It is important to understand that there is no rigorous, consensual definition of what constitutes classified information. Instead, in a practical sense, classified information is whatever the executive branch says it is.
(A minority of classified information, such as nuclear weapons design information, is specified and protected by statute. The remainder, the large majority, is classified by executive order.)
In 1997, the Central Intelligence Agency declassified the total intelligence budget for that year ($26.6 billion). But intelligence budget figures from three, four and five decades earlier remain classified. Why? Because the CIA says so!
One might argue that it should be the other way around — budget figures from the remote past should be declassified while more recent figures should perhaps be classified. But such logic is foreign to CIA classification policy, and to the classification system as a whole.
By far the most sensitive government document Secrecy News has obtained in recent years is a January 2006 military manual that explains in nearly 200 pages of detail exactly how to use a particular type of weapon that is known to pose a significant terrorist threat.
If there is anything that should be classified in the interests of national security, this manual would seem to be it. Yet it is unclassified. Distribution is “unlimited.”
The conclusion that emerges from the chaos of government information policy is that the classification system is essentially an administrative tool used by the executive branch for its own internal purposes. It is a poor index of what is sensitive and what is not.
Americans are paying too much for almost everything, because the United States has long treated its trucking industry as an artifact to be preserved rather than as an opportunity for innovation.
These ideas aim to advance the detailed policy solutions needed to foster public trust and implement fairness in the adoption of AI across diverse domains, from healthcare and government benefits to rural access, education, and worker protections.
The evidence is clear: algorithmic pay-setting is established in app-based work, and payroll/timekeeping failures show how software can produce systemic wage harm at scale
While a few states have taken steps to implement decision-making mechanisms for certain AI systems, too many leaders are simply accepting narratives about AI’s purported public benefit at face value – jumping to the “how” of AI implementation before thoroughly vetting potential systems and deciding whether they are appropriate to use at all.