Latest / Government and geopolitics
OMB memo M-26-04 makes 'unbiased AI' terms mandatory in federal LLM contracts
Implementing the July 2025 'Preventing Woke AI' order, M-26-04 requires agencies to write 'truth-seeking' and 'ideological neutrality' requirements into new LLM contracts and to update procurement policies by 11 March 2026. Vendors must at minimum supply acceptable-use policies, model or system cards and a feedback channel, and agencies may demand more, such as system-prompt, red-teaming and bias-evaluation details.
Why it matters
Any lab selling LLMs to US agencies now faces disclosure duties and contractual neutrality tests that can be used to exclude a vendor.
Line of Thought
Follow this story
Pick any item to keep going. Your path builds up above as a line you can share.
Curated lines through this story
Directly linked
Connections our researchers recorded
- DevelopmentWhite House publishes America's AI Action Plan and three executive orders23 Jul 2025 · Policy · USPuts into practice: Puts into effect the 'Preventing Woke AI' EO signed alongside the plan
What led here
Earlier developments on the same thread
- DevelopmentNY Comptroller finds New York City barely enforces its AI hiring law2 Dec 2025 · Enforcement or ruling · US-NY
- DevelopmentOpenAI, Anthropic and Google offer frontier AI to US agencies for about $1 via GSA6 Aug 2025 · Procurement · US
- DevelopmentPentagon AI office awards Anthropic, Google, OpenAI and xAI up to $200m each14 Jul 2025 · Procurement · US
- DevelopmentOMB replaces Biden-era rules with new memos on federal AI use and AI buying3 Apr 2025 · Policy · US
What happened next
Later developments on the same thread
- DevelopmentHegseth orders an 'AI-first' military and 'any lawful use' terms for AI contracts12 Jan 2026 · Policy · US
- DevelopmentPentagon labels Anthropic a supply-chain risk after it refuses to drop two limits27 Feb 2026 · Enforcement or ruling · US
- DevelopmentOpenAI agrees to deploy its models on Department of War classified networks28 Feb 2026 · Deal · US
- DevelopmentAppeals court upholds Pentagon's supply-chain-risk label on Anthropic25 Sep 2026 · Enforcement or ruling · US
Same story elsewhere
What other countries and bodies did on this
- DevelopmentRBI drafts model risk rules covering AI/ML for all regulated lenders24 Jun 2026 · Rule change · IN
- DevelopmentRBI publishes FREE-AI framework for AI in India's financial sector13 Aug 2025 · Report · IN
- DevelopmentILO refined index: one in four jobs worldwide exposed to generative AI20 May 2025 · Research · INTL
Rules in play
Laws and guidance this touches
- RuleOMB M-26-04 ('Woke AI' procurement rules)US · In force · 11 Dec 2025
- RuleDirective on Automated Decision-MakingCA · In force · 1 Apr 2019
- RuleCouncil of Europe AI ConventionINTL · Enacted, not yet in force · 17 May 2024
- RuleColorado AI Act (SB 24-205)US-CO · Superseded · 17 May 2024
- RuleColorado DOI Reg. 10-1-1US-CO · In force · 20 Aug 2025
Threads by topic: Government buying Bias Frontier models