people with specific lived experiences.

The PSTF should be charged with the following recommendations to start.

Recommendation 2. Integrate a peer support option into the 988 Lifeline.

Under the authority of the PSTF, integrating a peer support option into the 988 Lifeline could bypass additional Congressional action.

In the first pass, the integrated peer support option should cover only a few kinds of lived experiences (e.g., suicide and behavioral crises, veteran-connected disability).

  1. Before scaling the peer support service across the United States, the PSTF should implement a pilot program to test and refine protocols. The PSTF should select 988 Lifeline centers in states that already have a strong peer support program, and it should pilot peer support call and text services. The pilot program should select a few peer support needs to test first (e.g., suicide and behavioral crises, veteran-connected disability), as a narrow scope will be easier to assess for the first pass.
  2. The PSTF should then incorporate feedback from this pilot program as it scales up and integrates a peer support option into the nationwide 988 Lifeline.
  3. The PSTF should coordinate with the Federal Communications Commission (FCC) to add a caller menu option for peer support into the 988 Lifeline. (E.g., “Press 4 for peer support.”)
  4. The PSTF should coordinate with the FCC and U.S. Digital Service to implement a telephone triage, so the service user can submit a request for a specific kind of peer support and then be routed to a PSW on shift who has a lived experience that best matches the submitted request.
  5. The PSTF should facilitate the 988 Lifeline’s partnership with existing local and national peer support organizations and warmlines. Partnering with these existing organizations would bolster the 988 Lifeline’s capacity to provide peer support. Furthermore, states that already have a strong peer support certification in place could easily integrate their trained PSWs into the 988 Lifeline services using their existing infrastructure and expertise.
  6. The PSTF should coordinate with the FCC to incorporate a peer support text messaging option within the 988 Lifeline’s text and chat services. (For example, service users could text “PEER” to 9-8-8 and be routed to a PSW on shift.)
  7. The PSTF should implement a public campaign to the general public clarifying that the 988 Lifeline will remain as a suicide and crisis hotline, but it will also provide access to broader peer support services. The available peer support services should be clearly outlined on the 988 Lifeline website.

PSWs should work in call centers alongside 988 Lifeline phone and text specialists. 

Recommendation 3. Develop an action plan to fund and sustain the integration of the peer support service.

The peer support service would need funding to become integrated into the 988 Lifeline. The government has recently approved funds for mental health initiatives, such as with the Restoring Hope for Mental Health and Well-Being Act of 2022. In 2023, HHS announced an additional $200 million in funding for the 988 Lifeline, but overall, HHS has granted almost $1.5 billion in total toward the 988 Lifeline. Similar avenues of funding should help jumpstart the integration of the peer support service.

For continued maintenance, fees for the peer support service should apply in a similar fashion to 9-1-1. That is, service users should not pay each time they access the 988 Lifeline or its integrated peer support service. Some U.S. states have already passed 9-8-8 implementation legislation that allows a monthly flat fee to be collected through telephone and wireless service providers, as permitted in the “National Suicide Hotline Designation Act of 2020”. The remaining states should be encouraged to pass similar legislation. The fee should remain in an account that is spent only for the maintenance of the 988 Lifeline and peer support services. If a fee is collected, then the FCC should provide an annual report on these fees and their usage.

Funding for the peer support service’s integration into the 988 Lifeline would entail the following:

Recommendation 4. Establish state-level standardized peer support training and certification. 

PSWs should learn skills that are commonly taught to 988 Lifeline specialists, including active listening and recovery-oriented language. They should learn how to share their own stories, navigate challenging conversations, maintain boundaries, and self-care. Their training should be standardized at the state level, and it should be based on the “National Model Standards for Peer Support Certification” established by SAMHSA. 

While most U.S. states have some version of peer support certification for mental health and/or substance use recovery, the PSTF should work with advocacy organizations to ensure that the state-level standardized training accounts for the peer support needs and demographics of each state. The peer support training should address the needs of people with a diverse range of lived experiences. Peer support needs can also be studied using caller outcome data that are recorded by 988 Lifeline specialists.

The PSTF should recruit trainers who will train and mentor the PSWs. To start, the PSTF should liaise within SAMHSA and with existing peer support organizations to coordinate this recruitment.

Recommendation 5. Establish a nation-wide system for the recruitment and training of PSWs.

Recruit individuals who would like to use their own lived experience to help other community members. Recruitment should occur through hospitals, clinics, as well as peer-run and community-based organizations to maximize recruitment pathways. 

Current 988 Lifeline phone and text specialists could also be good candidates for the peer support service. Many are motivated by their own lived experiences and are already trained to handle calls across multiple helplines. 988 Lifeline specialists are instructed to focus on the service user seeking help and not share about themselves. However, specialists’ own lived experiences––if they are comfortable sharing them––represent untapped potential.

Once recruited, individuals who are training to become PSWs should attend live training sessions online with a peer cohort.

Recommendation 6. Expand the peer support service to cover a diverse range of lived experiences.

After successfully establishing the peer support service, it should be expanded upon to provide support for a broad and diverse range of lived experiences. The available peer support services should continue to be updated and outlined on the 988 Lifeline website.

Conclusion

A peer support service should be integrated into the 988 Lifeline. A service that caters to all kinds of lived experience paves way for a more empowered and self-reliant community. It fosters a societal mindset of helping each other based on shared lived experience in an empathic and healthy way. Peer support empowers both the PSW and the service user. While the service user finds empathy and understanding, the PSW finds a renewed sense of purpose and confidence. In this way, peer support would alleviate loneliness and isolation that result from a variety of causes while instilling longer-term resilience into the community.

This action-ready policy memo is part of Day One 2025 — our effort to bring forward bold policy ideas, grounded in science and evidence, that can tackle the country’s biggest challenges and bring us closer to the prosperous, equitable and safe future that we all hope for whoever takes office in 2025 and beyond.

PLEASE NOTE (February 2025): Since publication several government websites have been taken offline. We apologize for any broken links to once accessible public data.

Frequently Asked Questions
What is a service user?

A service user refers to any individual who uses the peer support service for assistance.

What if peer support workers (PSWs) become distressed or emotionally fatigued?

Peer support work can be empowering and healing for those in recovery. However, as with any emotionally challenging work, PSWs benefit from ongoing support and supervision. Some interactions can be draining, especially if they hit too close to home. It is common for 988 Lifeline call centers to have a trained staff member, such as a psychologist or therapist, on call for such situations. A PSW could use this resource to debrief or process as needed. Additionally, during recruitment, PSW candidates should be screened to ensure they are comfortable speaking about their own lived experience and helping others who are going through the same experience.

How would PSWs be compensated?

Most PSWs would be compensated in the same way as their 988 Lifeline call and text specialist counterparts. They would be compensated with rigorous ongoing training, support, and resources for recovery and self-care in return for their time. They would also benefit from a social support system of fellow PSWs. Some individuals who may be ideal candidates for peer support work may struggle to find employment and health insurance. They will be less likely to volunteer if they do not have a living wage. To maximize the range of lived experiences available, certain individuals should be eligible for financial compensation.

For the peer support phone and text service, how is a caller or texter connected with a PSW who has the same lived experience?

A service user’s phone call or text should be routed to a 988 Lifeline peer support operator or telephone triage, where the service user is asked to say aloud (or type) what kind of support they are searching for (e.g., “I want to speak with someone who has been in prison.”) Then, the service user’s call (or text) would be routed to a PSW (under an alias) anywhere in the United States who 1) is currently on shift and 2) has a lived experience that is closest to the service user’s request. Each PSW would have associated keywords, such as “former incarceration,” “PTSD,” or “living alone,” which means that they are trained to connect with any service user about those lived experiences. The PSW would follow the service user’s lead in the conversation, and the PSW could share parts of their own lived experience when appropriate.

Why should the peer support service include a text messaging option?

Text messaging can be more accessible than a phone line for youths and people with disabilities. The 988 Lifeline includes a text messaging option for this same reason.

Why should we integrate another option into the 988 Lifeline when some states still struggle to find sustainable ways to fund and staff the 988 Lifeline as it currently is?

It is true that both the regular 988 Lifeline and peer support service would provide resources and emotional support. By incorporating peer support into the 988 Lifeline, existing local and national peer support organizations would be eligible to partner with the 988 Lifeline and bolster its capacity to provide peer support. This endeavor will help with some staffing concerns. Furthermore, the peer support service would cover a diverse range of lived experiences extending beyond mental health. Therefore, more individuals may be motivated to join the 988 Lifeline staff to share their unique lived experience and help others who feel the same way.


Integrating a versatile peer support service into the 988 Lifeline transforms the latter into a service that every single American can use. (Every American has some unique lived experience.) The service’s versatility may help incentivize the remaining states to pass legislation to collect a 988 Lifeline fee through telephone and wireless service providers.


Finally, peer support is a cost-effective, preventative intervention. It should help remove the burden on other federal services, thereby reducing overall spending in time.

What happens next?

After implementing the action plan outlined in this memo, a subsequent memo should outline how to integrate peer support work into the community in person and on a large scale. Incorporating peer support into the 988 Lifeline would show its effectiveness to the public, policymakers, and healthcare professionals. This credibility would bolster endeavors to integrate peer support work into community settings throughout the country (e.g., behavioral health centers, hospitals and emergency rooms, and community clinics). These endeavors could also lead to professionalizing peer support, so PSWs can be reimbursed through Medicaid programs and health insurance.

Antitrust in the AI Era: Strengthening Enforcement Against Emerging Anticompetitive Behavior

The advent of artificial intelligence (AI) has revolutionized business practices, enabling companies to process vast amounts of data and automate complex tasks in ways previously unimaginable. However, while AI has gained much praise for its capabilities, it has also raised various antitrust concerns. Among the most pressing is the potential for AI to be used in an anticompetitive manner. This includes algorithms that facilitate price-fixing, predatory pricing, and discriminatory pricing (harming the consumer market), as well as those which enable the manipulation of wages and worker mobility (harming the labor market). More troubling perhaps is the fact that the overwhelming majority of the AI landscape is controlled by just a few market players. These tech giants—some of the world’s most powerful corporations—have established a near-monopoly over the development and deployment of AI. Their dominance over necessary infrastructure and resources makes it increasingly challenging for smaller firms to compete.

While the antitrust enforcement agencies—the FTC and DOJ—have recently begun to investigate these issues, they are likely only scratching the surface. The covert and complex nature of AI makes it difficult to detect when it is being used in an anticompetitive manner. To ensure that business practices remain competitive in the era of AI, the enforcement agencies must be adequately equipped with the appropriate strategies and resources. The best way to achieve this is to (1) require the disclosure of AI technologies during the merger-review process and (2) reinforce the enforcement agencies’ technical strategy in assessing and mitigating anticompetitive AI practices.

Challenge & Opportunity

Since the late 1970s, antitrust enforcement has been in decline, in part due to a more relaxed antitrust approach put forth by the Chicago school of economics. Both the budgets and the number of full-time employees at the enforcement agencies have steadily decreased, while the volume of permitted mergers and acquisitions has risen (see Figure 1). This resource gap has limited the ability of the agencies to effectively oversee and regulate anticompetitive practices.

Figure 1. Merger Enforcement vs. Total Filings

Source: Department of Justice Antitrust Division Operations

Changing attitudes surrounding big business, as well as recent shifts in leadership at the enforcement agencies—most notably President Biden’s appointment of Lina Khan to FTC Chair—have signaled a more aggressive approach to antitrust law. But even with this renewed focus, the agencies are still not operating at their full potential. 

This landscape provides a significant opportunity to make some much-needed changes. Two areas for improvement stand out. First, agencies can make use of the merger review process to aid in the detection of anticompetitive AI practices. In particular, the agencies should be on the look-out for algorithms that facilitate price-fixing, where competitors use AI to monitor and adjust prices automatically, covertly allowing for tacit collusion; predatory pricing algorithms, which enable firms to undercut competitors only to later raise prices once dominance is achieved; and dynamic pricing algorithms, which allow firms to discriminate against different consumer groups, resulting in price disparities that may distort market competition. On the labor side, agencies should screen for wage-fixing algorithms and other data-driven hiring practices that may suppress wages and limit job mobility. Requiring companies to disclose the use of such AI technologies during merger assessments would allow regulators to examine and identify problematic practices early on. This is especially useful for flagging companies with a history of anticompetitive behavior or those involved in large transactions, where the use of AI could have the strongest anticompetitive effects.

Second, agencies can use AI to combat AI. Research has demonstrated that AI can be more effective in detecting anticompetitive behavior than other traditional methods. Leveraging such technology could transform enforcement capabilities by allowing agencies to cover more ground despite limited resources. While increasing funding for these agencies would be requisite, AI nonetheless provides a cost-effective solution, enhancing efficiency in detecting anticompetitive practices, without requiring massive budget increases.

The success of these recommendations hinges on the enforcement agencies employing technologists who have a deep understanding of AI. Their knowledge on algorithm functionality, the latest insights in AI, and the interplay between big data and anticompetitive behavior is instrumental. A detailed discussion of the need for AI expertise is covered in the following section.

Plan Of Action

Recommendation 1. Require Disclosure of AI Technologies During Merger-Review.

Currently, there is no formal requirement in the merger review process that mandates the reporting of AI technologies. This lack of transparency allows companies to withhold critical information that may help agencies determine potential anticompetitive effects. To effectively safeguard competition, it is essential that the FTC and DOJ have full visibility of businesses’ technologies, particularly those that may impact market dynamics. While the agencies can request information on certain technologies further in the review process, typically during the second request phase, a formalized reporting requirement would provide a more proactive approach. Such an approach would be beneficial for several reasons. First, it would enable the agencies to identify anticompetitive technologies they might have otherwise overlooked. Second, an early assessment would allow the agencies to detect and mitigate risk upfront, rather than having to address it post-merger or further along in the merger review process, when remedies may be more difficult to enforce. This is particularly applicable with regard to deep integrations that often occur between digital products post-merger. For instance, the merger of Instagram and Facebook complicated the FTC’s subsequent efforts to challenge Meta. As Dmitry Borodaenko, a former Facebook engineer, explained: 

“Instagram is no longer viable outside of Facebook’s infrastructure. Over the course of six years, they integrated deeply… Undoing this would not be a simple task—it would take years, not just the click of a button.”

Lastly, given the rapidly evolving nature of AI, this requirement would help the agencies identify trends and better determine which technologies are harmful to competition, under what circumstances, and in which industries. Insights gained from one sector could inform investigations in other sectors, where similar technologies are being deployed. For example, the DOJ recently filed suit against RealPage, a property management software company, for allegedly using price-fixing algorithms to coordinate rent increases among competing landlords. The case is the first of its kind, as there had not been any previous lawsuit addressing price-fixing in the rental market. With this insight, however, if the agencies detect similar algorithms during the merger review process, they would be better equipped to intervene and prevent such practices.

There are several ways the government could implement this recommendation. To start, The FTC and DOJ should issue interpretive guidelines specifying that anticompetitive effects stemming from AI technologies are within the purview of the Hart-Scott-Rodino (HSR) Act, and that accordingly, such technologies should be disclosed in the pre-merger notification process. In particular, the agencies should instruct companies to report detailed descriptions of all AI technologies in use, how they might change post-merger, and their potential impact on competition metrics (e.g., price, market share). This would serve as a key step in signaling to companies that AI considerations are integral during merger review. Building on this, Congress could pass legislation mandating AI disclosures, thereby formalizing the requirement. Ultimately, in a future round of HSR revisions, the agencies could incorporate this mandate as a binding rule within the pre-merger framework. To avoid unnecessary burden on businesses, reporting should only be required when AI plays a significant role in the company’s operations or is expected to post-merger. What constitutes a ‘significant role’ should be left to the discretion of the agencies but could include AI systems central to core functions such as pricing, customer targeting, wage-setting, or automation of critical processes.

Recommendation 2. Reinforce the FTC and DOJ’s Technical Strategy in Assessing and Mitigating Anticompetitive AI Practices.

Strengthening the agencies’ ability to address AI requires two actions: integrating computational antitrust strategies and increasing technical expertise. A wave of recent research has highlighted AI as a powerful tool in helping detect anticompetitive behavior. For instance, scholars at the Stanford Computational Antitrust Project have demonstrated that methods such as machine learning, natural language processing, and network analysis can assist with tasks, ranging from uncovering collusion between firms to distinguishing digital markets. While the DOJ has already partnered with the Project, the FTC could benefit by pursuing a similar collaboration. More broadly, the agencies should deepen their technical expertise by expanding workshops and training with AI academic leaders. Doing so would not only provide them with access to the most sophisticated techniques in the field, but would also help bridge the gap between academic research and real-world implementation. Examples may include the use of machine learning algorithms to identify price-fixing and wage-setting; sentiment analysis, topic modeling, and other natural language processing tools to detect intention to collude in firm communications; or reverse-engineering algorithms to predict outcomes of AI-driven market manipulation. 

Leveraging such computational strategies would enable regulators to analyze complex market data more effectively, enhancing the efficiency and precision of antitrust investigations. Given AI’s immense power, only a small—but highly skilled—team is needed to make significant progress. For instance, the UK’s Competition and Markets Authority (CMA) recently stood up a Data, Technology and Analytics unit, whereby they implement machine learning strategies to investigate various antitrust matters. For the U.S. agencies to facilitate this, the DOJ and FTC should hire more ML/AI experts, data scientists, and technologists, who could serve several key functions. First, they could conduct research on the most effective methods for detecting collusion and anticompetitive behavior in both digital and non-digital markets. Second, based on such research, they could guide the implementation of selected AI solutions in investigations and policy development. Third, they could perform assessments of AI technologies, evaluating the potential risks and benefits of AI applications in specific markets and companies. These assessments would be particularly useful during merger review, as previously discussed in Recommendation 1. Finally, they could help establish guidelines for transparency and accountability, ensuring the responsible and ethical use of AI both within the agencies and across the markets they regulate.

To formalize this recommendation, the President should submit a budget proposal to Congress requesting increased funding for the FTC and DOJ to (1) hire technology/AI experts and (2) provide necessary training for other selected employees on AI algorithms and datasets. The FTC may separately consider using its 6(b) subpoena powers to conduct a comprehensive study of the AI industry or of the use of AI practices more generally (e.g., to set prices or wages). Finally, the agencies should strive to foster collaboration between each other (e.g., establishing a Joint DOJ-FTC Computational Task Force), as well as with those in academia and the private sector, ​​to ensure that enforcement strategies remain at the cutting edge of AI advancements.

Conclusion

The nation is in the midst of an AI revolution, and with it comes new avenues for anticompetitive behavior. As it stands, the antitrust enforcement agencies lack the necessary tools to adequately address this growing threat.

However, this environment also presents a pivotal opportunity for modernization. By requiring the disclosure of AI technologies during the merger review process, and by reinforcing the technical strategy at the FTC and DOJ, the antitrust agencies can strengthen their ability to detect and prevent anticompetitive practices. Leveraging the expertise of technologists in enforcement efforts can enhance the agencies’ capacity to monitor levels of competition in markets, as well as allow them to identify patterns between certain technologies and violations of antitrust.

Given the rapid pace of AI advancement, a proactive effort triumphs over a reactive one. Detecting antitrust violations early allows agencies to save both time and resources. To protect consumers, workers, and the economy more broadly, it is imperative that the FTC and DOJ adapt their enforcement strategies to meet the complexities of the AI era.

This action-ready policy memo is part of Day One 2025 — our effort to bring forward bold policy ideas, grounded in science and evidence, that can tackle the country’s biggest challenges and bring us closer to the prosperous, equitable and safe future that we all hope for whoever takes office in 2025 and beyond.

PLEASE NOTE (February 2025): Since publication several government websites have been taken offline. We apologize for any broken links to once accessible public data.

Using Pull Finance for Market-driven Infrastructure and Asset Resilience

The incoming administration should establish a $500 million pull-financing facility to ensure infrastructure and asset resiliency with partner nations by catalyzing the private sector to develop cutting-edge technologies. The increasing frequency of extreme weather events, which caused over $200 billion in global economic losses in 2023, is disrupting global supply chains and exacerbating migration pressures, particularly for the U.S. Investing in climate resilience abroad offers a significant opportunity for U.S. businesses in technology, engineering, and infrastructure, while also supporting job creation at home. 

Pull-finance mechanisms can maximize the efficiency and impact of U.S. investments, fostering innovation and driving sustainable solutions to address global vulnerabilities. Unlike traditional funding which second-guesses the markets by supporting only selected innovators, pull financing drives results by relying on the market to efficiently allocate resources to achievement, fostering competition and rewarding the most impactful solutions. Managed and steered by the U.S. government, the pull-financing facility would fund infrastructure and asset resiliency results delivered by the world’s cutting-edge innovators, mitigating the effects of extreme weather events and ultimately supporting U.S. interests abroad. 

Challenge and Opportunity 

The increasing frequency and severity of extreme weather events pose significant risks to global economic stability, with direct implications for U.S. interests. In 2023 alone, natural disasters caused over $200 billion in global economic losses with much of the damage concentrated in regions critical to global supply chains. U.S. businesses that depend on these supply chains face rising costs and disruptions, which translate into higher costs for U.S. businesses and consumers, undermining economic competitiveness.

Beyond the economic dimension, these vulnerabilities exacerbate socio-political pressures. Climate-induced displacement is accelerating, with 32.6 million people internally displaced by disasters in 2022. Most displaced individuals that cross borders migrate to countries neighboring their own, which are ill-equipped to handle the influx, often further destabilizing fragile states. For the U.S., this translates into increased migration pressures at its southern border, where natural disasters are already a driving force behind migration from Central America. Addressing these root causes through proactive resiliency investments abroad would reduce long-term strain on the U.S. and bolster stability in strategically important regions.

In addition to economic and social risks, resilience is now a key front in global competition. The People’s Republic of China has rapidly expanded its influence in developing nations through initiatives like the Belt and Road, financing over $200 billion in energy and infrastructure projects since 2013. A significant portion of these projects focus on resiliency investments, enabling China to position itself as a partner of choice for nations with asset and infrastructure exposure. This growing influence comes at the expense of U.S. global leadership.

In the context of these challenges, it is especially concerning that much of the U.S.’s existing spending may not be achieving the results it could. A recent audit of USAID climate initiatives highlights concerns around limited transparency and effectiveness in its development funding. The inefficient use of this funding is leaving opportunities on the table for U.S. businesses and workers. Global investments in adaptation and resiliency are projected to reach $500 billion annually by 2050. Resilience projects abroad could open substantial markets for American engineering, technology, and infrastructure firms. For instance, U.S.-based companies specializing in resilient agriculture, flood defense systems, advanced irrigation technologies, and energy infrastructure stand to benefit from increased demand. Domestically, the manufacturing and export of these solutions could generate significant economic activity, supporting high-quality jobs and revitalizing industrial sectors.

Pull finance presents an opportunity to increase the cost effectiveness of resiliency funding—and ensure this funding achieves U.S. interests. Pull finance mechanisms like results-based financing and Advance Market Commitments (AMC) reward successful solutions that meet specific criteria, promoting private sector engagement and market-driven problem-solving. Unlike traditional “push” financing, which funds chosen teams or projects directly, pull financing sets a goal and allows any innovator who reaches it to claim the reward, fostering      competitive problem-solving without pre-selected winners. This approach includes various mechanisms – such as prize challenges, milestone payments, advance market commitments, and subscription models – each suited to different issues and industries.

Pull financing is particularly effective for addressing complex challenges with unclear or emerging solutions, or in areas with limited commercial incentives. It has proven successful in various contexts, such as the first Trump Administration’s rapid development of COVID-19 vaccines through Operation Warp Speed and GAVI’s introduction of the pneumococcal vaccine in low-income countries. These initiatives highlight how pull financing can stimulate breakthrough innovations that efficiently address immediate needs in collaboration with private actors through effective incentives.  

Pull finance can be used to efficiently advance infrastructure and asset resilience goals while also providing opportunities for U.S. innovators and industry. By stimulating demand for critically needed technologies for development like resilient seeds and energy storage solutions, as detailed in Box 1, well-designed pull finance would help link U.S. technology innovators to addressing needs of U.S. partners. As such, pull finance can play a critical role in positioning the U.S. as a partner of first choice for countries seeking to access U.S. innovation to meet resilience needs.

What would the design of a pull financing mechanism look like in practice?  

Resilient Seeds

Agriculture in Africa is highly susceptible to extreme weather events, with limited adoption of effective farming technologies. Developing new seed varieties capable of withstanding these events and optimizing resource use has the potential to yield significant societal benefits.

While push financing can support the development of resource-efficient and productive seeds, it often lacks the ability to ensure they meet essential quality standards, like flavor and appearance, and are user-friendly across farming, transport and marketing stages. In contrast, pull financing can effectively incentivize private sector innovation across all critical dimensions, including end-user take-up. 

A pull mechanism for resilient seeds, using a milestone payment mechanism, could cover a portion of R&D costs initially, with additional payments tied to successful lab trials. Depending on the obstacles to scaling – whether they arise from the innovator/distributor side or the farmer side – a small per-user payment to the innovator or per-user subsidy could help sustain market demand.

The design and scale of a pull financing mechanism to promote the rollout of new seeds and crop varieties will largely depend on the market readiness of the various seed types involved. Establishing effective pull mechanisms for seed development is estimated to cost between $50 million and $100 million, aiming for significant outreach to farmers. Along with supporting improved livelihoods for farmers, this small investment would open opportunities for U.S. technology innovators and companies. 

Pull Finance Initiative for Infrastructure and Asset Resiliency in the Caribbean

The Caribbean is one of the regions most vulnerable to extreme weather events, making it critical to engage the private sector in developing and adopting technologies suited to Small Island Developing States (SIDS). Challenges such as limited demand and high costs hinder innovation and investment in these small markets, leaving key areas like agriculture and access underserved. Overcoming these market failures requires innovative approaches to create sustainable incentives for private sector involvement.

Pull finances offers a promising solution to drive resiliency in SIDS. By tying payments to measurable outcomes, this approach will incentivize the development and deployment of technologies that might otherwise remain inaccessible. 

For example, pull finance could be used to stimulate the creation of energy storage solutions designed to withstand extreme weather conditions in remote areas. This could be help address the critical needs of SIDS’ such as Guyana which face energy security challenges linked to extreme weather conditions, especially in remote and dispersed areas. Energy storage technologies exist, but companies are not motivated to invest in tailored innovation for local needs because end-users cannot pay prices that compensate for innovation efforts. Pull finance could address this by committing to purchase an amount large enough that nudges companies to develop a tailored product, without raising market prices. Success would require partnerships with local SMEs, caps in installation costs, and specifications on storage capacity, along with relevant technology partners such as those in the U.S.. This approach would support immediate adaptation needs and lay the foundation for sustainable, market-driven solutions that ensure long-term resilience for SIDS.

Plan of Action 

The new administration should establish a dedicated pull-financing facility to accelerate the scale-up and deployment of development solutions with partner nations. In line with other major U.S. climate initiatives, this facility could be managed by USAID’s Bureau for Resilience, Environment and Food Security (REFS), with significant support from USAID’s Innovation, Technology, and Research (ITR) Hub, in partnership with the U.S. Department of State. By leveraging USAID’s deep expertise in development and SPEC’s strategic diplomacy, this collaboration would ensure the facility addresses LMIC-specific needs while aligning with broader U.S. objectives.

The recent audit of USAID climate initiatives referenced above highlights concerns on the limited transparency and effectiveness in its climate funding. Thus, we recommend that USAID assesses the impact of its climate spending under the 2020-2024 administration and reallocates a portion of funds from less effective or stalled initiatives to this new facility. We recognize that it may be challenging to quickly identify $500 million in underperforming projects to close and reassign. Therefore, in addition to reallocating existing resources, we strongly recommend appealing to new funding for this initiative. This approach will ensure the new facility has the financial backing it needs to drive meaningful outcomes. Additional resources could also be sourced from large multilateral organizations such as the World Bank.

To enhance the facility’s impact, we recommend the active participation of agencies such as the National Oceanic and Atmospheric Administration (NOOA), particularly through the Climate and Societal Interactions Division (CSI)   in the Steering Committee,

We propose that this facility draw on the example of the UK’s planned Climate Innovation Pull Facility (CIPF), a £185 million fund which aims to fund development-relevant pull finance projects in LMICs such as those proposed by the Center for Global Development and Instiglio. This can be achieved through the following steps:

Recommendation 1. Establish the pull-finance facility, governance and administration with an initial tranche of $500 million. 

The initiative proposes establishing a pull-finance facility with an initial fund of $500 million. This facility will be overseen by a steering board chaired by USAID and comprising senior representatives from USAID, the State Department, NOOA , which will set the strategic direction and make final project selections. 

A facility management team, led by USAID, will be responsible for ensuring the successful implementation of the facility, including the selection and delivery of 8 to 16 projects. The final number of projects will depend on the launch readiness of prioritized technologies and their potential impact, with the selection process guided by criteria that align with the facility’s strategic goals. The facility management team will also be responsible for contracting with project and evaluation partners, compliance with regulations, risk management, monitoring and evaluation, as well as payouts. Additionally, the facility management team will provide incubation support for selected initiatives, including technical consultations, financial modeling, contracting expertise, and feasibility assessments.

Designing pull financing mechanisms is complex and requires input from specialized experts, including scientists, economists, and legal advisors, to identify suitable market gaps and targets. An independent Technical Advisory Group (TAG) led by USAID and comprised of such experts should be established to provide technical guidance and quality assurance. The TAG will identify priority resilience topics, such as reducing crop-residue burning or developing resilient crops. It will also focus on sectors where the U.S. can enhance its global competitiveness, which faces high upfront costs and risks. Additionally, the TAG will be responsible for technical review and recommendations of the shortlisted project proposals to inform final selection, as well as provide general advice and challenge to the facility management team and steering board.

We suggest starting with $500 million as the minimum required to be credible and relevant as well as responsive to the scale of global need. Further, experience shows that pull mechanisms need to be of sufficient scale to sustainably shift markets. For instance, GAVI’s pneumococcal vaccine AMC entailed a $1.5 billion commitment and Frontier’s carbon capture AMC likewise entails over $1 billion in commitments.  

Recommendation 2. Set up a performance management system to measure, assess and ensure impact.

The U.S. pull financing facility will implement a robust monitoring, evaluation, and learning (MEL) framework to track and enhance its impact and drive ongoing improvement through feedback and learning. 

The facility manager will develop a logical framework (logframe) that includes key performance indicators (KPIs) and a progress and risk dashboard to track monthly performance. These tools will enable effective monitoring of progress, assessment of impact, and proactive risk management, allowing for quick responses to unexpected challenges or underperformance.

Monthly check-ins with an independent evaluation partner, along with oversight from a dedicated MEL committee, will ensure consistent and rigorous evaluation as well as continuous learning. Additionally, knowledge management and dissemination activities will facilitate the sharing of insights and best practices across the program.

Recommendation 3. Establish a knowledge management hub to facilitate the sharing of results and insights and ensure coordination across pull-financing projects. 

The hub will work closely with community partners and stakeholders – such as industry and tech leaders and manufacturers – in areas like resiliency-focused finance and innovation to build strong support and develop resources on essential topics, including the effectiveness of pull financing and optimal design strategies. Additionally, the hub will promote collaboration across projects focused on similar technological and production advancements, generating synergies that enhance their collective impact and benefits.

Once the proof of concept is established through clear evidence and learning, the facility will likely secure further stakeholder buy-in and attract additional funding for a scale up phase covering a larger portfolio of projects. 

Conclusion 

The federal government should establish a $500 million pull-financing facility to accelerate technologies for resilience in the face of growing development challenges. This initiative will unlock high-return investments and increase cost effectiveness of resiliency spending, driving economic and geopolitical goals. Managed and steered by USAID and the State Department, with support from NOOA, the facility would foster breakthroughs in critical areas like resilient infrastructure, energy, and technology, benefiting both U.S. businesses and our international partners. By investing strategically, the U.S. can ensure both national and global stability.

The authors thank FAS for the reviews and feedback, along with Ranil Dissanayake, Florence Oberholtzer, and Laura Mejia Villada for their valuable contribution to this piece.

This action-ready policy memo is part of Day One 2025 — our effort to bring forward bold policy ideas, grounded in science and evidence, that can tackle the country’s biggest challenges and bring us closer to the prosperous, equitable and safe future that we all hope for whoever takes office in 2025 and beyond.

PLEASE NOTE (February 2025): Since publication several government websites have been taken offline. We apologize for any broken links to once accessible public data.

Frequently Asked Questions
What are some of the main constraints to pull financing and how can they be overcome?

Pull financing mechanisms, such as prize competitions, milestone payments, and Advanced Market Commitments (AMCs) often face regulatory and legal challenges due to their dependency on successful outcomes for funding disbursement (CGD, 2021; CGD, 2023). First, it can make cashflow management challenging as federal law requires that legally binding financial commitments be made if the necessary appropriated funds are available, resulting in upfront scoring of costs, even if the actual expenditures occur years later. The uncertainty surrounding innovation and payouts can also create risk aversion, as most funding accounts are not “no-year” accounts, meaning committed funds can expire if competition goals are unmet within the designated timeframe.


To mitigate these constraints, agencies can use budgetary workarounds like no-year appropriations, allowing them to reallocate de-obligated funds from canceled competitions to new initiatives. Other options include employing credit-type scoring to discount costs based on the likelihood of non-payment and making non-legally binding commitments backed by third parties, such as international institutions, to avoid these challenges altogether.

What is the expected timeline for the establishment and cost breakdown for this fund?

The entire fund is expected to span a maximum of five () years. The initial 12 months will concentrate on identifying eight (8) to 16 projects through comprehensive due diligence and providing incubation support. In the subsequent four (4) years, the focus will shift to project delivery.

How is this initiative different from existing U.S. initiatives such as USAID’s Climate Finance for Development Accelerator (CFDA)?

In contrast to the traditional push-funding approach of the CFDA program, our proposed pull-finance initiative introduces a unique market-shaping component aimed at driving key infrastructure and resilience solutions to fruition. In contrast to CFDA, pull finance addresses demand-side risks by providing demand-side guarantees of a future market for the technology or solution. It also mitigates R&D risk by combining incentives for research and development, ensuring that a viable market exists once the technology is developed. This approach helps accelerate market creation and innovation in high-risk, high-innovation sectors where demand or technological maturity is uncertain.

A National Guidance Platform for AI Acquisition

Streamlining the procurement process for more equitable, safe, and innovative government use of AI

The federal government’s approach to procuring AI systems serves two critical purposes: it not only shapes industry and academic standards but also determines how effectively AI can enhance public services. By leveraging its substantial purchasing power responsibly, the government can encourage high-quality, inclusive AI solutions that address diverse citizen needs while setting a strong precedent for innovation and accountability. Guidance issued in October 2024 by the White House’s Office of Management and Budget (OMB) gives recommendations on how agencies should use AI systems, focusing on public trust and data transparency. However, it is unclear how these guidelines align with general procurement regulations like the Federal Acquisition Regulation (FAR).

To reduce bureaucratic hurdles and encourage safe government innovation, the General Services Administration (GSA) should develop a digital platform that guides federal agencies through an “acquisition journey” for AI procurement. This recommendation is for streamlining guidance for procuring AI systems and should not be confused with the use of AI to simplify the procurement process.  The platform should be intuitive and easy to navigate by clearly outlining the necessary information, requirements, and resources at each process stage, helping users understand what they need at any point in the procurement lifecycle. Such a platform would help agencies safely procure and implement AI technologies while staying informed on the latest guidelines and adhering to existing federal procurement rules. GSA should take inspiration from Brazil’s well-regarded Public Procurement Platform for Innovation (CPIN). CPIN helps public servants navigate the procurement process by offering best practices, risk assessments, and contract guidance, ensuring transparency and fairness at each stage of the procurement process.

Challenges and Opportunities

The federal government’s approach to AI systems is a crucial societal benchmark, shaping standards that ripple through industries, academia, and public discourse. Along with shaping the market, the government also faces a delicate balancing act when it comes to its own use of AI: it must harness AI’s potential to dramatically enhance efficiency and effectiveness in public service delivery while simultaneously adhering to the highest AI safety and equity standards. As such, the government’s handling of AI technologies carries immense responsibility and opportunity. 

The U.S. federal government procures AI for numerous different tasks—from analyzing weather hazards and expediting benefits claims to processing veteran feedback. Positive impacts could potentially include faster and more accurate public services, cost savings, better resource allocation, improved decision-making based on data insights, and enhanced safety and security for citizens. However, risks can include privacy breaches, algorithmic bias leading to unfair treatment of certain groups, over-reliance on AI for critical decisions, lack of transparency in AI-driven processes, and cybersecurity vulnerabilities. These issues could erode public trust, inhibit the adoption of beneficial AI, and exacerbate existing social inequalities.

The federal government has recently published several guidelines on the acquisition and use of AI systems within the federal government, specifically how to identify and mitigate systems that may impact public trust in these systems. For example:

This guidance, coupled with the already extensive set of general procurement regulations such as the Federal Acquisition Regulation (FAR ), can be overwhelming for public servants. In conversations with the author of this memo, stakeholders, including agency personnel and vendors, frequently noted that they needed clarification about when impact and risk assessments should occur in the FAR process.

How can government agencies adequately follow their mandate to provide safe and trustworthy AI for public services while reducing the bureaucratic burden that can result in an aversion to government innovation? A compelling example comes from Brazil. The Public Procurement Platform for Innovation (CPIN), managed by the Brazilian Ministry of Development, Industry, Commerce, and Services (MDIC), is an open resource designed to share knowledge and best practices on public procurement for innovation. In 2023, the platform was recognized by the Federal Court of Auditors (TCU—the agency that oversees federal procurement) as an essential new asset in facilitating public service. The CPIN helps public servants navigate the procurement process by diagnosing needs and selecting suitable contracting methods through questionnaires. Then, it orients agencies through a procurement journey, identifying what procurement process should be used, what kinds of dialogue the agency should have with potential vendors and other stakeholders, guidance for risk assessments, and contract language. The platform is meant to guide public servants through each stage of the procurement process, ensuring they know their obligations for transparency, fairness, and risk mitigation at any given time. CPIN is open to the public and is meant to be a resource, not new requirements that supplant existing mandates by Brazilian authorities. 

Here in the U.S., the Office of Federal Procurement (OFFP) within the Office of Management and Budget (OMB) in partnership with the General Services Administration (GSA) and the Council of Chief AI Officers (CAIO), should develop a similar centralized resource to help federal agencies procure AI technologies safely and effectively. This platform would ensure agencies have up-to-date guidelines on AI acquisition integrated with existing procurement frameworks.

This approach is beneficial because:

GSA has created similar tools before. For example, the Generative AI Acquisition Resource Guide assists federal buyers in procuring and implementing generative AI technologies by describing key considerations, best practices, and potential challenges associated with acquiring generative AI solutions. However, this digital platform would go one step further and align best practices, recommendations, and other AI considerations within the processes outlined in the FAR and other procurement methods. 

Plan of Action

Recommendation 1. Establish a Working Group led by the OMB OFPP, with participation from GSA, OSTP, and the CAIO Council, tasked with systematically mapping all processes and policies influencing public sector AI procurement.

This includes direct AI-related guidance and tangential policies such as IT, data management, and cybersecurity regulations. The primary objective is identifying and addressing existing AI procurement guidance gaps, ensuring that the forthcoming platform can provide clear, actionable information to federal agencies. To achieve this, the working group should:

Conduct a thorough review of current mandates (see the FAQ for a non-exhaustive list of current mandates), executive orders, OMB guidance, and federal guidelines that pertain to AI procurement. This includes mapping out the requirements and obligations agencies must meet during acquisition. Evaluate if these mandates come with explicit deadlines or milestones that need to be integrated into the procurement timeline (e.g., AI risk assessments, ethics reviews, security checks)

Conduct a gap analysis to identify areas where existing AI procurement guidance needs to be clarified, completed, or updated. Prioritize gaps that can be addressed by clarifying existing rules or providing additional resources like best practices rather than creating new mandates to avoid unnecessary regulatory burdens. For example, guidance on handling personally identifiable information within commercially available information, guidance on data ownership between government and vendors, and the level of detail required for risk assessments. 

Categorize federal guidance into two main buckets: general federal procurement guidance (e.g., Federal Acquisition Regulation [FAR]) and agency-specific guidelines (e.g., individual AI policies from agencies such as DoD’s AI Memos or NASA’s Other Transaction Authorities [OTAs]). Ensure that agency-specific rules are clearly distinguished on the platform, allowing agencies to understand when general AI acquisition rules apply and when specialized guidance takes precedence. Since the FAR may take years to update to reflect agency best practices, this could help give visibility to potential gaps. 

Recommendation 2. The OMB OFPP-GSA-CAIO Council Working Group should convene a series of structured engagements with government and external stakeholders to co-create non-binding, practical guidance addressing gaps in AI procurement to be included in the platform.

These stakeholders should include government agency departments (e.g., project leads, procurement officers, IT departments) and external partners (vendors, academics, civil society organizations). The working group’s recommendations should focus on providing agencies with the tools, content, and resources they need to navigate AI procurement efficiently. Key focus areas would include risk management, ethical considerations, and compliance with cybersecurity policies throughout the procurement process. The guidance should also highlight areas where more frequent updates will be required, particularly in response to rapid developments in AI technologies and federal policies.

Topics that these stakeholder convenings could cover include: 

Procurement Process

Transparency 

Resources:

Recommendation 3. The OPPF, in collaboration with GSA and the United States Digital Service (USDS) should then develop an intuitive, easy-to-navigate digital platform that guides federal agencies through an “acquisition journey” for AI procurement.

While the focus of this memo is on the broader procurement of AI systems, this digital platform could also benefit from the incorporation of AI, for example, by using a chatbot that is able to refer government users to the specific regulations governing their use cases. At each process stage, the platform should clearly outline the necessary information collected during the previous phases of this project to help users understand exactly what is needed at any given point in the procurement lifecycle.

The platform should serve as a central repository that unites all relevant AI procurement requirements, guidance from federal regulations (e.g., FAR, OMB memos), and insights from stakeholder convenings (e.g., vendors, academics, civil society). Each procurement stage should feature the most up-to-date guidance, ensuring a comprehensive and organized resource for federal employees.

The system should be designed for ease of navigation, potentially modeled after Brazil’s CPIN, which is organized like a city subway map. Users can begin with a simple questionnaire recommending a specific “subway line” or procurement process. Each “stop” along the line would represent a key stage in the procurement journey, offering relevant guidance, requirements, and best practices for that phase. 

OPPF and GSA must regularly update the platform to reflect the latest federal AI and procurement policies and evolving best practices from government, civil society, and industry sources. Regular updates ensure that agencies use the most current information, especially as AI technologies and policies evolve rapidly.

The Federal Acquisition Institute within OFPP should create robust training programs to familiarize public servants with the new platform and how to use it effectively. These programs should explain how the platform supports AI acquisition and links to broader agency AI strategies.

  1. Roll out the platform gradually through agency-specific capacity-building sessions, demonstrating its utility for different departments. These sessions should show how the resource can help public servants meet their AI procurement needs and align with their agency’s strategic AI goals.
  2. Develop specialized training modules for different government stakeholders. For example, project teams might focus on aligning AI systems with mission objectives, procurement specialists on contract compliance, and IT departments on technical evaluations and cybersecurity.
  3. To ensure broad understanding and transparency, host public briefings for external stakeholders such as vendors, civil society organizations, and researchers. These sessions would clarify AI procurement requirements, fostering trust and collaboration between the public and private sectors.

Conclusion

The proposed centralized platform would represent a significant step forward in streamlining and standardizing the acquisition of AI technologies across federal agencies. By consolidating guidance, resources, and best practices into a user-friendly digital interface, this initiative would address gaps in the current AI acquisition landscape without increasing bureaucracy. This initiative supports individual agencies in their AI adoption efforts. It promotes a cohesive, government-wide approach to responsible AI implementation, ultimately benefiting both public servants and the citizens they serve.

This action-ready policy memo is part of Day One 2025 — our effort to bring forward bold policy ideas, grounded in science and evidence, that can tackle the country’s biggest challenges and bring us closer to the prosperous, equitable and safe future that we all hope for whoever takes office in 2025 and beyond.

PLEASE NOTE (February 2025): Since publication several government websites have been taken offline. We apologize for any broken links to once accessible public data.

Frequently Asked Questions
What do federal agencies have to consider when procuring AI technologies?

There are so many considerations based on a particular agency’s many needs. A non-exhaustive list of legislation, executive orders, standards and other guidance relating to innovation procurement and agency use of AI can be found here. One approach to top-level simplification and communication is to create something similar to Brazil’s city subway map, discussed above.

Could this be a platform be used for other types of procurement, especially innovation procurement in general?

The original Brazilian CPIN is designed for general innovation procurement and is agnostic to specific technologies or services. However, this memo focuses on artificial intelligence (AI) in light of recent guidance from the Office of Management and Budget (OMB) and the growing interest in AI from both the Biden Administration and the incoming Trump Administration. Establishing a platform specifically for AI system procurement could serve as a pilot for developing a broader innovation procurement platform.

How does this platform ensure compliance with safety, equity, and ethical standards in AI procurement?

The platform seeks to ensure responsible public sector AI by mitigating information asymmetries between government agencies and vendors, specifically by:



  • Incorporating the latest OMB guidelines on AI system usage, focusing on human rights, safety, and data transparency. These guidelines are seamlessly integrated into each step of the procurement process.

  • Throughout the “acquisition journey,” the platform should include clarifying checkpoints where agencies can demonstrate how their procurement plans align with established safety, equity, and ethical standards.

  • Prompting agencies to consider how procured AI systems will address context-specific risks by integrating agency-specific guidance (e.g., the Department of Labor’s AI Principles) into the existing AI procurement frameworks.

Enhancing Local Capacity for Disaster Resilience

Across the United States, thousands of communities, particularly rural ones, don’t have the capacity to identify, apply for, and manage federal grants. And more than half of Americans don’t feel that the federal government adequately takes their interests into account. These factors make it difficult to build climate resilience in our most vulnerable populations. AmeriCorps can tackle this challenge by providing the human power needed to help communities overcome significant structural obstacles in accessing federal resources. Specifically, federal agencies that are part of the Thriving Communities Network can partner with the philanthropic sector to place AmeriCorps members in Community Disaster Resilience Zones (CDRZs) as part of a new Resilient Communities Corps. Through this initiative, AmeriCorps would provide technical assistance to vulnerable communities in accessing deeply needed resources. 

There is precedent for this type of effort. AmeriCorps programming, like AmeriCorps VISTA, has a long history of aiding communities and organizations by directly helping secure grant monies and by empowering communities and organizations to self-support in the future. The AmeriCorps Energy Communities is a public-private partnership that targets service investment to support low-capacity and highly vulnerable communities in capitalizing on emerging energy opportunities. And the Environmental Justice Climate Corps, a partnership between the Environmental Protection Agency (EPA) and AmeriCorps, will place AmeriCorps VISTA members in historically marginalized communities to work on environmental justice projects. 

A new initiative targeting service investment to build resilience in low-capacity communities, particularly rural communities, would help build capacity at the local level, train a new generation of service-oriented individuals in grant writing and resilience work, and ensure that federal funding gets to the communities that need it most.

Challenge and Opportunity  

A significant barrier to getting federal funding to those who need it the most is the capacity of those communities to search and apply for grants. Many such communities lack both sufficient staff bandwidth to apply and search for grants and the internal expertise to put forward a successful application. Indeed, the Midwest and Interior West have seen under 20% of their communities receive competitive federal grants since the year 2000. Low-capacity rural communities account for only 3% of grants from the Federal Emergency Management Agency (FEMA)’s flagship program for building community resilience. Even communities that receive grants often lack the capacity for strong grant management, which can mean losing monies that go unspent within the grant period.

This is problematic because low-capacity communities are particularly vulnerable to natural disasters from flooding to wildfires. Out of the nearly 8,000 most at-risk communities with limited capacity to advocate for resources, 46% are at risk for flooding, 36% are at risk for wildfires, and 19% are at risk for both.

Ensuring communities can access federal grants to help them become more climate resilient is crucial to achieving an equitable and efficient distribution of federal monies, and to building a stronger nation from the ground up. These objectives are especially salient given that there is still a lot of federal money available through the Inflation Reduction Act (IRA) and the Infrastructure Investment and Jobs Act (IIJA) that low-capacity communities can tap into for climate resilience work. As of April 2024, only $60 billion out of the $145 billion in the IRA for energy and climate programs had been spent. For the IIJA, only half of the nearly $650 billion in direct formula funding had been spent. 

The Biden-Harris Administration has tried to address the mismatch between federal resilience funding and community capacity in a variety of ways. The Administration has deployed resources for low-capacity communities, agencies tasked with allocating funds from the IRA and IIJA have held information sessions, and the IRA and IIJA contain over a hundred technical assistance programs. Yet there still is not enough support in the form of human capacity at the local level to access grants and other resources and assistance provided by federal agencies. AmeriCorps members can support communities in making informed decisions, applying for federal support, and managing federal financial assistance. Indeed, state programs like the Maine Climate Corps, include aiding communities with both resilience planning and emergency management assistance as part of their focus. Evening the playing field by expanding deployment of human capital will yield a more equitable distribution of federal monies to the communities that need it the most. 

AmeriCorps’ Energy Communities initiative serves as a model for a public-private partnership to support low-capacity communities in meeting their climate resilience goals. Over a three-year period, the program will invest over $7.8 million from federal agencies and philanthropic dollars to help communities designated by the Interagency Working Group on Coal & Power Plant Communities & Economic Revitalization on issues revolving around energy opportunity, environmental cleanup, and economic development to help communities capitalize on emerging energy opportunities. 

There is an opportunity to replicate this model towards resilience. Specifically, the next Administration can leverage the Federal Emergency Management Agency (FEMA’s) Community Disaster Resilience Zone (CDRZ) designations to target AmeriCorps support to the communities that need it most. Doing so will not only build community resilience, but will help restore trust in the federal government and its programs (see FAQ).

Plan of Action

The next administration can support vulnerable communities in building climate resilience by launching a new Resilient Communities Corps through AmeriCorps. The initiative can be launched through a three-part Plan of Action: (1) find a philanthropic partner to fund AmeriCorps placements in CDRZs, (2) engage federal agencies that are part of the Thriving Communities Network to provide resilience training and support to Corps members, and (3) use the CDRZ designations to help guide where AmeriCorps members should be placed.  

Recommendation 1. Secure philanthropic funding 

American service programs have a history of utilizing philanthropic monies to fund programming. The AmeriCorps Energy Communities is funded with philanthropic monies from Bloomberg Philanthropies. California Volunteers Fund (CVF), the Waverly Street Foundation, and individual philanthropists helped fund the state Climate Corps. CVF has also provided assistance and insights for state Climate Corps officials as they develop their programs. 

A new Resilient Communities Corps under the AmeriCorps umbrella could be funded through one or several major philanthropic donors, and/or through grassroots donations. Widespread public support for AmeriCorps’ ACC that transcends generational and party lines presents the opportunity for new grassroots donations to supplement federal monies allocated to the program along with tapping the existing network of foundations, individuals, companies, and organizations that have provided past donations. The Partnership for the Civilian Climate Corps (PCCC), which has had a history of collaborating with the ACC’s federal partners, would be well suited to help spearhead this grassroots effort. 

America’s Service Commissions (ASC), which represents state service commissions, can also help coordinate with state service commissions to find local philanthropic monies to fund AmeriCorps work in CDRZs. There is precedent for this type of fundraising. Maine’s state service commission was able to secure private monies for one Maine Service Fellow. The fellow has since worked with low-capacity communities in Maine on climate resilience. ASC can also work with state service commissions to identify current state, private, and federally funded service programming that could be tapped to work in CDRZs or are currently working in CDRZs. This will help tie in existing local service infrastructure.  

Recommendation 2. Engage federal agencies participating in the Thriving Communities Network and the American Climate Corps (ACC) interagency working group. 

Philanthropic funding will be helpful but not sufficient in launching the Resilient Communities Corps. The next administration should also engage federal agencies to provide AmeriCorps members participating in the initiative with training on climate resilience, orientations and points of contact for major federal resilience programs, and, where available, additional financial support for the program. The ACC’s interagency working group has centered AmeriCorps as a multiagency initiative that has directed resources and provides collaboration in implementing AmeriCorps programming. The Resilient Communities Corps will be able to tap into this cross-agency collaboration in ways that align with the resilience work already being done by partnership members. 

There are currently four ACC programs that are funded through cooperation with other federal agencies. These are the Working Lands Climate Corps with the U.S. Department of Agriculture (USDA)’s Natural Resources and Conservation Service, AmeriCorps NCCC Forest Corps with the USDA Forest Service, Energy Communities AmeriCorps with the Department of Interior and the Department of Commerce, and the Environmental Justice Corps, which was announced in September 2024 and will launch in 2025, with the EPA. The Resilient Communities Corps could be established as a formal partnership with one or more federal agencies as funding partners.

In addition, the Resilient Communities Corps can and should leverage existing work that federal agencies are doing to build community capacity and enhance community climate resilience. For instance, USDA’s Rural Partners Network helps rural communities access federal funding while the EPA’s Environmental Justice Thriving Communities Technical Assistance Centers Program provides training and assistance for communities to build the capacity to navigate, develop proposals, and manage federal grants. The Thriving Communities Network provides a forum for federal agencies to provide technical assistance to communities trying to access federal monies. Corps members, through the network, can help federal agencies provide communities they are working with building capacity to access this technical assistance. 

Recommendation 3. Use CDRZ designations and engage with state service commissions to guide Resilient Communities Corps placements

FEMA, through its National Risk Index, has identified communities across the country that are most vulnerable to the climate crisis and need targeted federal support for climate resilience projects. CDRZs provide an opportunity for AmeriCorps to identify low-capacity communities that need their assistance in accessing this federal support. With assistance from partner agencies and philanthropic dollars, the AmeriCorps can fund Corps members to work in these designated zones to help drive resources into them. As part of this effort, the ACC interagency working group should be broadened to include the Department of Homeland Security (which already sponsors FEMA Corps).

In 2024, the Biden-Harris Administration announced Federal-State partnerships between state service commissions and the ACC. This partnership with state service commissions will help AmeriCorps and partner agencies identify what is currently being done in CDRZs, what is needed from communities, and any existing service programming that could be built up with federal and philanthropic monies. State service commissions understand the communities they work with and what existing programming is currently in place. This knowledge and coordination will prove invaluable for the Resilient Communities Corps and AmeriCorps more broadly as they determine where to allocate members and what existing service programming could receive Resilient Communities Corps designation. This will be helpful in deciding where to focus initial/pilot Resilient Communities Corps placements.

Conclusion

A Resilient Communities Corps presents an incredible opportunity for the next administration to support low-capacity communities in accessing competitive grants in CDRZ-designated areas. It will improve the federal government’s impact and efficiency of dispersing grant monies by making grants more accessible and ensure that our most vulnerable communities are better prepared and more resilient in the face of the climate crisis, introduce a new generation of young people to grant writing and public service, and help restore trust in federal government programs from communities that often feel overlooked.

This action-ready policy memo is part of Day One 2025 — our effort to bring forward bold policy ideas, grounded in science and evidence, that can tackle the country’s biggest challenges and bring us closer to the prosperous, equitable and safe future that we all hope for whoever takes office in 2025 and beyond.

PLEASE NOTE (February 2025): Since publication several government websites have been taken offline. We apologize for any broken links to once accessible public data.

Frequently Asked Questions
How much would a Resilient Communities initiative cost?

Funding for one AmeriCorps member in each of FEMA’s 483 designated Community Disaster Resilience Zones would cost around $14,500,000 per year. This is with an estimate of $30,000 per member. However, this figure will be subject to change due to overhead and living adjustment costs.

Why focus on CDRZ communities? Aren’t there lots of other communities that could also benefit from support?

There are many communities that could benefit from additional support when it comes to building resilience. Headwater Economics, a research institute in Montana, has flagged that the CDRZ does not account for all low-capacity communities hampered in their efforts to become more climate resilient. But the CDRZ designation does provide a federal framework that can serve as a jumping-off point for AmeriCorps to begin to fill capacity gaps. These designations, identified through the National Risk Index, provide a clear picture for where federal, public and private monies are needed the most. These communities are some of the most vulnerable to climate change, lack the resources for resilience work, and need the human capacity to access them. Because of these reasons, the CDRZ communities provide the ideal and most appropriate area for the Resilient Communities Corps to first serve in.

How would the Resilient Communities initiative help restore local trust in federal government?

Funding for national service programming, particularly for the ACC, has bipartisan support. 53% of likely voters say that national service programming can help communities face climate-related issues.


On the other hand, 53% of Americans also feel that the federal government doesn’t take into account “the interests of people like them.” ACC programming, like what Maine’s Climate Corps is doing in rural areas, can help reach communities and build support among Americans for government programs that can be at times met with hostility.


For example, in Maine, the small and politically conservative town of Dover-Foxcroft applied for and was approved to host a Maine Service Fellow (part of the Maine Climate Corps network) to help the local climate action committee to obtain funding for and implement energy efficiency programs. The fellow, a recent graduate from a local college, helped Dover-Foxcroft’s new warming/cooling emergency shelter create policies, organized events on conversations about climate change, wrote a report about how the county will be affected by climate change, and recruited locals at the Black Fly Festival to participate in energy efficiency programs.


Like the Maine Service Fellows, Resilient Communities Corps members will be integral members of the communities in which they serve. They will gather essential information about their communities and provide feedback from the ground on what is working and what areas need improvement or are not being adequately addressed. This information can be passed up to the interagency working groups that can then be relayed to colleagues administering the grants, improving information flow, and creating feedback channels to better craft and implement policy. It also presents the opportunity for representatives of those agencies to directly reach out to those communities to let them know they have been heard and proactively alert residents to any changes they plan on making.

Ensuring the Next Generation of STEM Talent through K–12 Research Programming

Labor shortages persist in the United States in a variety of STEM (science, technology, engineering, and mathematics) fields. To address these shortages, the next administration should establish a national, federally funded initiative involving the public and private sectors to develop a more robust and diverse pipeline of STEM talent. The Next Generation of STEM Talent Through K–12 Research Programming Initiative will remove significant barriers to participation in STEM careers through enhanced K–12 STEM programs such as science fairs and robotics competitions, as well as through strengthened federal support for teacher training to actively engage K–12 students in STEM research.

Challenge and Opportunity

Need for a Stronger STEM Pipeline in the United States

The 2024 Federal Strategic Plan for Advancing STEM Education and Cultivating STEM Talent from the National Science and Technology Council (NSTC) notes that “The United States must “inspire, educate, train, and innovate in STEM fields and STEM careers, so that through unparalleled access and opportunity, the nation can leverage the full potential of its STEM talent and ensure the country’s national security, economic prosperity, and global competitiveness.” Indeed, a vigorous domestic STEM workforce that innovates quickly to confront national challenges is a central driver for economic growth. Yet while the number of degrees awarded in STEM fields has increased since 2000 in the United States, labor shortages persist in certain fields requiring STEM degrees. These fields include computer science, data science, electrical engineering, and software development.

Fostering STEM talent across the country “is critical both to enable all individuals to achieve their own aspirations in STEM fields and careers and to ready the nation to pursue new opportunities.” Yet, the rest of the world is outpacing the United States when it comes to upper-level STEM education. The United States awarded nearly 800,000 first university degrees (i.e., associate’s and bachelor’s degrees) in science and engineering (S&E) in 2016. However, the European Union (EU) top six  countries (France, Germany, Italy, Poland, Spain, and the U.K., then part of the EU) produced more than 700,000 equivalent degrees—and China 1.7 million (in 2015)—around the same timeframe. In 2020, the United States came in third in terms of the most first university degrees in science and engineering (900,000), lagging behind nations such as India (2.5 million) and China (2 million).

The data are more complex but equally worrisome at the doctorate level. As of 2019, the United States no longer awards the largest number of science and engineering (S&E) doctoral degrees of any country. It was surpassed by China, with the United States awarding 42,000 and China awarding 43,000 that year. Comparisons of doctoral-degree production in the United States with doctoral-degree production in other nations need to account for the fact that a substantial number of U.S. S&E doctorate recipients are students on temporary visas. However, many of these doctorate recipients stay in the United States for jobs after obtaining their degrees. Moreover, the United States also lags peer nations when it comes to the percentage of S&E doctorates awarded out of all doctorates awarded. This figure is 44% for the United States, behind China (nearly 60%), Sweden (55%), Taiwan (53%), India (50%), and the U.K. (48%).

We as a nation must prepare by strengthening the STEM pipeline and closing the gap between demand for and supply of STEM talent. This effort must also focus on creating a diverse and inclusive STEM talent pool. Only by drawing on the talents of all its citizens can the United States effectively maintain and grow the national innovation base that supports key economic sectors. This broader participation in STEM “fosters closer alignment between societal needs and research, enhances public understanding and trust in science, facilitates uptake of research results throughout society, and supports evidence-based policymaking.”

If the United States is to keep pace and ensure continued innovation and prosperity, it must up its game on STEM education and training. Because of the time and training required to become a scientist or engineer, this effort must begin without delay. The COVID-19 pandemic emphasized the need for a robust STEM workforce. Scientists raced to discover more about the virus itself and its impact, as well as to develop vaccines and treatments safely and in record time. Engineers designed new equipment and ways to manufacture needed personal protective equipment (PPE) and ventilators. Computer scientists, statisticians, epidemiologists, and big-data scientists collaborated to make sense of pandemic data and model outcomes to inform public-health policies. Similar crises will inevitably arise in the future. 

Engaging Learning Experiences with Well-Trained Educators are Even More Important Because of Pandemic Learning Losses 

The coronavirus pandemic led to a significant disruption in K-12 education. Even with students back in classrooms, the negative impact of this disruption is clear and will have myriad effects on the STEM talent pipeline into the future.

Chronic absenteeism nationwide (based on students missing at least 10% of a school year) surged from 15% in 2018 to 28% in 2022, showing that post-pandemic school attendance has reduced test scores. Student attendance is instrumental to their success. As absenteeism increased, test scores declined. 

This standardized test score decline is seen across the globe where middle and high school students are still struggling academically in the years since the start of the pandemic. The Program for International Student Assessment, taken by 15-year-olds, found record decreases in scores between 2018 and 2022, where math scores decreased by 15 points and reading scores by 10 points. When students have fewer math skills, it reduces the number of students likely to become STEM experts, which narrows the pool of future scientists and engineers.

Students experienced years of learning loss, along with disruption to their social and emotional development. When compared to peer nations, U.S. children are not equipped with the high-level reading, math and digital problem-solving skills needed for the fastest-growing jobs especially in a global economy that is highly competitive. The most vulnerable students are also the most negatively impacted. Gaps already present in 2019 between high-poverty and higher-income school districts increased during the pandemic and have not closed.

Launching the Next Generation of STEM Talent Through K–12 Research Programming Initiative

The next administration should launch the Next Generation of STEM Talent Through K–12 Research Programming Initiative, coordinated by the White House Office of Science and Technology Policy (OSTP) through a working group of federal agency representatives, to strengthen the STEM pipeline in the United States. The initiative would provide an additional $25 million per year for 10 years to select agencies to support K–12 research programs (such as science fairs and robotics competitions) that inspire critical thinking and encourage young people to pursue STEM careers. The new funds would also be used to train educators and community- based scientists to become K–12 research mentors, expand research programs at the local and national levels, and build an interagency tracking mechanism to coordinate and evaluate the success of these programs. These activities directly support the five interdependent pillars outlined in the Committee on STEM Education (CoSTEM) 2024 Federal Strategic Plan for Advancing STEM Education and Cultivating STEM Talent:

Since almost 16% of the 2.1 million federal employees in the United States occupy a STEM position, this initiative would directly benefit the Federal Government—and, by extension, U.S. civil society. Students and educators involved with this initiative would increase their awareness of Federal Government STEM occupations and develop a mental contract with participating U.S. agencies that will impact future career choices. This initiative should also involve the private sector, as many companies and their trade associations are also in need of STEM talent and lead programs that the initiative could leverage. In 2021, out of 146.4 million people ages 18 to 74 working in the United States, 34.9 million (24%) were in STEM occupations. Only the federal government has the resources and infrastructure to undertake and coordinate this public-private partnership.

Inclusivity is an indispensable aspect and opportunity of this new initiative. To foster development of STEM skills, the 2023 Progress Report on the Implementation of the Federal Science, Technology, Engineering, and Mathematics (STEM) Education Strategic Plan emphasized that “the nation must engage in a collaborative effort to ensure that everyone has access to high quality STEM education throughout their lifetimes.” Access to STEM education and representation in STEM fields is unequally distributed in the United States. Women, differently abled persons, and three ethnic or racial groups—Blacks or African Americans, Hispanics or Latinos, and American Indians or Alaska Natives—are significantly underrepresented in science and engineering education and employment. In 2021, a greater share of men (29%) than women (18%) worked in STEM occupations, even though men and women represented similar proportions of the total workforce (52% men and 48% women). Similarly, Blacks/African Americans and Hispanics/Latinos make up about 28% of the overall population but only 13% of the STEM workforce. Research suggests there are many individuals—especially women, minorities, and children from low-income families—who would have developed highly impactful inventions had they been exposed to innovation in childhood. The Next Generation of STEM Talent Through K–12 Research Programming Initiative is designed to help find those “lost Einsteins”. 

There also will be an emphasis placed on rural students who do not have adequate mentors and educational systems currently in place. Studies have shown that underserved minority and rural communities often do not have access to the same educational opportunities as more affluent white communities, and this impacts the careers they will pursue. The pandemic exposed the enormous gaps between the country’s poorest and wealthiest schools around access to basic technology and live remote instruction, as well as the percentages of students who teachers report were not logging in or making contact.

The Federal Strategic Plan for Advancing STEM Education and Cultivating STEM Talent cites one of its pillars as STEM Research and Innovation Capacity. Informal learning, especially participation in research programs such as science fairs or robotic competitions, is one way to inspire critical thinking in young people and foster long-term interest in STEM. Research funded by the National Science Foundation shows that participating in a science research project increases student interest in STEM careers. These competitions provide students with opportunities to create solutions to real-life problems, encouraging innovation, which is a critical component of economic growth and entrepreneurial talent development.

There is flexibility in how opportunities are delivered to students. When schools were shut down in 2020-2021, the Society for Science converted its STEM Research Grants program, an opportunity for teachers to receive up to $5,000 for classroom resources and/or transportation to research sites, to include STEM Research kits full of resources that students were able to bring home to complete STEM research outside of school. The Society for Science has continued to provide home and school options for the resources teachers receive from this program. Relatedly, the Society for Science launched a new Research at Home website to support this work.

No matter if the vehicle for delivery is from educators providing materials to be used at home or at school, success in this area requires training teachers to be effective research mentors.  In line with CoSTEM’s Federal Objective for Training STEM Educators, an excellent prototype for such training is the Research Teachers Conferences run by the Society for Science. The Research Teachers Conferences convene high-school and middle school STEM research teachers annually to share best practices, troubleshoot challenges, and establish a network of support for each other. Nearly 2,000 teachers each year request the opportunity to attend these conferences, but funding for 2024 was only available for 275—highlighting the pent-up demand for STEM research training. More training is also needed to help professional scientists become more effective research mentors for K–12 students, and they, too, need training to ensure optimal effectiveness. The Next Generation of STEM Talent Through K–12 Research Programming Initiative is designed to train a collaborative community of K–12 research mentors working throughout the United States.

There are already many hands-on programs designed to increase the STEM talent pool by providing research-based and problem-solving learning opportunities to K–12 students: especially underrepresented minorities, girls, or students from rural communities. These programs range in size from small to large and in scope from local to federal. Programs are run by institutions such as nonprofit organizations, colleges and universities, scientific societies, and even industry trade associations. For example, the American Chemical Society has provided economically disadvantaged high-school students with paid summer-research internships for more than 50 years. Students participating in the internship program work under the guidance of professional scientists who have been trained to be research mentors. The Society for Science’s Advocate Program provides mentors to support underserved students in submitting research projects to science competitions. Funding for these types of K–12 STEM programs comes from a myriad of sources, including philanthropic foundations and individuals, companies, and local, state, and federal governments. But there is currently no widespread coordination among these programs or sharing of best practices. There is also little rigorous evaluation to determine program success. The Next Generation of STEM Talent Through K–12 Research Programming Initiative will provide leadership to align complementary efforts and additional funding to support assessment and scale-up of practices proven effective.

Only the federal government has the ability to accomplish the three objectives outlined above. But as the 2024 STEM Plan states, “the federal government alone cannot produce the STEM talent needed for the entire country. Multi-agency and multi-sector partnerships and ecosystem development, including with international counterparts, are necessary to achieve a vision for STEM in America.”  

Plan of Action

The Next Generation of STEM Talent Through K–12 Research Programming Initiative should have four major components:

Component 1. White House leadership, coordination, tracking, and evaluation

The next president should sign an Executive Order (EO) launching a national Next Generation of STEM Talent Through K–12 Research Programming Initiative led by the White House Office of Science and Technology Policy (OSTP). The initiative would oversee and strengthen federal support for teacher training and program development designed to actively engage students in STEM research and problem-solving. 

The EO should also establish an OSTP-led working group like the Committee on STEM Education (CoSTEM), the NSTC group that wrote Charting a Course for Success: America’s Strategy for STEM Education. CoSTEM – with its mandate to review STEM education programs, investments, and activities, and the respective assessments of each, in federal agencies to ensure that they are effective – serves as a model for this initiative. While CoSTEM coordinates the interagency working groups focused on different aspects of STEM, particularly the Interagency Working Group to Engage Students where Disciplines Converge (IWGC) and the Interagency Working Group to Develop and Enrich Strategic Partnerships (SP-IWG), this new working group would coordinate relevant activities across federal agencies and their subunits, with the goal to gather the leading scientists, administrators and educators doing this work outside of federal agencies, leveraging the organizational power of the federal government to provide the resources and infrastructure to coordinate this public-private partnership.

While some federal agencies already have directly relevant programs in place, other agencies could help identify offices and programs essential to the initiative’s success. The working group should issue an open call for nonprofit organizations with expertise in research-based STEM learning and teacher/mentor training to participate as advisors to the working group. The working group could also include representatives from existing programs that help expand research-based and problem-solving STEM experiences at the K–12 level. The EO should task the working group with developing a strategic national action plan that includes metrics to monitor the initiative’s success, as well as with creating a centralized database that can track, monitor, and evaluate programs funded by the initiative. The working group should periodically report to the Executive Office of the President on the initiative’s progress.

Overall goals of the initiative would be to:

  1. Ensure an abundance of qualified applicants from a variety of backgrounds—including variety in gender, race, or socioeconomic status—for all STEM jobs in the United States.
  2. Train teachers to provide students with research-based STEM education opportunities throughout their K–12 education experiences
  3. Create a comprehensive database to track programs (and their participants) aligned with and/or funded by the initiative.
  4. Rigorously and fairly evaluate programs aligned with and/or funded by the initiative, quickly communicating evaluation findings in ways that help programs adjust to best serve students and educators.

Quantitative targets to assess progress towards these goals include:

  1. Improving the extent to which demographics of applicants to STEM jobs in the United States reflect demographics of the United States as a whole.
  2. Availability of project-based STEM learning at publicly funded K–12 schools, as well as student access to opportunities (e.g., science fairs) where they can share the results of their work.
  3. Grow the pool of qualified STEM research educators to 100,000 in the next 10 years, so that all schools have access to the needed number of trained educators.

Component 2. Federal budget commitments

A few agencies—such as the National Aeronautics and Space Administration (NASA), the National Security Agency (NSA), and the Department of Defense (DoD)—currently directly support aspects of this initiative. Yet at least 20 federal agencies (full list found in FAQ) and their subunits have a clear stake in developing the STEM workforce and hiring STEM graduates. 

Each of these federal units will need dedicated funding to support the initiative, including by:

We estimate that an average allocation of $25 million per year for 10 years per relevant federal unit would be sufficient to get the initiative off the ground. These funds alone are not enough to develop the STEM workforce to the level needed in the United States. However, consistent federal funding for K–12 research programming (and associated teacher training) would provide a solid foundation for addressing the shortfalls outlined at the beginning of this memo. To maximize the initiative’s impact, additional funding should be allocated specifically for coordination and evaluation. Evaluations should be carried out every few years, and findings used to inform funding priorities and program structure as needed. Emphasis should be placed on allocating funds to expand access to high-quality STEM experiences for underserved and underrepresented students.

Component 3. Meaningful agency participation

The working group will identify existing federal programs that could be expanded to achieve the initiative’s goal. The working group will also identify agencies that have relevant missions but currently lack relevant programs.

Component 4. Partnership with non-federal organizations to provide programmatic content and complementary actions

The working group should partner with third-party organizations that already offer programs and resources (financial and in-kind) relevant to the initiative. These include but are not limited to:

The working group itself should aim to have representatives from underrepresented groups in STEM to ensure a wide variety of voices are represented as part of the leadership of this initiative. 

The next administration can use the power of the federal government to help such third-party organizations scale up and strengthen programs that have already proven effective, resulting in more teachers and scientists trained and more K–12 students able to participate in science and engineering research projects.

Precedents

The initiative outlined in this memo can and should build on multiple outstanding federal precedents. One example is the DoD’s Defense STEM Education Consortium (DSEC). The DSEC is a collaborative partnership among academia, industry, nonprofit organizations, and government that aims to broaden STEM literacy and develop a diverse and agile workforce with the technical excellence to defend our nation. Many smaller federal programs provide teacher training in various STEM fields. The next administration should leverage and potentially refocus such existing programs to emphasize critical-thinking skills and research-based programs at the K–12 level.

Conclusion

The next administration should seize the opportunity to reinvigorate the STEM talent pool in the United States by creating the Next Generation of STEM Talent Through K–12 Research Programming Initiative. The initiative will motivate participation in STEM careers by making participation in hands-on STEM research and problem-solving opportunities a standard component of K–12 education. Failure to replenish and grow our domestic STEM talent pool will lead to a decline in national innovation and economic progress and an inability to meet the moment in future times of challenge, such as a next pandemic. Only the federal government can address this need at the scale and pace needed.

Inclusivity is an indispensable aspect of this initiative. Building a robust STEM workforce in the United States requires us as a nation to draw on the talent of all Americans. The Next Generation of STEM Talent Through K–12 Research Programming Initiative will rediscover our country’s ”lost Einsteins”: the underrepresented minorities, women and underserved students from rural communities who have the capacity to deliver transformative contributions to STEM if only they were provided opportunities to do so.

This action-ready policy memo is part of Day One 2025 — our effort to bring forward bold policy ideas, grounded in science and evidence, that can tackle the country’s biggest challenges and bring us closer to the prosperous, equitable and safe future that we all hope for whoever takes office in 2025 and beyond.

PLEASE NOTE (February 2025): Since publication several government websites have been taken offline. We apologize for any broken links to once accessible public data.

Frequently Asked Questions
How does this idea complement existing actions already undertaken by the federal government?

The current and previous administrations have taken multiple actions that serve as a foundation for achieving the goals of Ensuring the Next Generation of STEM Talent Through K-12 Research Programming. “Agencies across the federal government are united in their commitment to developing STEM talent so that all individuals and communities can grow, aspire, and thrive, allowing the United States to reach its full potential.” While there are groups within the Federal government that are doing similar work, like the Committee on STEM Education (CoSTEM) with its mandate to review science, technology, engineering, and mathematics (STEM) education programs, investments, and activities, and the respective assessments of each, in federal agencies to ensure that they are effective, this initiative is unique in that the main leadership would be a coalition of leaders and organizations outside of the government, with a government agency like OSTP coordinating, rather than CoSTEM’s focus on interagency working groups (IWGs). CoSTEM serves as a model, with the goal to gather the best of the best who are doing this work outside of federal agencies and leverage the organizational power of the government to provide the resources and infrastructure to coordinate this public-private partnership.

What other federal initiatives can we use as a model of effective collaboration?

An outstanding example of a federal initiative that works is DoD STEM’s Defense STEM Education Consortium (DSEC). Aligned to the Federal STEM Education Strategic plan, the Defense Science, Technology, Engineering, and Mathematics Education Consortium (DSEC) is a collaborative partnership among academia, industry, non-for-profit organizations, and government that aims to broaden STEM literacy and develop a diverse and agile workforce with the technical excellence to defend our Nation. By addressing and prioritizing critical STEM challenges, DoD is investing in evidence-based approaches to inspire and develop the Nation’s science and technology workforce.


This multi-year effort includes elements focused on STEM enrichment programs for students and educators, STEM workforce engagement, program evaluation, and public outreach. These efforts will allow DoD to improve access for students to pursue STEM careers and consider Defense laboratories as a place of employment. Through strategic investment in STEM education and outreach activities, the effort will provide students with more exposure to educational and career opportunities, as well as DoD research. The program includes scholarships, internships/apprenticeships, teacher training, and conferences.

Are there existing programs under federal agencies that would benefit from the coordination proposed by the Next Generation of STEM Talent Through K–12 Research Programming Initiative?

Yes. One example that could be adapted for K-12 are the existing programs funded by the National Science Foundation, which provides summer research experiences nationally and internationally to college students. Those programs involve partnerships with universities and non-for-profit scientific societies.


In 2022, the Department of Education launched its YOU Belong in STEM initiative to strengthen STEM education nationwide by implementing and scaling high-quality PreK through university STEM education for all, which is well aligned with the goals of this proposal. Similar programs exist in many other federal agencies, but they are not coordinated nor specifically directed to building the STEM pipeline.

Is there another agency in the Federal Government besides the White House Office of Science and Technology Policy that makes sense to coordinate this effort?

Although OSTP is the obvious coordinating group, in the event it could not undertake a project of this size, the National Science Foundation in concert with the Department of Education, given its YOU Belong in STEM initiative, would be an appropriate coordinator.


A few agencies—such as the National Aeronautics and Space Administration (NASA), the National Security Agency (NSA), and the Department of Defense (DoD)—currently directly support aspects of this initiative. Yet at least 20 federal agencies and their subunits have a clear stake in developing the STEM workforce and hire STEM graduates.

Since there are so many organizations in the private sector noted in the proposal doing similar types of programs, why should the federal government step in?

While there are dozens if not hundreds of organizations doing similar types of programs, they are underfunded, uncoordinated, and under-evaluated. They are not uniformly distributed throughout the U.S., and their goals are also diffuse. Only the federal government is positioned to create the umbrella to coordinate such programs, track them, and evaluate them.

If $250 million of funding is not available, what would a less ambitious version of this proposal look like?

The OSTP working group would need to prioritize the number of federal agencies and choose those that have the most at stake from this proposal—agencies that specifically need STEM workforce to carry out their mission. Narrowed in this way, the number of units might drop by 50%, from 20 to 10. The working group could also focus on those agencies that currently have robust programs in the same space and build on those. For example, more money might be given to the DSEC program.


In addition, funding should be continued and increased for the National Science Foundation to evaluate these programs from a rigorous viewpoint to determine whether they are succeeding. Continued funding would be dependent on the results of these evaluations.

An Innovation Agenda for Addiction: Breakthrough Medicines That Scale

The federal government should expand the FDA’s priority review voucher program (PRV) and provide market exclusivity advantages to encourage the development of medications for addiction. 

Taken together, substance use disorders (alcohol, cigarettes, and other drugs) cause more deaths in the U.S. every year than cancer or heart disease and cause devastating downstream social harms. Despite this, only 3% of eligible patients received substance use disorder (SUD) medication, a result of low uptake and efficacy of existing medications and a lack of options for patients addicted to stimulants. This is due to a near-total absence of pharmaceutical research and development activity. To make population level impact to reduce harms from opioids, methamphetamine, cocaine, alcohol, and cigarettes, we must address the broken market dynamics in addiction medicine. 

The PRV program should be expanded to cover opioid use disorder, alcohol use disorder, stimulant use disorder, and smoking. In addition, drugs that are approved for these SUD indications should have extended exclusivity and sponsors that develop these medications should receive vouchers to extend exclusivity for other medications.

Challenge and Opportunity 

Addiction policy efforts on both the left and the right have struggled. Despite substantial progress reducing smoking, 29 million Americans still smoke cigarettes and feel unable to quit and 480,000 Americans die each year from smoking. While overdose deaths from opioids, cocaine, and methamphetamine have fallen slightly from their peak in 2022, they are still near record highs, three  times higher than 20 years ago. Alcohol deaths per capita have doubled since 1999. 

Roughly 60% of all crimes and 65% of violent crimes are related to drugs or alcohol; and the opioid crisis alone costs the United States $1.5 trillion a year. Progress in reducing addiction is held back because people with a substance use disorder take medication. This low uptake has multiple causes: in opiate use disorder, uptake is persistently low despite recent relaxations of prescription rules, with patients reporting a variety of reasons for refusal; treatments for alcohol use disorder have modest effects; and there are no approved treatments for stimulant use disorder. Only three percent take SUD medications, as shown in figure 1 below [link to image]. In brief, only 2% of those suffering alcohol use disorder, 13% of those with opiate use disorder, 2% of smokers, and approximately 0% of illicit stimulant users are receiving medication, giving a weighted average of about 3%.

There has been rapid innovation in the illicit market as synthetic opioids and expanded meth production have lowered price and increased strength and availability. Meanwhile, there has been virtually no innovation in medicines to prevent and treat addiction. The last significant FDA approval for opioid use disorder was buprenorphine in 2002; progress since then has been minimal, with new formulations or dosing of old medications. For alcohol use disorder, the most recent was acamprosate in 2004 (and it is rarely prescribed due to limited efficacy and three times a day dosing).

None of the ten largest pharmaceutical companies have active addiction medicine programs or drug candidates, and the pharmaceutical industry as a whole has only pursued minimal drug development. According to the trade association BIO, “Venture investment into companies with novel addiction drug programs over the last 10 years is estimated at $130M, 270 times less than oncology.”

There are promising addiction drug candidates being studied by academics but without industry support they will never become medicines. If pharmaceutical companies spent just 10% of what they spend on obesity therapies, we might quickly make progress.

For example, GLP-1 medicines like Ozempic and Mounjaro have strong anti-addictive effects across substances. Randomized trials and real-world patient health record studies show dramatic drops in consumption of drugs and alcohol for patients taking a GLP-1. Many addiction scientists now consider these compounds to be the biggest breakthrough in decades. However, Novo Nordisk and Eli Lilly, who own the drugs currently in the market, do not plan to run phase 3 addiction trials on them, due to fear of adverse events in substance use disorder populations. The result is that a huge medical opportunity is stuck in limbo indefinitely. Fortunately, Lilly has recently signaled that they will run trials on related compounds, but remain years from approval.

Conversations with industry leaders make clear that large pharmas avoid SUD indications for several reasons. First, their upside appears limited, since current SUD medications have modest sales. Second, like other psychiatric disorders, the problem is challenging given the range and complexity of neurological targets and the logistical challenges of recruiting people with substance use disorder as participants. Finally, companies face downside reputational and regulatory risk if participants, who face high baseline rates of death from overdose regardless, were to die in trials. In the case of Ozempic and Mounjaro, sponsors face an obstacle some have termed the “problem of new uses” – clinical trials of an already lucrative drug for a new indication carry downside risk if new side effects or adverse events are reported. 

Image from Charting the fourth wave, based on CDC data

Plan of Action

Market Shaping Interventions

Recommendation 1. Expand the FDA priority review voucher (PRV) program to include addiction medicine indications.

The FDA priority review voucher (PRV) program incentivizes development of drugs for rare pediatric and infectious diseases by rewarding companies who get drugs approved with a transferable voucher that accelerates FDA approval. These vouchers are currently selling for an average of $100M. The PRV program doesn’t cost the government any money but it makes drug development in the designated categories much more lucrative. The PRV program has proven very successful, leading to a surge in approvals of medications.

As a neglected market with urgent unmet medical and public health needs, and which also promises to benefit the broader public by reducing the negative externalities of addiction, addiction medicine is a perfect fit for the PRV program. Doing so could transform the broken market dynamics of the field. The advantage of the PRV program is that it does not require substantial new congressional appropriations, though it will require Congress giving the FDA authority to expand the PRV program, as it has done previously to add other disease areas.

Recommendation 2. Extend exclusivity for addiction medicines and incentivize pursuit of new indications

Market exclusivity is a primary driver of pharmaceutical industry revenue. Extending exclusivities would have a very large effect on industry behavior and is needed to create sufficient incentives. The duration of exclusivity for alcohol use disorder, opioid use disorder, stimulant use disorder, and smoking cessation indications should be extended along with other incentives.  

For precedent, there are already a number of FDA programs that extend medication exclusivity, including ‘orphan drug exclusivity’ and the qualified infectious disease product (QIDP) program. Like rare diseases and antibiotics, addiction is a market that requires incentives to function effectively. In addition, successful treatments, given the negative externalities of addiction, have public benefit beyond the direct medical impact, and deserve additional public incentives.

Recommendation 3. Modernize FDA Standards of Efficacy for Substance Use Disorder Trials

A significant barrier to pharmaceutical innovation in SUDs is outdated or unpredictable efficacy standards sometimes set by the FDA for clinical trials. Efficacy expectations for substance use disorder indications are often rooted in abstinence-only and other binary measure orientations that the scientific and medical community has moved past when evaluating substance use disorder harms.

This article in the American Journal of Drug and Alcohol Abuse demonstrates that binary outcome measures like ‘number of heavy drinking days’ (NHDD) can underestimate the efficacy of treatments. This recent report from NIAAA on alcohol trial endpoints recommends a shift away from abstinence-based endpoints and towards more meaningful consumption-based endpoints. This approach should be adopted by the FDA for all SUD treatments, not just alcohol.

There are some indications that the FDA has begun modernizing their approach. This recent paper from NIH and FDA on smoking cessation therapies provides updated guidance that moves in the right direction.

More broadly, the FDA should work to adopt endpoints and standards of efficacy that mirror standards in other disease areas. This shift is best achieved through new guidance or statements issued by the FDA, which would offer positive assurance to pharmaceutical companies that they have achievable paths to approval. Predictability throughout the medication development life cycle is absolutely essential for companies considering investment.

Congress should include statements in upcoming appropriations and authorizations that state:

  1. The FDA should adopt non-binary standards of efficacy for addiction treatments that are aligned with standards for other common disorders and the FDA shall, within 12 months, report on the standards employed for substance use disorder relative to other prevalent chronic conditions and report steps to eliminate disparities in evidentiary standards and issue new guidance on the subject.
  2. The FDA should publish clear guidance on endpoints across SUDs to support planning among pharmaceutical companies considering work in this field.

Conclusion

Sustained focus and investment in diabetes and heart disease treatments has enabled medical breakthroughs. Addiction medicine, by contrast, has been largely stagnant for decades. Stimulating private-sector interest in addiction medicine through regulatory and exclusivity incentives, as well as modernized efficacy standards, is essential for disrupting the status quo. Breakthroughs in addiction medicine could save hundreds of thousands of lives in the US and provide long-term relief for one of our most intractable social problems. Given the negative externalities of addiction, this would also have enormous benefits for society at large, reducing crime and intergenerational trauma and saving money on social services and law enforcement.

This action-ready policy memo is part of Day One 2025 — our effort to bring forward bold policy ideas, grounded in science and evidence, that can tackle the country’s biggest challenges and bring us closer to the prosperous, equitable and safe future that we all hope for whoever takes office in 2025 and beyond.

PLEASE NOTE (February 2025): Since publication several government websites have been taken offline. We apologize for any broken links to once accessible public data.

Frequently Asked Questions
Why doesn’t the private sector target SUD? Why is government incentive necessary?

Per author conversations with industry leaders, private sector interest in SUD medication development is limited for the following reasons:



  • The upside of pursuing SUD indications appears limited, since current SUD medications, which are generally targeted for specific substances, have modest sales.

  • Even with preliminary evidence that GLP-1 drugs may be efficacious for some SUD indications (e.g, alcohol, opiates, and tobacco), companies are reluctant to pursue label expansion for SUD. As described previously, with already lucrative drugs, companies face a downside risk (termed the “problem of new uses”) from running large clinical trials, and possibly uncovering new side effects or incurring random adverse events which could harm reputation and existing markets.

  • In the specific case of SUD, this downside risk might be especially large, since people with substance use disorder have high baseline rates of overdose and death.


Moreover, there is an argument that a treatment for SUD is a public good, to the degree that it ameliorates the negative externalities of addiction – increasing the case for more public-sector incentives for SUD treatment. The end result is that medical treatments for SUD are stuck in an indefinite limbo, with private-sector interest in SUD, as documented previously, being very low.

Why are we optimistic about SUD medications?

The current lack of effective and widely used SUD medications is disheartening, but this is in the context of private sector disinterest and scant funding. Even modest successes in SUD treatment have the potential to kickstart an innovation loop, akin to the rush of biotech companies hastening to enter the obesity treatment field. Prior to the success of the GLP-1 drugs, obesity treatment had been moribund, and viewed pessimistically in light of drugs that had limited efficacy or had been withdrawn for side effects like suicidality or cardiovascular issues.


An SUD success like GLP-1 for obesity has the potential to kindle a similar rush of interest; the challenge is the initiation of that cascade. Given the very low levels of investment in SUD treatments, there is potential low-hanging fruit that, given sufficient funding, could be trialed and deployed.

What are the innovations in the illicit drug market?

There has been rapid innovation in the field of addiction, but it’s been happening on the wrong side: addiction-inducing technologies are becoming more powerful, while SUD treatments have largely stagnated. This innovation is most evident in synthetic opioids and methamphetamine.


Compared to heroin, fentanyl is about 25x stronger (on a per-weight basis), and hence, much easier to smuggle. As the Commission on Combating Synthetic Opioid Trafficking put it:


Single-digit metric tonnage of pure fentanyl is not a large amount and could easily fit into a shipping container or a truck trailer, which seriously challenges interdiction…Perhaps as much as 5 MT [metric tons] of pure fentanyl would be needed to satisfy the entire annual U.S. consumption for illegally supplied opioids.


Moreover, as a recent Scientific American article documented, innovations in fentanyl production, including the use of safer precursors and methods that don’t require sophisticated equipment, mean that fentanyl production is now decentralized, and resistant to attempts by law enforcement to shut it down.


As fentanyl has come to dominate the opioid supply over the past 10 years, overdose deaths have risen dramatically. New synthetic opioids and non-opioids like xylazine are also becoming common.


At the same time, due to advances in production techniques in Mexico, methamphetamine production has skyrocketed in recent decades while purity has improved. Worst of all, unlike heroin, fentanyl is easily combined with meth and cocaine in pills and powder.


The DEA has highlighted the presence of “super labs” in Mexico capable of producing hundreds of pounds of meth per batch.


Together, these three innovations (fentanyl, cheap meth, and new combinations) have led to a 400% increase in overdose deaths in the past 20 years. Without equally powerful innovations to reduce addiction rates, we will never make long-term and sustainable progress.

Restoring U.S. Leadership in Manufacturing

Manufacturing is a critical sector for American economic well-being. The value chains in the American economy that rely on manufactured goods account for 25% of employment, over 40% of gross domestic product (GDP), and almost 80% of research and development (R&D) spending in the United States. Yet U.S. leadership in manufacturing is eroding. U.S. manufacturing employment plummeted by one-third—and 60,000 U.S. factories were closed—between 2000 and 2010 (Bonvillian and Singer, Advanced Manufacturing: The New American Innovation Policies, MIT Press, 2018, 52, 265.) Only some 18% of the production jobs lost in the United States during the Great Recession were recovered in the following decade, and production output only returned to its pre–Great Recession levels in 2018. This hollowing out of U.S. manufacturing has been largely driven by international competition, particularly from China. China passed the United States in 2011 as the largest global manufacturing power in both output and value added. The nations have literally traded places: China now has 31% of world manufacturing output while the U.S. has dropped to 16%. The U.S. has not just lost leadership in low-price commodity goods: As part of its massive trade deficit in 2023 of $733 billion in overall manufactured goods, the U.S. ran a deficit of $218 billion in advanced technology goods.

Declining U.S. manufacturing has sharply curtailed a key path to the middle class for those with high school educations or less, thereby exacerbating income inequality nationwide. As a country, we are increasingly leaving a large part of our working class behind an ever-advancing upper middle class. The problems plaguing the domestic manufacturing sector are multifold: American manufacturing productivity is historically low; the supporting ecosystem for small and midsized manufacturers has thinned out so they are slow to adopt process and technology innovations; manufacturing firms lack access to financing when they seek to scale up production for new innovations; manufacturing is poorly supported by our workforce-education system; and we have disconnected our innovation system from our production systems.

The United States can address many of these problems through concerted efforts in advanced manufacturing. Advanced manufacturing means introducing new production technologies and processes to significantly lower production costs and raise efficiency. This would reposition the United States to better compete internationally. Advanced manufacturing also requires that we reconnect innovation with production. A milestone in advanced manufacturing came in 2012, when the federal government established the first of 17 Advanced Manufacturing Institutes with two more planned. Each institute in this network is organized around developing new advanced technologies, from 3D printing to digital production to biofabrication. Each also represents a collaboration among industry, government, and academic institutions. Three federal agencies invest a total of approximately $200 million per year in the institutes—an amount matched by industry and states. 

The manufacturing institutes have proven successful to date. But they cannot reboot American manufacturing alone; other efforts are needed. Key U.S. trading partners and competitors spend far more on maintaining their manufacturing base and investing in advanced manufacturing compared to their GDP than the U.S. does. To restore U.S. leadership in manufacturing and rebuild manufacturing as a route to quality jobs for Americans, the federal government must double down on advanced manufacturing nationwide. Specifically, the next administration should:

More detail on these and related recommendations is provided below. 

Challenge

Obstacles facing U.S. manufacturing

The United States faces multiple obstacles related to manufacturing:

From 2000 to 2010, U.S. manufacturing employment fell precipitously, from about 17 million to under 12 million. While employment declined in all manufacturing sectors, those most prone to globalization (such as textiles and furniture) were especially affected. This is largely attributable to China’s competitive entry into manufacturing, which experts have estimated caused the loss of 2.4 million U.S. manufacturing jobs.

The U.S. manufacturing decline has adversely affected economic well-being in numerous historically industrialized regions—especially for the men without college degrees who have long led U.S. manufacturing employment. Full-year employment of men with a high school diploma but without a college degree dropped from 76% in 1990 to 68% in 2013, and the share of these men who did not work at all rose from 11% to 18%. While real wages have grown over recent decades for men and women with college degrees, they fell for men without college degrees. These trends have affected the working class overall and are particularly worrying for the socioeconomic mobility of minority populations in the United States. African Americans and Hispanics have long comprised a significant portion of the manufacturing workforce, and manufacturing jobs have long been a critical route to enter the middle class. With manufacturing’s decline, this avenue upward has significantly narrowed. While post-COVID workforce shortages slightly improved the economic position of the working class for the past two years, a deep gulf of economic inequality remains.

Opportunity

Although lagging behind countries such as Germany, Japan, Korea, Taiwan, and China in manufacturing innovation, the United States is still a world leader on R&D innovation overall. There is a valuable opportunity to leverage domestic capabilities in R&D innovation to bolster manufacturing capabilities in. A variety of cutting-edge technologies—including new sensor and control systems, big data and analytics, robotics, artificial intelligence (AI), complex simulation and modeling, advanced materials and composites, biofabrication, mass customization (the ability to produce small customized lots at mass-production costs through 3D printing and computerized controls), nanofabrication, and photonics—have potential applications in manufacturing. Using such technologies to create new advanced-manufacturing paradigms could transform manufacturing efficiency, productivity, and returns on investment. 

A national advanced manufacturing initiative would yield multiple benefits. Investment in advanced manufacturing could restore U.S. manufacturing leadership and therefore help employment; in addition, certain advanced-manufacturing technologies (e.g., 3D printing) could re-localize supply chains and generate additional jobs. Pursuing innovative manufacturing methods could improve production cost efficiency, enabling the United States to compete successfully with nations where labor costs are lower. Production innovation will also generate better products and create new product markets. Finally, robust domestic manufacturing capabilities are increasingly understood to be essential to national security. Advanced manufacturing constitutes a “leap ahead”: If the U.S. could lead on the mix of these new innovative technologies and related processes, it could again be a manufacturing leader. 

Plan of Action

A manufacturing transformation requires the following steps:

Recommendation 1. Create a new White House Office of Advanced Manufacturing to coordinate across agencies and develop a strategic plan for positioning the United States as a world leader in advanced manufacturing.

The White House Office of Advanced Manufacturing should work with the National Economic Council and the President’s Council of Advisors for Science and Technology (PCAST)—as well as industry, university, and government leaders—to develop a public/private national advanced manufacturing strategic plan. Staffing and support could come from Manufacturing USA (the network organization for the institutes) and the three federal agencies that fund the institutes (DOD, DOE, and DOC), as well as from other involved agencies. The plan should:

The strategy should also establish a mechanism and timeline for periodic updating of the plan.

Recommendation 2. Create a New Financing Mechanism for Scale-Up of Advanced Manufacturing

While venture capital has been a major force financing innovation in the U.S., it is now focused overwhelmingly on software, biotech, and various services sectors, not on “hard tech” – innovations that must be manufactured. This means there are few mechanisms to scale up manufacturing production in the U.S. 

Shifting to advanced manufacturing requires scale-up financing to support it. This is particularly a problem for small and midsized manufacturers who are reluctant to take the risk of significant investments in new advanced manufacturing technologies. We have a number of models for scale-up financing: The CHIPS Act, to reestablish U.S. manufacturing of advanced semiconductor chips because of national security concerns, provided grant authority of $39 billion for new fabrication facilities as well as low-cost financing and investment tax credits. Operation Warp Speed in 2020 used guaranteed contracts to COVID-19 vaccine makers to reduce their risks and assure production. Senator Chris Coons and colleagues have proposed an industrial finance corporation for innovative manufacturing. The DOD in 2022 created a new Office of Strategic Capital for technology scale up, although it appears focused more on national security rather than broader economic security goals. The Biden Administration proposed repurposing an established federal bank, the Ex-Im Bank, to provide manufacturing scale-up support along with its long-standing export financing role. The DOE’s Loan Programs Office has since 2005 provided financing for scale-up of new energy technologies. In addition, states have often provided financing to firms to support regional job creation. 

There are a range of possible manufacturing scale-up mechanisms. Because it is an existing bank fluent with industry lending and risk, Ex-Im appears a logical contender for significant expansion of its program into manufacturing scale-up financing. It has a broad authorization that would enable it to do so and has embarked on initial lending in this area. This uses the capability of an existing institution and avoids having to create a new one. Applying this and other mechanisms with expanded lending authority for advanced manufacturing scale up could be a solution to this serious manufacturing system gap. 

Recommendation 3. Improve the Advanced Manufacturing Institute Program 

A key part of the manufacturing transformation involves improving the federally funded advanced manufacturing institute program, also called Manufacturing USA (a.k.a. the institutes). The institutes, which were formed to help close the gap between R&D innovation and production innovation, involved the critical actors required for developing advanced manufacturing technologies: industry, universities, community colleges, and federal, state, and local government. 

Established based on recommendations from the industry- and university-led Advanced Manufacturing Partnership (AMP) in 2012, the 17 institutes are designed to create new production paradigms in different production areas, shared across the supply chains of large and small firms and across industry sectors. The institutes are partially funded by federal agencies: nine are funded by the Department of Defense (DOD), seven by the Department of Energy (DOE), and one (plus the two upcoming institutes funded through the CHIPS Act) by the Department of Commerce (DOC). Total federal funding for the institutes is currently approximately $200 million per year. Each federal dollar is typically matched by about two dollars from industry and state governments. 

Why institutes? One key reason is that the great majority of the U.S. manufacturing sector firms are small and midsized, producing nearly half of U.S. output, that—as noted above—typically do not perform in-house R&D and have difficulty accessing the production innovation they need to compete. Challenges facing small and midsize manufacturing firms became even greater in the 2000s, when U.S. manufacturing output stagnated and factories closed. Moreover, even larger firms with R&D capabilities need to collaborate to share the major risks and costs of transitioning to new production paradigms. The institutes address these challenges and needs by acting as test beds—providing a range of industries and firms with opportunities to collaborate on, test, and prove prototypes for advanced production technologies and processes. The institutes also help fill manufacturing talent gaps, training technical workers to use advanced technologies and to develop processes and routines for introducing advanced technologies into established production systems. And, as noted, they can bring the critical actors together (industry, academia, government). Any manufacturing technology program to work has to connect these actors. 

In short, the manufacturing institutes help fill gaps in the U.S. manufacturing innovation system by: 

Although technology development is a long-term project, the institutes have already delivered some notable results. Institute-supported achievements to date include:

These successes notwithstanding, the current institute program cannot solve the systemic problems plaguing U.S. manufacturing. The next administration should dramatically scale up and expand the role of the institutes as part of an effort to usher in a new era of advanced manufacturing nationwide.

The institutes’ collaborative, cost-shared, public/private model is designed to engage all critical stakeholders. But with U.S. manufacturing contributing $2.3 trillion to GDP, federal spending of just $200 million annually to support the institutes (even if those funds are matched by industry and state governments) will not make a large impact. Specific recommendations for the institutes are:

Increase funding for the Manufacturing USA institute program 

The institutes are delivering successes, but they are not funded at a level commensurate with their important role. The level of seed funding has significantly declined in the institutes supported by the DOE and DOD. While the DOC’s single institute has kept up its funding level, the others have faced reductions in the federal share after their initial terms of five years. This has limited, in particular, their programs with small and midsized companies and in workforce education. The federal investment per institute needs to return to well above the $50–70 million level initially awarded per institute over their initial five years. Given what our international competitors are investing, an institute program of at least $500 million a year is needed. The new national strategic plan for advanced manufacturing (described above) could guide funding requests. 

Action steps

The next president should seek additional funding for additional institutes, with a goal of increasing the total federal support for institutes of $500 million The next president should also ask other federal agencies with significant research budgets and mission stakes in the industrial economy—such as the National Aeronautics and Space Administration (NASA), the U.S. Department of Agriculture (USDA), and the Department of Health and Human Services (HHS)—to consider sponsoring institutes. Funding levels per institute should be determined by a combination of institute performance, national-strategy recommendations, and the particulars of proposed projects.

Instead of term-limiting institutes, the agencies have been establishing a formal process for determining whether an institute’s term should be extended, but they need to provide funding at least comparable to the institute’s initial term

The institutes were initially established with five-year fixed terms. But the job of the institutes is not done in five years—addressing the deep structural issues in U.S. manufacturing innovation will require sustained effort over decades. Continuing federal seed funding is still needed at least at levels of their initial five-year terms to leverage private sector investment. However, the agencies have been going in different directions on subsequent funding. Congress has recognized that the institutes should not face fixed terms. But this does not mean that all institute terms should be automatically renewed. The federal government should extend those that are working well, end those that aren’t, and require improvements in others. Like any experiment, the institutes will engender successes and failures. The institute network needs a governance process that recognizes that. 

Two of the three federal agencies that currently fund and oversee the institutes—DOD, and DOC—have been implementing performance metrics aligned with the strategic goals of the institutes. DOD and DOC have developed a formal evaluation process that each institute must go through when it approaches the end of its term and applies for term renewal. DOE is now following their lead, subject to availability of funding. This review process was specifically recommended by a National Academies study. Agencies should consider elements such as an institute’s progress on its technology-development roadmaps, the impact of its current and planned technology development, the reach of its workforce-education efforts, involvement of small and midsized firms, and continued support from and cost-sharing by industry and states. The evaluation process and criteria should be carefully developed such that they can be conducted by an impartial, third-party expert review team. Finally, the evaluation process should be as consistent as possible across the entire institute network.

Where the evaluation concludes that an institute is making adequate progress, the evaluation team should recommend that its funding be renewed for an additional five-year term. If progress is inadequate, the institute could be terminated or recommended for renewal contingent on specific improvements. In cases where an inadequate institute has responsibility for an essential technology area, the evaluation team could also recommend seeking different leadership and organizational changes. All evaluations—even those for institutes deemed to be making adequate progress—could provide recommendations for improvements.

It is not enough to establish a formal review process; if an institute is performing well, it should be extended for an additional term at funding levels at least comparable to the funding level it received for its initial term. The agencies have gone in different directions on this key point. DOC has extended its institute with full funding; DOD and DOE provide only limited funding to the extent that it is available. Given the gravity of U.S. manufacturing problems, this needs to change. 

Action steps

When institutes are meeting their mission goals, they should extend their terms at funding levels at least comparable to what these institutes initially received. 

Use the institutes to strengthen industry supply chains by bringing all supply-chain participants into demonstration facilities

Most of the institutes have established hands-on and virtual demonstration and design facilities accessible to small and midsized firms participating in the institutes. These facilities are important because smaller firms are less likely to adopt new production technologies unless they can see how those technologies would work within production lines, train employees in using new technologies, and estimate the potential efficiency gains new technologies could yield for them.

But many advanced-manufacturing technologies—such as digital-production technologies—cannot be implemented unless adopted by all participants in a given supply chain. The institutes need to bring in participants of industrial supply chains in addition to individual firms in order to disseminate advanced-manufacturing technologies most successfully.

Action steps

The institutes should develop programs whereby larger manufacturers can bring in other participants in their supply chains as new technologies become ready for adoption. Manufacturing USA should support supply-chain-level demonstration and testing using institute demonstration facilities. NIST’s Manufacturing Extension Partnership (MEP) programs should be asked to assist in these efforts given their access to small manufacturing firms. In general, the new administration should support expanded collaboration and system-wide thinking at the implementation stage of advanced-manufacturing technologies and ensure that new technologies and processed developed at institutes move deep into the supply chain, not just to larger firms.

Network the manufacturing institutes to package their different technology advances to be integrated and readily useable by manufacturers

Because companies will want to adopt a series of new technologies, collaboration across the institutes will likely be essential to improving workforce education. Cross-institute collaboration will also help private-sector participants build “factories of the future” that integrate multiple advanced-manufacturing technologies. The role of Manufacturing USA needs to be expanded to support collaboration in these areas as well as cost sharing, dissemination of best practices in intellectual property, membership organization, involvement of small and midsized firms, and joint access to facilities and equipment. Yet agency cooperation to sustain the network has been limited at the time when the need for a collaborative network is becoming clearer. Companies, particularly smaller companies, don’t want individual stovepiped technologies. They want robotics and digital production technologies and analytics plus new areas such as 3D printing—and they need to be interoperable. They don’t want one-off technologies. Presenting manufacturers with integrated packages of technologies with demonstrated efficiency increases will be key to adoption. Similarly, companies would like workforce skills packaged and integrated across new technology fields. 

Action steps

The next administration should expand the role of Manufacturing USA to help the institutes, participating companies, and the workforce as a whole confront the challenges of implementing advanced manufacturing. In particular integrating multiple advanced-manufacturing technologies from multiple institutes to be offered to manufacturers as a package will be key.

Recommendation 4. Bring the strength of the U.S. early-stage R&D innovation system to bear on advanced manufacturing

Federal R&D agencies do not have significant research portfolios around advanced manufacturing technologies and processes, contributing to a serious innovation gap in U.S. manufacturing. While federal research agencies have made massive contributions to information technologies and life science advances, for example, they have not focused on manufacturing technologies. Given our societal challenge in manufacturing, they need to develop this approach. These new research efforts must be connected to and involve industry to help identify the most significant technology opportunities. Because the institutes work on applied and later-stage research, they rely on “feeder systems” for early-stage research. When the institutes aren’t connected closely enough to early-stage research systems—when they have limited knowledge of what research is being carried out and limited capacity to inform the research agenda—they risk “stranded technology” problems. In other words, they will be limited in their ability to capitalize on new results and/or to keep developing concepts progressing. The incoming administration should strive to better link the federal basic-research system (including DOD’s 6.1–6.3-level research programs, NSF’s engineering and computing research and its new directorate for Technology, Innovation and Partnerships, and DOE research at ARPA-E, OS, and EERE) both to industry and to the institutes and their technology focus areas. 

Action steps

The next administration should work through OSTP, with its agency-convening authority, to require federal entities responsible for basic research to develop and implement significant research portfolios around advanced manufacturing. Industry should advise and participate in this research. In addition, OSTP, in cooperation with the proposed White House Advanced Manufacturing Office, should form joint R&D agency and manufacturing institute planning and roadmapping processes to support the institutes’ technology focus areas. Such an effort will assist agencies in developing and highlighting research activities that complement the institutes’ technology-development activities, and vice versa. 

Recommendation 5. Link new manufacturing technologies emerging from the R&D agencies and the manufacturing institutes to acquisition by the Department of Defense 

DOD has a massive annual procurement program; this program could be used to introduce advanced production technologies and processes that could help DOD lower costs and improve system performance. DOD has done this before. When computer numerically controlled (CNC) equipment was developed at MIT in the 1950s, DOD saw that it could significantly improve the precision and performance in missiles and other technologies. It therefore required its manufacturing suppliers to adopt CNC equipment and helped finance its introduction at these firms. Application at these firms spread CNC equipment to virtually all manufacturers today. We could use this same approach today to get advanced manufacturing technologies in place.

DOD did not select the technology focus areas for its nine manufacturing institutes by accident; these technologies support its industrial base and new systems need them. Thus, the institutes are critical for ensuring that DOD possesses the domestic capacity to produce new innovations at scale. Technologies produced by non-DOD-funded institutes are often relevant to DOD as well. For example, the power electronics coming out of a DOE-funded institute will yield not only improved energy efficiency but improved electronics and power systems in general, which are important to DOD. Another DOE-funded institute is developing advanced composites that could significantly improve DOD operating platforms.

Unlike other agencies, DOD not only has a major acquisition system, but it is connected to its R&D system—it can research, develop, and build new technologies. Most private-sector manufacturing firms, particularly smaller firms, tend to be risk- and cost-averse and are hence often reluctant to lead on production in new areas. DOD can fill this gap by using its acquisition system to support testing, design prototyping, and initial procurement for new technologies coming out of the institutes. This would benefit the nation by jump-starting deployment of emerging innovations and would benefit DOD by providing the agency with early access to technologies not yet available on the private market. 

Action steps

The incoming administration should direct DOD to (1) review its relevant demonstration, testing, and acquisition processes so they can be used for implementing advanced manufacturing technologies, (2) identify specific options for the agency to leverage these processes to procure emerging manufacturing technologies it needs, including from the institutes, and (3) identify changes to existing regulations and systems that would help link DOD acquisition with advanced manufacturing innovation emerging from R&D agencies, industry and institutes. ( For example, DOD may be able to reinstitute a form of its industrial/modernization incentive program or apply its Defense Production Act Title III authorities.) The next administration should then take prompt action to implement recommendations arising from the review. 

Recommendation 6. Work to create a new workforce education system designed to prepare workers for jobs in advanced manufacturing

Germany has long gained productivity improvements from a famously well-trained manufacturing workforce based on an apprenticeship system. In contrast, U.S. companies have generally tried to get productivity gains from capital plant and equipment investments and ignored the workforce side. German firms understand, however, that productivity gains from new equipment will soon spread worldwide, while a gain from a high-quality workforce will be enduring and provide a long-term competitive edge. Like Germany, the U.S. needs both inputs. 

Unfortunately, American workforce-education systems are largely broken. The U.S. has a deep disconnect between the education system and workplaces, which lies at the heart of many of its workforce problems. (See detailed discussion of these issues and corresponding recommendations in Bonvillian and Sarma, Workforce Education.) Other causes include:

Complicating efforts to establish new and improved workforce-education systems is the fact existing systems depend heavily on actors in complex, established “legacy” sectors that are hard to change. At the federal level, only a modest NSF program in Advanced Technological Education (ATE), through community colleges, provides education and training in advanced manufacturing. The manufacturing institutes are also developing workforce education efforts at both the technician and engineering levels, but these are also modest in scope. NIST’s Manufacturing Extension Partnership is encouraging new workforce programs for its SME participants. Neither the Department of Labor (DOL) nor the Department of Education (DOEd), however, has a program dedicated to education or training in advanced manufacturing. These existing programs need to be strengthened. The institutes, for example, through their unique blend of academic, public, and private-sector participation, are well positioned to help build a skilled advanced-manufacturing workforce. The institutes also have the deep technical expertise needed to effectively guide the content and structure of new workforce-education manufacturing modules in emerging technologies. 

If the U.S. wants to adopt advanced manufacturing, its workforce must be ready for it. This requires rebuilding much of workforce education. Reforms are needed at all levels. Community colleges must introduce advanced manufacturing curricula, create short programs more adapted to upskilling workers already in the workforce, establish certificates around particular skills that stack toward degrees, and turn around low completion rates. At the federal level, disconnected Labor and Education Department programs need to be integrated and efforts to expand registered apprenticeship programs accelerated. At the industry level, firms must collaborate with each other and with community colleges to build new training and apprenticeship programs, including youth apprenticeships starting in high school. Solutions will have to be pursued in the U.S. federal system, since education is largely state and local government-led. These federal agencies need to improve community college funding and support, integrate advanced manufacturing programs across community colleges, promote apprenticeships and work closely with area firms on reforms and curricula. 

Action steps

The White House Office of Advanced Manufacturing (OAM) proposed above, working with Manufacturing USA, DOD, DOL, DOEd, NSF ATE, and NIST, should launch a coordinated effort to identify best practices in workforce education at the involved agencies, institutes, and elsewhere, including for online education. OAM and Manufacturing USA should also develop an education “commons” of shared advanced-manufacturing courses, modules, and materials. OAM, the workforce education agencies noted above and Manufacturing USA should ensure that current and future workforce-education efforts are coordinated for a unified cross-agency effort. Working with existing manufacturing skills standards groups, OAM and these agencies will likely also need to establish standards for certifications in advanced manufacturing fields, so that certifications can be earned at one place and recognized at another. Manufacturing USA Institutes can play an important role because of their expertise in their specific technology areas. Expert teams will need to be assembled to develop these training resources and standards, as well as to evaluate their effectiveness. Because workforce education tends to suffer from the “tragedy of the commons”—many want it but few want to pay for it—the incoming administration should support federal funding for all of the above efforts and ensure that relevant participating agencies include these efforts in their budget requests. 

Recommendation 7. Develop an ongoing assessment of advanced production capabilities in critical technologies emerging in other nations

In 2023, pursuant to the CHIPS and Science Act, NSF sponsored a study commission to develop a pilot program to assess international innovation and manufacturing capability in critical technology sectors. The U.S. doesn’t undertake such a systematic critical technology assessment, and it has become a major gap in our capabilities and understanding. We are integrated into an  international economy that we helped create and now face massive trade deficits in manufactured and advanced technology goods. The U.S. is failing to track our increasingly innovative competitors and their implementation of emerging technologies. In effect, in a highly competitive world the U.S. is flying blind. The report provides a framework for undertaking this kind of assessment. Adoption of a critical technology assessment of competitor nations and the U.S.’s comparative status is important to both national and economic security; we need a benchmarking system on where we stand and the new approaches others are adopting that we need to pursue. 

If located in the intelligence or defense agencies, this would help ensure longer-term sustainable support, and these agencies could also collaborate with key non-defense agencies. Since national security is now so closely tied to economic security, this is an appropriate role for intelligence or defense organizations. It is important to have a sustained assessment effort led by a talented staff team that can build expertise in the complex assessment process, which is also connected to draw on university and industry experts.

Action steps

The next administration should direct the Director of National Intelligence or DOD (working with DOC, NSF, and DOE, which also have relevant technical capabilities) to conduct an ongoing assessment of the progress of other leading nations in critical technologies and advanced manufacturing. The assessment should examine the strategic goals, internal organization, and funding levels of international advanced-manufacturing initiatives. Such an assessment should emphasize aspects of advanced manufacturing where the U.S. industrial base, including from a national security perspective, has a significant stake in future technology, such as advanced materials, semiconductors and computing, composites, photonics, functional fibers, power electronics, biofabrication, and a suite of digital tools that are finding applications in manufacturing and other sectors (e.g., AI, machine learning, the Internet of Things, robotics, simulations and modeling, data analytics, and quantum sensors and computing). Such an assessment would help us understand the status of critical elements of its industrial base and inform the focus areas and technology-development agendas of the various institutes and agencies. The assessment would also benefit the United States as a whole by guiding national manufacturing strategy and ensuring that the institutes are used to maximize the country’s global competitiveness in manufacturing.

Implementation

The need to bolster the U.S. manufacturing has become a political imperative for both parties. Many aspects of the recommendations detailed above could be implemented quickly by presidential directives and would not require legislation. Increasing funding for the institutes and advanced manufacturing generally is, of course, the exception. But congressional approval for increased advanced manufacturing funding seems likely. Despite the sharp political divides of the past decade, Congress overwhelmingly passed bipartisan legislation in 2014 authorizing the institutes and amended that legislation in 2019—a strong political signal of political support around this issue. Congress has also been willing to back advanced manufacturing in appropriations bills each year. A bipartisan congressional manufacturing caucus, and deep understanding by a number of key congressional figures of issues related to advanced manufacturing, provide a solid foundation of expected legislative-branch support for executive-branch actions to further advanced manufacturing.

Conclusion

Production plays a disproportionate role in U.S. economic well-being. As international competitors move rapidly on advanced manufacturing while U.S. manufacturing capabilities stagnate, the U.S. economy is increasingly vulnerable. These are national security issues as well as economic security issues, and the two are increasingly merged. Public-private models, (particularly Manufacturing USA and the manufacturing institutes), organized around advanced manufacturing, offer a promising model for helping reverse these trends and restoring U.S. leadership in manufacturing. The institutes bring together the key actors: industry, universities, community colleges, and government (federal, state, and local). But the institutes as they now stand simply do not have the capacity to affect the U.S. manufacturing sector at the scale needed. 

The recommendations detailed here provide the next administration with a roadmap for launching a nationwide effort to strengthen advanced manufacturing—an effort that builds on the institutes’ successes and significantly expands their capabilities, roles, and impacts. Briefly summarized, the recommendations are:

Douglas Brinkley’s book American Moonshot tells how, in 1961, President Kennedy mobilized the American public around a new space mission. The mission was rationalized in part on Cold War competition but also on the dramatic mission-related technology advances—from communication satellites to STEM education to computing—that the president argued would (and did) boost the economy. The direct tie between advanced manufacturing and the future of the American economy is, frankly, far more visible to the public than the space race. Strong presidential leadership could unify public support around a shared goal of manufacturing leadership and building quality jobs in a period of political fracture. There is a dramatic competitive aspect: China has already passed the United States on manufacturing output while we as a nation play catch-up, and China’s increased economic power and the corresponding. decline in the U.S. share of world manufacturing output have important implications for the future of democracy and world leadership. Furthering advanced manufacturing in the United States also involves rethinking and rebuilding our workforce-education systems, another potentially highly popular imperative. And, finally, advanced manufacturing includes not only government but industry, universities, community colleges, and nonprofits as well. In short, advanced manufacturing can unite nearly all American institutions—and nearly all Americans.

This action-ready policy memo is part of Day One 2025 — our effort to bring forward bold policy ideas, grounded in science and evidence, that can tackle the country’s biggest challenges and bring us closer to the prosperous, equitable and safe future that we all hope for whoever takes office in 2025 and beyond.

PLEASE NOTE (February 2025): Since publication several government websites have been taken offline. We apologize for any broken links to once accessible public data.

Kickstarting Collaborative, AI-Ready Datasets in the Life Sciences with Government-funded Projects

In the age of Artificial Intelligence (AI), large high-quality datasets are needed to move the field of life science forward. However, the research community lacks strategies to incentivize collaboration on high-quality data acquisition and sharing. The government should fund collaborative roadmapping, certification, collection, and sharing of large, high-quality datasets in life science. In such a system, nonprofit research organizations engage scientific communities to identify key types of data that would be valuable for building predictive models, and define quality control (QC) and open science standards for collection of that data. Projects are designed to develop automated methods for data collection, certify data providers, and facilitate data collection in consultation with researchers throughout various scientific communities. Hosting of the resulting open data is subsidized as well as protected by security measures. This system would provide crucial incentives for the life science community to identify and amass large, high-quality open datasets that will immensely benefit researchers.

Challenge and Opportunity 

Life science has left the era of “one scientist, one problem.” It is becoming a field wherein collaboration on large-scale research initiatives is required to make meaningful scientific progress. A salient example is Alphafold2, a machine learning (ML) model that was the first to predict how a protein will fold with an accuracy meeting or exceeding experimental methods. Alphafold2 was trained on the Protein Data Bank (PDB), a public data repository containing standardized and highly curated results of >200,000 experiments collected over 50 years by thousands of researchers.

Though such a sustained effort is laudable, science need not wait another 50 years for the ‘next PDB’. If approached strategically and collaboratively, the data necessary to train ML models can be acquired more quickly, cheaply, and reproducibly than efforts like the PDB through careful problem specification and deliberate management. First, by leveraging organizations that are deeply connected with relevant experts, unified projects taking this approach can account for the needs of both the people producing the data and those consuming it. Second, by centralizing plans and accountability for data and metadata standards, these projects can enable rigorous and scalable multi-site data collection. Finally, by securely hosting the resulting open data, the projects can evaluate biosecurity risk and provide protected access to key scientific data and resources that might otherwise be siloed in industry. This approach is complementary to efforts that collate existing data, such as the Human Cell Atlas and UCSD Genome Browser, and satisfy the need for new data collection that adheres to QC and metadata standards.

In the past, mid-sized grants have allowed multi-investigator scientific centers like the recently funded Science and Technology Center for Quantitative Cell Biology (QCB, $30M in funding 2023) to explore many areas in a given field. Here, we outline how the government can expand upon such schemes to catalyze the creation of impactful open life science data. In the proposed system, supported projects would allow well-positioned nonprofit organizations to facilitate distributed, multidisciplinary collaborations that are necessary for assembling large, AI-ready datasets. This model would align research incentives and enable life science to create the ‘next PDBs’ faster and more cheaply than before.  

Plan of Action 

Existing initiatives have developed processes for creating open science data and successfully engaged the scientific community to identify targets for the ‘next PDB’ (e.g., Chan Zuckerberg Initiative’s Open Science program, Align’s Open Datasets Initiative). The process generally occurs in five steps:

  1. A multidisciplinary set of scientific leaders identify target datasets, assessing the scale of data required and the potential for standardization, and defining standards for data collection methods and corresponding QC metrics.
  2. Collaboratively develop and certify methods for data acquisition to de-risk the cost-per-datapoint and utility of the data.
  3. Data collection methods are onboarded at automation partner organizations, such as NSF BioFoundries and existing National Labs, and these automation partners are certified to meet the defined data collection standards and QC metrics.
  4. Scientists throughout the community, including those at universities and for-profit companies, can request data acquisition, which is coordinated, subsidized, and analyzed for quality.
  5. Data becomes publicly available and is hosted in a stable, robustly maintained database with biosecurity, cybersecurity, and privacy measures in perpetuity for researchers to access. 

The U.S. Government should adapt this process for collaborative, AI-ready data collection in the life sciences by implementing the following recommendations:  

Recommendation 1. An ARPA-like agency — or agency division — should launch a Collaborative, AI-Ready Datasets program to fund large-scale dataset identification and collection.

This program should be designed to award two types of grants:

  1. A medium-sized “phase 1” award of $1-$5m to fund new dataset identification and certification. To date, roadmapping dataset concepts (Steps 1-2 above) has been accomplished by small-scale projects of $1-$5M with a community-driven approach. Though selectively successful, these projects have not been as comprehensive or inclusive as they could otherwise be. Government funding could more sustainably and systematically permit iterative roadmapping and certification in areas of strategic importance.
  2. A large “phase 2” award of $10-$50m to fund the collection of previously identified datasets. Currently, there are no funding mechanisms designed to scale up acquisition (Steps #3-4 above) for dataset concepts that have been deemed valuable and derisked. To fill this gap, the government should leverage existing expertise and collaboration across the nonprofit research ecosystem by awarding grants of $10-50m for the coordination, acquisition, and release of mature dataset concepts. The Human Genome project is a good analogy, wherein a dataset concept was identified and collection was distributed amongst several facilities.

Recommendation 2. The Office of Management and Budget should direct the NSF and NIH to develop plans for funding academics and for-profits traunched on data deposition.

Once an open dataset is established, the government can advance the use and further development of that dataset by providing grants to academics that are traunched on data deposition. This approach would be in direct alignment with the government’s goals for supporting open, shared resources for AI innovation as laid out in section 5.2 of the Executive Order on the Safe, Secure, and Trustworthy Development and Use of Artificial Intelligence. 

Agencies’ approaches to meeting this priority could vary. In one scenario, a policy or program could be established in which grantees would use a portion of the funds disbursed to them to pay for open data acquisition at a certified data provider. Analogous structures have enabled scientists to access other types of shared scientific infrastructure, such as the NSF’s ACCESS program. In the same way that ACCESS offers academics access to compute resources, it could be expanded to offer academic access to data acquisition resources at verified facilities. Offering grants in this way would incentivize the scientific community to interact with and expand upon open datasets, as well as encourage compliance through traunching. 

Efforts to support use and development of open, certified datasets could also be incorporated into existing programs, including the National AI Research Resource, for which complementary programs could be developed to provide funding for standardized data acquisition and deposition. Similar ideas could also be incorporated into core programs within NSF and NIH, which already disburse funds after completion of annual progress reports. Such programs could mandate checks for data deposition in these reports.

Conclusion 

Collaborative, AI-Ready datasets would catalyze progress in many areas of life science, but realizing them requires innovative government funding. By supporting coordinated projects that span dataset roadmapping, methods and standards development, partner certification, distributed collection, and secure release on a large scale, the government can coalesce stakeholders and deliver the next generation of powerful predictive models. To do so, it should combine small-sized, mid-sized, and traunched grants in unified initiatives that are orchestrated by nonprofit research organizations, which are uniquely positioned to execute these initiatives end-to-end. These initiatives should balance intellectual property protection and data availability, and thereby help deliver key datasets upon which new scientific insights depend.

This action-ready policy memo is part of Day One 2025 — our effort to bring forward bold policy ideas, grounded in science and evidence, that can tackle the country’s biggest challenges and bring us closer to the prosperous, equitable and safe future that we all hope for whoever takes office in 2025 and beyond.

PLEASE NOTE (February 2025): Since publication several government websites have been taken offline. We apologize for any broken links to once accessible public data.

Frequently Asked Questions
What is involved in roadmapping dataset opportunities?

Roadmapping dataset opportunities, which can take up to a year, requires convening experts across multiple disciplines, including experimental biology, automation, machine learning, and others. In collaboration, these experts assess both the feasibility and impact of opportunities, as well as necessary QC standards. Roadmapping culminates in determination of dataset value — whether it can be used to train meaningful new machine learning models.

Why should data collection be centralized but redundant?

To mitigate single-facility risk and promote site-to-site interoperability, data should be collected across multiple sites. To ensure that standards and organization holds across sites, planning and documentation should be centralized.

How should automation partners be certified?

Automation partners will be evaluated according to the following criteria:



  • Commitment to open science

  • Rigor and consistency in methods and QC procedures

  • Standardization of data and metadata ontologies


More specifically, certification will depend upon the abilities of partners to accommodate standardized ontologies, capture sufficient metadata, and reliably pass data QC checks. It will also require partners to have demonstrated a commitment to data reusability and replicability, and that they are willing to share methods and data in the open science ecosystem.

Should there be an embargo before data is made public?

Today, scientists have no obligation to publish every piece of data they collect. In an Open Data paradigm, all data must eventually be shared. For some types of data, a short, optional embargo period would enable scientists to participate in open data efforts without compromising their ability to file patents or publish papers. For example, in protein engineering, the patentable product is the sequence of a designed protein, making immediate release of data untenable. An embargo period of one to two years is sufficient to alleviate this concern and may even hasten data sharing by linking it to a fixed length of time after collection, rather than to publication. Whether or not an embargo should be implemented and its length should be determined for each data type, and designed to encourage researchers to participate in acquisition of open data.

How do we ensure biosecurity of the data?

Biological data is a strategic resource and requires stewardship and curation to ensure it has maximum impact. Thus, data that is generated through the proposed system should be hosted by high-quality providers that adhere to biosecurity standards and enforce embargo periods. Appropriate biosecurity standards will be specific to different types of data, and should be formulated and periodically reevaluated by a multidisciplinary group of stakeholders. When access to certified, post-embargo data is requested, the same standards will apply as will export controls. In some instances, for some users, restricting access may be reasonable. For offering this suite of valuable services, hosting providers should be subsidized through reimbursements.

Launch the Next Nuclear Corps for a More Flexible Nuclear Regulatory Commission

The Nuclear Regulatory Commission (NRC), the Nation’s regulator of civilian nuclear technologies, should shift agency staff, resources, and operations more flexibly based on emergent regulatory demands. The nuclear power industry is demonstrating commercialization progress on new reactor concepts that will challenge the NRC’s licensing and oversight functions. Rulemaking on new licensing frameworks is in progress, but such regulation will fall short without changes to the NRC’s staffing. Since the NRC is exempt from civil service laws under the Atomic Energy Act (AEA) of 1954, the agency should use AEA flexible hiring authorities to launch the Next Nuclear Corps, a staffing program to shift capacity based on emergent, short-term workforce needs. The NRC should also better enable hiring managers to meet medium-term workforce needs by clarifying guidance on NRC’s direct hire authority.

Challenge and Opportunity

Policymakers, investors, and major energy users, such as data centers and industrial plants, are interested in new nuclear power because it promises unique value. New nuclear power technologies could add either additional base load or variable power to electrical systems. Small modular or micro reactors could provide independent power to military bases, many of which are connected to power grids and vulnerable to disruption. Local governments can stimulate economies with high-paying and safe jobs at nuclear plants. The average nuclear power plant also has the lowest lifecycle greenhouse gas emissions compared to other available electricity-generating technologies, including wind, solar, and hydropower. Current efforts to expand nuclear power are different from those of the 1970s and 1980s, the most recent decades of significant building. Proposals today include building plants designed similarly to plants of those decades or even restarting power operations at up to three closed plants; but more activity is focused on commercializing advanced and small modular reactors, diverse concepts incorporating innovations in reactor design, fuel types, and safety systems. The government has partnered with private companies to develop and demonstrate advanced reactors since the inception of nuclear technology in the 1950s, but today several companies demonstrate advanced technical and business progress toward commercialization. 

Innovation in nuclear power challenges the NRC’s status-quo approaches to licensing and oversight. Rulemaking on new regulatory frameworks is necessary and in progress, but changes to the agency’s staffing and operations are also needed. Over time, Congress, the President, and the Commission itself have adjusted the agency’s operations in response to shifts in international postures, comprehensive national energy plans, and accidents or emerging threats at nuclear plants, but the NRC’s ability to respond to sudden changes in the nuclear industry is a long-standing challenge. To become more flexible, NRC initiated Project Aim in 2014 after expectations of significant industry growth, spurred in part by tax incentives in the Energy Policy Act of 2005, were not realized due to record-low natural gas prices. More recent assessments from the Government Accountability Office (GAO) and NRC Office of Inspector General (OIG) acknowledge the challenge of workload forecasting in an unpredictable nuclear industry, but counterintuitively, some recommendations focus on improving the ability to workforce plan two years or more in advance. Renewed expectations of growth, spurred by interest from policymakers and energy customers, reinforces a point from the 2015 Project Aim final report that, “…effectiveness, efficiency, agility, flexibility, and performance must improve for the agency to continue to succeed in the future.”

Congress also called on the NRC to become more responsive to current developments as expressed in legislation enacted with bipartisan support. Across the Fiscal Responsibility Act of 2023 and the ADVANCE Act of 2024, Congress requires the NRC to update its mission statement to better reflect the benefits of civilian nuclear technology, establish regulatory frameworks for new technology, streamline environmental review, incentivize licensing of advanced nuclear technologies, and position itself and the United States as a leader in civilian nuclear power. Meeting expectations requires significant operational and workforce changes. Since NRC is exempt from civil service laws and operates an independent competitive merit system, widespread changes to the agency’s hiring practices will be determined by future Commissioners, including the President’s selection of Chair (and by extension, the Chair’s selection of the Executive Director for Operations (EDO)), and modifications to agreements between the NRC and the Office of Personnel Management (OPM). In the meantime, NRC is well equipped to increase hiring flexibility using authorities from existing law and regulations.

Plan of Action

Recommendation 1. The NRC EDO should launch the Next Nuclear Corps, a staffing program dedicated to shifting agency capacity based on short-term workforce needs.

The EDO should hire a new director to lead the Corps. The Corps director should report to the EDO and consult with the Office of the Chief Human Capital Officer (OCHCO) and division heads to develop Corps positions to address near-term priorities in competency areas that do not require in-depth training. Near-term priorities should be informed by the NRC’s existing yearly capacity assessments, but the Corps director should also rely on direct expertise and insights from branch chiefs who have a real-time understanding of industry activity and staffing challenges.

Recommendation 2. Hiring for the Corps should be executed under the special authority to appoint directly to the excepted service under 161B(a) of the Atomic Energy Act (AEA).

The ADVANCE Act of 2024 created new categories of hires to fill critical needs related to licensing, oversight, and matters related to NRC efficiency. The EDO should execute the Corps under the new authorities in section 161B(a) of the AEA as it provides clear direction and structure for the EDO to make personnel appointments outside of the NRC’s independent competitive merit system described in Management Directive 10.1. 161B(a)(A) provides up to 210 hires at any time and 161B(a)(B) provides up to 20 additional hires each fiscal year which are limited to a term of four years. The standard service term should be one year as near-term workforce needs may be temporary because of the nature of the position or uncertainty in future demand.

The EDO should adopt the following practices to allow renewals of some positions from the prior year without reaching the limits described in the AEA:

Recommendation 3. The EDO should update Management Directives 10.13 and 10.1 to contain or reference the standard operating procedure for NRC’s mirrored version of OPM’s Direct Hire Authority.

The proposed Corps addresses emergent, short-term capacity needs, but internal policy clarity is needed to solve medium-term hiring challenges for hard-to-recruit positions. As far back as 2007, NRC hiring managers and human resources reported that DHA was highly desired for hiring flexibility. The NRC OIG closed Recommendation 2.1 from Audit of the U.S. Nuclear Regulatory Commission’s Vacancy Announcement Process in June 2024 because NRC updated Standard Operating Procedure for Direct Hire Authority with more details. However, management directives are the primary policy and procedure documents that govern the NRC’s internal functions. The EDO should update management directives to formally capture or reference this procedure so that NRC staff are better equipped to use DHA. Specifically, the EDO should:

Conclusion

The potential of new nuclear power plants to meet energy demand, increase energy security, and revitalize local economies depends on new regulatory and operational approaches at the NRC. Rulemaking on new licensing frameworks is in progress, but the NRC should also use AEA flexible hiring authorities to address emergent, short-term workforce needs that may be temporary based on shifting industry developments. The proposed Corps structure allows the EDO to quickly hire new staff outside of the agency’s competitive merit system for short-term needs while preserving flexibility to renew appointments if the capacity needs continue. For permanent hard-to-recruit positions, the EDO should clarify guidance for hiring managers on direct hire authority. The NRC is well equipped with existing authorities to meet emergent regulatory demand and renewed expectations of nuclear power growth.

This action-ready policy memo is part of Day One 2025 — our effort to bring forward bold policy ideas, grounded in science and evidence, that can tackle the country’s biggest challenges and bring us closer to the prosperous, equitable and safe future that we all hope for whoever takes office in 2025 and beyond.

PLEASE NOTE (February 2025): Since publication several government websites have been taken offline. We apologize for any broken links to once accessible public data.

Frequently Asked Questions
What types of jobs would be filled by the Corps?

The Corps director should create positions informed by the expertise and insights from agency leaders who have a real-time understanding of industry activity and present staffing challenges. Positions should cover all career levels and cover competency areas that do not require in-depth internal training or security clearances. The Corps should fill new positions created for special roles in support of other staff or teams, such as special coordinators, specialists, and consultants.

How is this different from the Presidential Management Fellows Program or similar programs?

The Corps is not a graduate-level fellowship or leadership development program. The Corps is specifically for short-term, rapid hiring based on emergent capacity needs that may be temporary based on the nature of the need or uncertainty in future demand.

Can Corps hires transition to permanent positions?

The Corps structure includes flexibility for a limited number of renewals, but it is not intended to recruit for permanent positions. Supervisors and hiring managers could choose to coordinate with the OCHCO to recruit off-boarding Corps members to other employment opportunities.

What is the recruiting strategy for the Corps?

The Corps director can identify talent through existing NRC recruiting channels, such as job fairs, universities, and professional associations, however, the Corps director should also establish new recruiting efforts through more competitive channels. Because the positions are temporary, the Corps can recruit from more competitive talent pools, such as talent seeking long term careers in private industry. Job seekers with long-term ambitions in the private nuclear sector and the NRC could both benefit from a one- or two-year period of service focused on a specific project.

Ready for the Next Threat: Creating a Commercial Public Health Emergency Payment System

There are many examples of groundbreaking success in the development of life-saving vaccines, diagnostics, and therapeutics to which we can point in the response to the COVID-19 pandemic, such as the introduction of mRNA vaccines and the accelerated path to its authorization through Operation Warp Speed. However, in anticipation of future known and unknown health security threats, including new pandemics, biothreats, and climate-related health emergencies, our answers need to be much faster, cheaper, and less disruptive to other operations. One path to a more permanent state of readiness is to create a commercial public health emergency payment system to use the full power of commercial healthcare reimbursement, providing clear and tunable market signals to catalyze investment in anticipation and in response to public health emergencies.

Challenge and Opportunity

There are two key problems we have yet to solve. First, how do we break out of the panic/neglect cycle of investments in emergency medical countermeasures (MCMs), i.e invest more in prevention and preparedness than in response? Secondly, how can we avoid any friction in capital (both public and private) once there is an emergency, while preserving the ability to fine tune incentives as needs evolve?

Many of the most promising and impactful tools applicable to emerging infectious diseases and public health emergency (PHE) management more broadly (e.g. wastewater surveillance, home testing, vaccinations, improved indoor air quality), face strong headwinds due to small, poorly defined, and/or unstable markets, significantly reducing private investment in them and relying on seemingly stochastic public investments by a fragmented set of federal, state, and local agencies.

Additionally, there is a prevailing assumption within the health security community that it operates largely outside the commercial U.S. healthcare system due to lack of private incentives (as opposed to, for instance, the development and use of cancer therapies). This, however, reflects a policy decision and not fundamental underlying market demand. And yet there remain two key realities to contend with: pandemic-related demand is largely unpredictable, and we have, thus far, been unable to effectively amortize pandemic costs into interpandemic periods.

U.S. health care insurers process more than nine  billion claims for payment each year  – a process which features a sophisticated, standardized accounting system that is widely understood by the entire healthcare industry; it is also a powerful signal of future market expectations that drives private and public R&D investment decisions.

Plan of Action 

The U.S. government should create a prospective healthcare reimbursement code set that can anticipate the need for any product or service in the context of PHEs, including MCMs and infrastructure-based countermeasures. The goal would be to provide clear market signals and pull incentives, to encourage and accelerate development and appropriate utilization of medical countermeasures (diagnostics, vaccines, therapies, early warning detection, and others). This would complement other strategies, such as advanced R&D investments made by federal funding agencies including the Biomedical Advanced Research & Development Authority (BARDA). Creating clear reimbursement pathways would likely immediately attract private investment in MCMs in ways that are notoriously difficult currently and help enable rapid response to public health emergencies.

The development and management of this reimbursement system should be housed within the Centers for Medicare and Medicaid Services (CMS), and would require introduction of legislation clearly authorizing CMS to pay for products and services under EUA and use of prospective payment policies. 

There are a variety of additional benefits to using a payment system like this for emergency response. This system could be a unique and core source of surveillance data, through conditions of payment policies, that can be used to provide intelligence and manage evolving emergencies such as outbreaks and pandemics, significantly reducing the need for additional data collection systems – need which proved to be a major bottleneck during COVID-19. Prospective codesets are not common in public and private insurance, but the existence of them would likely serve as a new and powerful tool for private investment in a capability that would be certain to have public health benefit were it to exist.

We propose that reimbursable services be categorized into three tiers.

Tier 1. Existing and prospective medical products/services under FDA EUA

This is the set of regulated medical products that would typically be considered for emergency use authorization by the FDA. It can include infectious disease diagnostics, vaccinations, and therapeutics. It can even include the use of repurposed generic drugs to mitigate potential drug shortages.

Tier 2. Existing public health products/services typically not requiring FDA approvals, but may have other regulatory hurdles

There are a variety of products and services not considered medical products but nonetheless play critical roles not only in response but in prevention of public health emergencies. Many of these technologies struggle to find stable markets in which to operate and are relegated to the sidelines. This includes genomic surveillance, remnant sample sequencing, or other surveillance testing capabilities, early warning systems, use of commercial wearables as a check engine light for possible infection prior to symptom onset. Many of these technologies need to be “always on” to be most effective.

Tier 3. Prospective public health products/services typically not requiring FDA approvals, but may have other regulatory hurdles

This category deviates from current models of healthcare reimbursement, because it is comprised of interventions that are carried out by service providers outside of healthcare delivery, but which nonetheless have high impact potential. This can include indoor air quality upgrades, or wastewater detection. Again, the commercialization of indoor air quality is in part impeded by poorly defined markets. 

Value-based care models

While the above items are largely “fee for service” payment models, we can also envision the use of value-based care models here, focusing on community outcomes and providing flexibility to communities or other systems to achieve them. This can include models to prevent hospitalization due to PHE (e.g. COVID-19), prevent community transmission (could be directed at local public health agencies or other agents), herd immunity vaccination rates, etc.

Conclusion 

One of the biggest frustrations amongst the life science and medical device communities in the COVID-19 response was that the government was not clear about target product profiles and advanced market commitments for the full range of products and services needed. These types of systems, if implemented early, would have sent clear and powerful signals to the market that would have quickly unlocked the necessary private sector investment needed to expedite product development needed for the response. Using the already widely adopted and understood commercial healthcare reimbursement system to incentivize and pay for prospective emergency product development and delivery is a novel, powerful, and turnkey approach to pandemic preparedness. 

This action-ready policy memo is part of Day One 2025 — our effort to bring forward bold policy ideas, grounded in science and evidence, that can tackle the country’s biggest challenges and bring us closer to the prosperous, equitable and safe future that we all hope for whoever takes office in 2025 and beyond.

PLEASE NOTE (February 2025): Since publication several government websites have been taken offline. We apologize for any broken links to once accessible public data.

From Strategy to Impact: Establishing an AI Corps to Accelerate HHS Transformation

To unlock the full potential of artificial intelligence (AI) within the Department of Health and Human Services (HHS), an AI Corps should be established, embedding specialized AI experts within each of the department’s 10 agencies. HHS is uniquely positioned for—and urgently requires—this investment in AI expertise, as it plays a pivotal role in delivering efficient healthcare to millions of Americans. HHS’s responsibilities intersect with areas where AI has already shown great promise, including managing vast healthcare datasets, accelerating drug development, and combating healthcare fraud. 

Modeled after the success of the Department of Homeland Security (DHS)’s existing AI Corps, this program would recruit top-tier professionals with advanced expertise in AI, machine learning, data science, and data engineering to drive innovation within HHS. While current HHS initiatives like the AI Council and AI Community of Practice provide valuable strategic guidance, they fall short in delivering the on-the-ground expertise necessary for meaningful AI adoption across HHS agencies. The AI Corps would fill this gap, providing the hands-on, agency-level support necessary to move beyond strategy and into the impactful implementation intended by recent federal actions related to AI. 

This memo uses the Food and Drug Administration (FDA) as a case study to demonstrate how an AI Corps member could spearhead advancements within HHS’s agencies. However, the potential benefits extend across the department. For instance, at the Centers for Disease Control and Prevention (CDC), AI Corps experts could leverage machine learning for more precise outbreak modeling, enabling faster, more targeted public health responses. At the National Institutes of Health (NIH), they could accelerate biomedical research through AI-driven analysis of large-scale genomic and proteomic data. Similarly, at the Centers for Medicare and Medicaid Services (CMS), they could improve healthcare delivery by employing advanced algorithms for patient data analytics, predicting patient outcomes, and enhancing fraud detection mechanisms.

Challenge and Opportunity

AI is poised to revolutionize not only healthcare but also the broad spectrum of services under HHS, offering unprecedented opportunities to enhance patient outcomes, streamline administrative processes, improve public health surveillance, and advance biomedical research. Realizing these benefits and defending against potential harms demands the effective implementation and support of AI tools across HHS. The federal workforce, though committed and capable, currently lacks the specialized expertise needed to fully harness AI’s potential, risking a lag in AI adoption that could impede progress.

The public sector is responding well to this opportunity since it is well positioned to attract leading experts to help leverage new technologies. However, for federal agencies, attracting technical experts has been a perennial challenge, resulting in major setbacks in government tech projects: Of government software projects that cost more than $6 million, only 13% succeed. 

Without introducing a dedicated AI Corps, existing employees—many of whom lack specialized AI expertise—would be required to implement and manage complex AI tools alongside their regular duties. This could lead to the acquisition or development of AI solutions without proper evaluation of their suitability or effectiveness for specific use cases. Additionally, without the necessary expertise to oversee and monitor these systems, agencies may struggle to ensure they are functioning correctly and ethically. As a result, there could be significant inefficiencies, missed opportunities for impactful AI applications, and an increased reliance on external consultants who may not fully understand the unique challenges and needs of each agency. This scenario not only risks undermining the effectiveness of AI initiatives but also heightens the potential for errors, biases, and misuse of AI technologies, ultimately hindering HHS’s mission and objectives.

HHS’s AI Strategy recognizes the need for AI expertise in government; however, its focus has largely been on strategic oversight rather than the operational execution needed on the ground, with the planned establishment of an AI Council and AI Community of Practice prioritizing policy and coordination. While these entities are crucial, they do not address the immediate need for hands-on expertise within individual agencies. This leaves a critical gap in the hands-on expertise required to safely implement AI solutions at the agency level. HHS covers a wide breadth of functions, from administering national health insurance programs like Medicare and Medicaid to conducting advanced biomedical research at the NIH, with each agency facing distinct challenges where AI could provide transformative benefits. However, without dedicated support, AI adoption risks becoming fragmented, underutilized, or ineffective.

For example, at the CDC, AI could significantly improve infectious disease surveillance systems, enabling more timely interventions and enhancing the CDC’s overall preparedness for public health crises, moving beyond traditional methods that often rely on slower, manual analysis. Furthermore, the Administration for Children and Families (ACF) could leverage AI to better allocate resources, improve program outcomes, and support vulnerable populations more effectively. There are great opportunities to use machine learning algorithms to accelerate data processing and discovery in fields such as cancer genomics and personalized medicine. This could help researchers identify new biomarkers, optimize clinical trial designs, and push forward breakthroughs in medical research faster and more efficiently. However, without the right expertise, these game-changing opportunities could not only remain unrealized but also introduce significant risks. The potential for biased algorithms, privacy breaches, and misinterpretation of AI outputs poses serious concerns. Agency leaders may feel pressured to adopt technologies they don’t fully understand, leading to ineffective or even harmful implementations. Embedding AI experts within HHS agencies is essential to ensure that AI solutions are deployed responsibly, maximizing benefits while mitigating potential harms.

This gap presents an opportunity for the federal government to take decisive action. By recruiting and embedding top-tier AI professionals within each agency, HHS could ensure that AI is treated not as an ancillary task but as a core component of agency operations. These experts would bring the specialized knowledge necessary to integrate AI tools safely and effectively, optimize processes, and drive innovation within each agency.

DHS’s AI Corps, launched as part of the National AI Talent Surge, provides a strong precedent for recruiting AI specialists to advance departmental capabilities. For instance, AI Corps members have played a vital role in improving disaster response by using AI to quickly assess damage and allocate resources more effectively during crises. They have also enhanced cybersecurity efforts by using AI to detect vulnerabilities in critical U.S. government systems and networks. Building on these successes, a similar effort within HHS would ensure that AI adoption moves beyond a strategic objective to a practical implementation, with dedicated experts driving innovation across the department’s diverse functions.

Case Study: The Food and Drug Administration (FDA)

The FDA stands at the forefront of the biotechnology revolution, facing the dual challenges of rapid innovation and a massive influx of complex data. Advances in gene editing, personalized medicine, and AI-driven diagnostics promise to transform healthcare, but they also present significant regulatory hurdles. The current framework, though robust, struggles to keep pace with these innovations, risking delays in the approval and implementation of groundbreaking treatments.

This situation is reminiscent of the challenges faced in the 1980s and 1990s, when advances in pharmaceutical science outstripped the FDA’s capacity to review new drugs, leading to the so-called “drug lag.” The Prescription Drug User Fee Act of 1992 was a pivotal response, streamlining the drug review process by providing the FDA with additional resources. However, the continued reliance on scaling resources may not be sustainable as the complexity and volume of data increase.

The FDA has begun to address this new challenge. For example, the Center for Biologics Evaluation and Research has established committees like the Artificial Intelligence Coordinating Committee and the Regulatory Review AI Subcommittee. However, these efforts largely involve existing staff who must balance AI responsibilities with their regular duties, limiting the potential impact. Moreover, the focus has predominantly been on regulating AI rather than leveraging it to enhance regulatory processes.

Placing an AI expert from the HHS AI Corps within the FDA could fundamentally change this dynamic. By providing dedicated, expert support, the FDA could accelerate its regulatory review processes, ensuring timely and safe access to innovative treatments. The financial implications are significant: the value of accelerated drug approvals, as demonstrated by the worth of Priority Review Vouchers (acceleration of four months = ~$100 million), indicates that effective AI adoption could unlock billions of dollars in industry value while simultaneously improving public health outcomes.

Plan of Action

To address the challenges and seize the opportunities outlined earlier, the Office of the Chief Artificial Intelligence Officer (OCAIO) within HHS should establish an AI Corps composed of specialized experts in artificial intelligence, machine learning, data science, and data engineering. This initiative will be modeled after DHS’s successful AI Corps and tailored to the unique needs of HHS and its 10 agencies.

Recommendation 1. Establish an AI Corps within HHS.

Composition: The AI Corps would initially consist of 10 experts hired to temporary civil servant positions, with one member allocated to each of HHS’s 10 agencies, and each placement lasting one to two years. These experts will possess a range of technical skills—including AI, data science, data engineering, and cloud computing—tailored to each agency’s specific needs and technological maturity. This approach ensures that each agency has the appropriate expertise to effectively implement AI tools and methodologies, whether that involves building foundational data infrastructure or developing advanced AI applications.

Hiring authority: The DHS AI Corps utilized direct hiring authority, which was expanded by the Office of Personnel Management under the National AI Talent Surge. HHS’s AI Corps could adopt a similar approach. This authority would enable streamlined recruitment of individuals into specific AI roles, including positions in AI research, machine learning, and data science. This expedited process would allow HHS to quickly hire and onboard top-tier AI talent.

Oversight: The AI Corps would be overseen by the OCAIO, which would provide strategic direction and ensure alignment with HHS’s broader AI initiatives. The OCAIO would also be responsible for coordinating the activities of the AI Corps, setting performance goals, and evaluating outcomes.

Budget and Funding

Estimated cost: The AI Corps is projected to cost approximately $1.5 million per year, based on an average salary of $150,000 per corps member. This estimate includes salaries and operational costs such as training, travel for interagency collaboration, and participation in conferences. 

Funding source: Funding would be sourced from the existing HHS budget, specifically from allocations set aside for digital transformation and innovation. Given the relatively modest budget required, reallocation within these existing funds should be sufficient. 

Recruitment and Training

Selection process: AI Corps members would be recruited through a competitive process, targeting individuals with proven expertise in AI, data science, and related fields. 

Training: Upon selection, AI Corps members would undergo an intensive orientation and training program to familiarize them with the specific needs and challenges of HHS’s various agencies. This also includes training on federal regulations, ethics, and data governance to ensure that AI applications comply with existing laws and policies.

Agency Integration

Deployment: Each AI Corps member would be embedded within a specific HHS agency, where they would work closely with agency leadership and staff to identify opportunities for AI implementation. Their primary responsibility would be to develop and deploy AI tools that enhance the agency’s mission-critical processes. For example, an AI Corps member embedded at the CDC could focus on improving disease surveillance systems through AI-driven predictive analytics, while a member at the NIH could drive advancements in biomedical research by using machine learning algorithms to analyze complex genomic data.

Collaboration: To ensure cross-agency learning and collaboration, AI Corps members would convene regularly to share insights, challenges, and successes. These convenings would be aligned with the existing AI Community of Practice meetings, fostering a broader exchange of knowledge and best practices across the department.

Case Study: The FDA

AI Corps Integration at the FDA

Location: The AI Corps member assigned to the FDA would be based in the Office of Digital Transformation, reporting directly to the chief information officer. This strategic placement would enable the expert to work closely with the FDA’s leadership team, ensuring that AI initiatives are aligned with the agency’s overall digital strategy.

Key responsibilities

Process improvement: The AI Corps member would collaborate with FDA reviewers to identify opportunities for AI to streamline regulatory review processes. This might include developing AI tools to assist with data analysis, automate routine tasks, or enhance decision-making capabilities.

Opportunity scoping: The expert would engage with FDA staff to understand their workflows, challenges, and data needs. Based on these insights, the AI Corps member would scope and propose AI solutions tailored to the FDA’s specific requirements.

Pilot projects: The AI Corps member would lead pilot projects to test AI tools in real-world scenarios, gathering data and feedback to refine and scale successful initiatives across the agency.

Conclusion

Establishing an AI Corps within HHS is a critical step toward harnessing AI’s full potential to enhance outcomes and operational efficiency across federal health agencies. By embedding dedicated AI experts within each agency, HHS can accelerate the adoption of innovative AI solutions, address current implementation gaps, and proactively respond to the evolving demands of the health landscape.

While HHS may currently have less technological infrastructure compared to departments like the Department of Homeland Security, targeted investment in in-house expertise is key to bridging that gap. The proposed AI Corps not only empowers agencies like the FDA, CDC, NIH, and CMS to enhance their missions but also sets a precedent for effective AI integration across the federal government. Prompt action to establish the AI Corps will position HHS at the forefront of technological innovation, delivering tangible benefits to the American public and transforming the way it delivers services and fulfills its mission.

This action-ready policy memo is part of Day One 2025 — our effort to bring forward bold policy ideas, grounded in science and evidence, that can tackle the country’s biggest challenges and bring us closer to the prosperous, equitable and safe future that we all hope for whoever takes office in 2025 and beyond.

PLEASE NOTE (February 2025): Since publication several government websites have been taken offline. We apologize for any broken links to once accessible public data.

Frequently Asked Questions
How will the AI Corps avoid becoming just another bureaucratic layer?

The AI Corps is designed to be the opposite of bureaucracy—it’s about action, not administration. These experts will be embedded directly within agencies, working alongside existing teams to solve real-world problems, not adding paperwork. Their mission is to integrate AI into daily operations, making processes more efficient and outcomes more impactful. By focusing on tangible results and measurable improvements, the AI Corps will be judged by its ability to cut through red tape, not create it.

What if AI Corps members are too ahead of the curve for existing agency cultures?

Innovation can present challenges, but the AI Corps is designed to address them effectively. These experts will not only bring technical expertise but also serve as facilitators who can translate advanced AI capabilities into practical applications that align with existing agency cultures. A key part of their role will be to make AI more accessible and understandable, ensuring it is valuable to all levels of staff, from frontline workers to senior leadership. Their success will depend on their ability to seamlessly integrate advanced technology into the agency’s everyday operations.

Why focus on AI when there are so many other pressing health issues?

AI isn’t just another tool; it’s a force multiplier that can help solve those other pressing issues more effectively. Whether it’s accelerating drug approvals at the FDA or enhancing public health responses across HHS, AI has the potential to improve outcomes, save time, and reduce costs. By embedding AI experts within agencies, we’re not just addressing one problem—we’re empowering the entire department to tackle multiple challenges with greater efficiency and impact.

What’s in it for the AI experts? Why would top talent join the AI Corps?

For top AI talent, the AI Corps offers a unique opportunity to make a difference at a scale that few private-sector roles can match. It’s a chance to apply their skills to public service, tackling some of the nation’s most critical challenges in healthcare, regulation, and beyond. The AI Corps members will have the opportunity to shape the future of AI in government, leaving a legacy of innovation and impact. The allure of making a tangible difference in people’s lives can be a powerful motivator for the right kind of talent.

Why not outsource AI talent or rely on consultants instead of building in-house expertise?

While outsourcing AI talent or using consultants can offer short-term benefits, it often lacks the sustained engagement necessary for long-term success. Building in-house expertise through the AI Corps ensures that AI capabilities are deeply integrated into the agency’s operations and culture. A notable example illustrating the risks of overreliance on external contractors is the initial rollout of HealthCare.gov. The website faced significant technical issues at launch due to coordination challenges and insufficient in-house technical oversight, which hindered public access to essential healthcare services. In contrast, recent successful government initiatives—such as the efficient distribution of COVID-19 test kits and the timely processing of economic stimulus payments directly into bank accounts—demonstrate the positive impact of having the right technical experts within government agencies.

How will the AI Corps collaborate with existing IT and data teams within agencies?

Collaboration is crucial to the AI Corps’ success. Instead of working in isolation, AI Corps members will integrate with existing IT and data teams, bringing specialized AI knowledge that complements the teams’ expertise. This partnership approach ensures that AI initiatives are well-grounded in the agencies’ existing infrastructure and aligned with ongoing IT projects. The AI Corps will serve as a catalyst, amplifying the capabilities of existing teams rather than duplicating their efforts.

Could the AI Corps inadvertently lead to job displacement within agencies?

The AI Corps is focused on augmentation, not replacement. The primary goal is to empower existing staff with advanced tools and processes, enhancing their work rather than replacing them. AI Corps members will collaborate closely with agency employees to automate routine tasks and free up time for more meaningful activities. A 2021 study by the Boston Consulting Group found that 60% of employees view AI as a coworker rather than a replacement. This reflects the intent of the AI Corps—to build capacity within agencies and ensure that AI is a tool that amplifies human effort, fostering a more efficient and effective workforce.

What does success look like for the HHS AI Corps program after one or two years?

Success for the AI Corps program means that each HHS agency has made measurable progress toward integrating AI and related technologies, tailored to their specific needs and maturity levels. Within one to two years, agencies might have established robust data infrastructures, migrated platforms to the cloud, or developed pilot AI projects that address key challenges. Success also includes fostering a culture of innovation and experimentation, with AI Corps members identifying opportunities and creating proofs of concept in low-risk environments. By collaborating across agencies, these experts support each other and amplify the program’s impact. Ultimately, success is reflected in enhanced capabilities and efficiencies within agencies, setting a strong foundation for ongoing technological advancement aligned with each agency’s mission.