Executive branch agencies often classify information inconsistently or unnecessarily. But when challenged, they will sometimes modify their practices.
These elementary but important facts were illustrated recently by Justice Department Inspector General Glenn A. Fine, who described the process by which his office’s investigative reports are reviewed by agency officials prior to release.
“We have seen a lot of times where [agency officials] said, well, that is too sensitive, that is classified, and then we will look and in another forum they have publicly released it,” Mr. Fine said at a recent congressional hearing (pdf, at pp. 28-29).
“I have been in situations where the FBI told me you can’t say that, that is classified. I said okay. And then a week later, an FBI employee will come to [to Congress] and say the same thing. And I say to myself why can’t we say it if they can?”
“A lot of times FBI will give a report to one person and they will say this is classified, and then the FBI will give it to another person and different amounts are classified,” Mr. Fine said. “It is not a precise science.”
“I have to say, this is not just restricted to the Department of Justice,” concurred Rep. William D. Delahunt, “it is throughout the executive branch…. I have seen what you just articulated happen time and time again.”
“It is my own belief that the classification process has become a tool, if you will, for the avoidance of embarrassment,” the Congressman said.
The good news is that, at least at the Justice Department, the Inspector General does not simply yield to the whims of the classifiers, but insists on a sensible rationale for secrecy.
“We push for explanations to make sure that it [classification] is not being used as a way to avoid embarrassment,” Mr. Fine said. “And I think generally, in our case, we have been pretty successful.”
The relative success of Mr. Fine’s office in limiting classification of Inspector General reports points to an untapped opportunity for curtailing overclassification more broadly. That is, agency inspectors general could be assigned to conduct regular audits of classification activity and to “push for explanations” of dubious secrets.
The inspectors general are already on site throughout most of the executive branch and they have the security clearances they need to function (though in some cases they may lack adequate funding for this and other purposes). They could provide an enormous force multiplier for the small staff of the Information Security Oversight Office which is nominally responsible for classification oversight government-wide. If the classification system is worth fixing, the Inspectors General could have an important role to play.
Mr. Fine’s remarks on classification appeared in a newly published hearing record entitled “City on the Hill or Prison on the Bay? Part III: Guantanamo–The Role of the FBI,” House Foreign Affairs Committee, June 4, 2008.
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.