Trump administration drafts strict AI guidelines as Pentagon designates Anthropic supply-chain risk

US draws up strict new AI guidelines amid Anthropic clash, FT reports

March 6 (Reuters) - The Trump administration has drawn up strict rules for civilian artificial intelligence contracts that would require AI companies to allow "any lawful" use of their ⁠models amid a stand-off between the Pentagon and Anthropic, the Financial Times reported on Friday. A draft of the new guidelines reviewed by the FT says AI ⁠groups seeking to do business with the government must grant the U.S. ⁠an irrevocable license to use their systems for all legal ⁠purposes, the report added. Reuters could not immediately ⁠verify the report. Reporting by Bipasha Dey in Bengaluru; Editing by Himani Sarkar Our Standards: The Thomson Reuters Trust Principles., opens new tab

US draws up strict new AI guidelines amid Anthropic clash: Report

The Trump administration is proposing strict rules for civilian AI contracts, requiring companies to permit "any lawful" use of their models. This comes after the Pentagon designated Anthropic a "supply-chain risk," barring its technology from military work due to disputes over safeguards. The proposed guidelines also mandate AI systems avoid partisan judgments and disclose foreign compliance modifications. The Trump administration has drawn up strict rules for civilian artificial intelligence contracts that would require AI companies to allow "any lawful" use of their models amid a stand-off between the Pentagon and Anthropic, the Financial Times reported on ⁠Friday. The ⁠report comes a day after the Pentagon formally designated Anthropic a "supply-chain risk" and barred government contractors from using the AI firm's technology in work for the U.S. military. That move followed a months-long dispute over the company's insistence on safeguards that the Defense Department says went too ⁠far. A draft of the guidelines reviewed by the FT says AI groups seeking business with the government ⁠must grant the U.S. an irrevocable license to use their systems for all legal purposes. The guidance from the U.S. General Services Administration (GSA) would apply to civilian contracts and is part of a broader government-wide effort to strengthen AI services procurement, the FT reported, adding that it mirrors measures the Pentagon is considering for military contracts. The White House and the GSA did not immediately respond to requests for comment. The draft ⁠from the GSA also mandates that contractors "must not intentionally encode partisan or ideological judgments into the AI systems data outputs," the FT reported. It also requires companies to disclose whether their models have been "modified or configured to comply with any non-U.S. federal government or commercial compliance or regulatory framework," the FT reported.

White House eyes 'any lawful use' mandate for AI firms in new draft By Investing.com

Investing.com -- The Trump administration has drafted aggressive new guidelines for civilian artificial intelligence contracts, requiring companies to permit "any lawful" use of their models by the U.S. government, according to the Financial Times. The move follows a high-stakes standoff between the Department of War and Anthropic, which culminated Thursday in the Pentagon designating the AI firm a "supply-chain risk." Market sentiment in the tech sector remained cautious following the report. Shares of AI-adjacent firms wavered as investors assessed the potential for a broader regulatory crackdown on "ethical" guardrails that conflict with national security objectives. The Nasdaq 100 had slipped 1.51% by 15:59 ET (20:59 GMT), while Microsoft Corporation (NASDAQ:MSFT) fell 0.42% and Alphabet Inc Class A (NASDAQ:GOOGL) was down 0.78% as the industry weighed the implications of the U.S. General Services Administration (GSA's) "irrevocable license" requirement. Anthropic designated a 'supply-chain risk' The new General Services Administration (GSA) rules portray a hardening stance within the military establishment. The Pentagon's designation of Anthropic as a "supply chain risk", a label typically reserved for foreign entities like Huawei, effectively bars any government contractor from utilizing the firm's technology. The "blacklisting" stems from a months-long dispute over Anthropic's refusal to waive safeguards against mass domestic surveillance and the use of its Claude models in lethal autonomous weapons. Secretary of War Pete Hegseth defended the move on Friday, stating that the U.S. requires "patriotic" technology partners who do not impose restrictive "red lines" on lawful operations. Anthropic has argued that the designation is legally unsound and plans to challenge the decision in court. The Trump administration has granted a six-month transition period for agencies to migrate away from Anthropic's systems. Ideological neutrality and disclosure mandates The newly published GSA draft mandates that AI contractors must not "intentionally encode partisan or ideological judgments" into their data outputs. The rule aims to strip what the administration describes as "embedded bias" or "wokeness" from government-funded models. Companies must now disclose if their models have been modified to comply with non-U.S. regulatory frameworks, such as the European Union's AI Act. Analysts at Evercore ISI noted that these requirements could force a "decoupling" of the American AI stack from global standards. Traders are now watching closely for a response from OpenAI, which reportedly stepped in to fill the Pentagon's vacuum shortly after Anthropic's exile.

US draws up strict new AI guidelines amid Anthropic clash, FT reports

March 6 (Reuters) - The Trump administration has drawn up strict rules for civilian artificial intelligence contracts that would require AI companies to allow "any lawful" use of their models amid a stand-off between the Pentagon and Anthropic, the Financial Times reported on Friday. A draft of the new guidelines reviewed by the FT says AI groups seeking to do business with the government must grant the U.S. an irrevocable license to use their systems for all legal purposes, the report added. Reuters could not immediately verify the report. (Reporting by Bipasha Dey in Bengaluru; Editing by Himani Sarkar)