A federal judge this week granted permission to J. William Leonard, the former director of the Information Security Oversight Office, to discuss three documents that were at issue in the trial of former National Security Agency official Thomas Drake.
Mr. Leonard, an expert witness for the Drake defense, had sought permission to publicly challenge the legitimacy of the classification of one of the documents cited in the indictment against Mr. Drake, which was ultimately dismissed.
The government had opposed the motion to lift the non-disclosure obligations in the protective order that bound Mr. Leonard. Government attorneys argued that Mr. Leonard had no standing to make such a request, which was filed by Mr. Drake’s public defenders James Wyda and Deborah L. Boardman. The government also said the request should be denied in order “to prevent a flood of similar claims by non-parties in other completed cases.” Instead, prosecutors suggested, Mr. Leonard could file a Freedom of Information Act request for the records in question.
But Judge Richard D. Bennett said that “the government’s arguments in this case are inapposite.” Even if the documents were made available to Mr. Leonard under FOIA, “he would not have been permitted to discuss them as he would remain bound by this Court’s Protective Order.”
Judge Bennett therefore formally lifted the Protective Order and granted Mr. Leonard permission to publicly discuss his concerns.
The documents themselves, and the complaint that Mr. Leonard submitted to the Information Security Oversight Office, were released by the National Security Agency under FOIA in July. (“Defense, Critique of NSA Classification Action Released,” Secrecy News, July 30.)
The complaint itself is still pending, and is awaiting a formal response from the Department of Justice, said the current ISOO director, John P. Fitzpatrick.
The challenge presented by Mr. Leonard extends well beyond the Drake case or the secrecy practices of the National Security Agency. Essentially, the question posed by the former ISOO director’s complaint is whether there is any threshold beyond which classification of information is so completely unjustified as to trigger third-party intervention to correct the problem. As of today, such corrective mechanisms are weak or nonexistent.
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.