Electronic Frontier Foundation (EFF) — analysis: Published analysis criticizing OpenAI's Pentagon deal for inadequate safeguards against AI-powered surveillance

NoteThe source directly confirms the claim. The EFF published analysis criticizing OpenAI's Pentagon deal for inadequate safeguards against AI-powered surveillance as of March 6, 2026 (matching the claimed date of 2026-03). The article's entire thesis is that the contractual language fails to prevent surveillance due to ambiguous wording and government interpretation flexibility. The subject (EFF's analysis) matches the claim perfectly.

NoteThe source text directly confirms the claim. It is an EFF-published analysis (Deeplinks Blog post dated March 6, 2026) that explicitly criticizes OpenAI's Pentagon deal for inadequate safeguards against AI-powered surveillance. The article's title and content demonstrate that EFF found the deal's protections insufficient, particularly regarding the vague language around 'applicable laws' and 'intentionally' that could allow surveillance to continue. The date matches the claimed timeframe (2026-03).