There is practically a universal consensus that the national security classification system has become dysfunctional and counterproductive. (Just what to do about it remains up in the air–more on that shortly).
That consensus was articulated again earlier this month in a speech by Joan Dempsey, formerly a senior Pentagon intelligence official, a Deputy Director of Central Intelligence, and executive director of the President’s Foreign Intelligence Advisory Board, and now a vice president at Booz Allen and Hamilton.
“Ninety-five percent of what we do shouldn’t be classified at all, or it should be a much lower level of classification,” Ms. Dempsey said. “We’re lazy about classification. We call things secret that are not secret. It hampers our ability to be effective as a community. It costs the country billions of unnecessary dollars, and it doesn’t provide us one additional capability. We’re our own worst enemy in that regard,” she said.
Ms. Dempsey spoke on April 14 at the University of Texas at Austin. Her talk, ironically enough, was entitled “Back to Black: An Argument for Removing U.S. Intelligence Activities from Public Scrutiny,” and amounted to a call for increased secrecy of intelligence operations. But her defense of intelligence secrecy, she said, was contingent on robust congressional oversight and was not intended to shield misconduct or to perpetuate overclassification. A webcast of the talk is available here (the discussion of classification begins at about 28:45).
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.
When properly structured — with specific numeric targets, secured financial obligations, independent monitoring, and meaningful enforcement — CBAs transform data center deals into durable community partnerships.