Politics

Anthropic refuses Pentagon’s AI military terms amid escalating dispute over surveillance and autonomous weapons

Saturday, 28 February 2026 5:36AM UTC

Anthropic chief Dario Amodei has rejected the US Department of Defense's broadening military use of its Claude AI model, sparking a high-stakes confrontation that threatens to reshape industry standards and government relations over AI safety and national security.

Anthropic’s chief executive, Dario Amodei, said on Thursday that the company “cannot in good conscience accede” to terms offered by the United States Department of Defense that would permit broader military use of its Claude model, widening a public confrontation that could see the firm lose federal business and face further government actions. According to Fortune, Anthropic described the Pentagon’s revised contract language as making “virtually no progress on preventing Claude’s use for mass surveillance of Americans or in fully autonomous weapons.”

Sources: Associated Press, Fortune.

The Defence Department insists it is not seeking tools for domestic mass surveillance and has said it will not deploy autonomous weapons without human oversight. Pentagon spokesman Sean Parnell reiterated on social media that the military “has no interest in using AI to conduct mass surveillance of Americans (which is illegal) nor do we want to use AI to develop autonomous weapons that operate without human involvement.” Still, senior defence officials have framed the negotiations as a matter of operational flexibility and urgency.

Sources: Fortune, Associated Press.

Tensions escalated after Defence Secretary Pete Hegseth gave Anthropic an ultimatum to accept the department’s terms by Friday or face contract termination; officials also warned of more severe steps such as declaring the company a supply chain risk or invoking the Defense Production Act. Amodei pushed back, arguing those threats were inconsistent with each other, labelled a security risk on one hand, deemed essential to national security on the other, and said Anthropic would prepare to facilitate a smooth transition to another provider if an agreement could not be reached.

Sources: Fortune, Axios.

The fallout has rapidly expanded beyond the immediate parties. The Associated Press reports that the administration ordered federal agencies to stop using Anthropic technology following the company’s refusal to remove safeguards, a move that drew criticism from AI safety advocates, some technologists and a number of lawmakers who warned the dispute risks politicising sensitive technology. Anthropic has signalled it may challenge government actions in court, calling any blacklisting legally unjustified.

Sources: Associated Press, Axios.

Other AI firms have responded in ways that underline the strategic stakes. Axios reports OpenAI has struck terms with the Pentagon that mirror the red lines Anthropic defended, prohibitions on domestic mass surveillance and requirements for human accountability in force decisions, and CEO Sam Altman has said his company will adopt similar ethical limits while continuing to work with defence customers. That parallel agreement complicates the Pentagon’s rationale for pressuring Anthropic and highlights an industry-wide debate about how to balance safety commitments with national security needs.

Sources: Axios, Axios.

The political temperature has become acute. President Trump and some Pentagon officials have publicly criticised Anthropic, with Axios reporting the company was labelled a “supply chain risk” and that the government cancelled a reported $200 million contract. Senate defence leaders have privately sought to mediate, according to Axios, signalling Congress may intervene to reconcile governance concerns with military operational requirements. Lawmakers on both sides of the aisle expressed unease about the public nature of the standoff and urged quieter, more constructive negotiations.

Sources: Axios, Axios.

Industry observers warn that the dispute could set a precedent shaping corporate behaviour and investor responses across the sector. Axios coverage highlights fears that blacklisting could pressure major technology investors and partners and prompt wider divestment, while former administration advisers described the move as potentially destructive to a key U.S. company. At the same time, calls from senators for binding AI governance frameworks for national security contexts reflect a growing consensus that policy solutions are needed to prevent similar crises in future.

Sources: Axios, Associated Press.

Source Reference Map

Inspired by headline at: ^[1]

Sources by paragraph:

Paragraph 1: ^[2], ^[1]
Paragraph 2: ^[1], ^[2]
Paragraph 3: ^[1], ^[3]
Paragraph 4: ^[2], ^[3]
Paragraph 5: ^[4], ^[5]
Paragraph 6: ^[3], ^[7]
Paragraph 7: ^[6], ^[2]

Source: Noah Wire Services

More on this

https://fortune.com/2026/02/27/dario-amodei-says-he-cannot-in-good-conscience-bow-to-pentagons-demands-over-ai-use-in-military/ - Please view link - unable to able to access data
https://apnews.com/article/b72d1894bc842d9acf026df3867bee8a - The Associated Press reports that the Trump administration has ordered all U.S. federal agencies to halt the use of Anthropic's AI technology. This action follows Anthropic's refusal to provide the military with unrestricted access to its chatbot, Claude. President Trump and Defense Secretary Pete Hegseth have labeled the company a national security risk, equating Anthropic’s restrictions to support for adversarial interests. The conflict centers on Anthropic’s refusal to allow its AI to be used for mass domestic surveillance or fully autonomous weapons, citing ethical safeguards. In response, Anthropic plans to challenge the government's actions in court, calling the designation legally unjustified. The move has drawn criticism from AI safety advocates, tech leaders, and some lawmakers, who suggest political motivations may be driving the decision. Meanwhile, rival OpenAI has swiftly secured a deal with the Department of Defense that respects the same AI usage limitations Anthropic defended. This high-profile dispute underscores rising tensions between AI firms and the government over the ethical use of advanced technologies.
https://www.axios.com/2026/02/27/anthropic-pentagon-supply-chain-risk-claude - Axios reports that President Trump has announced that the U.S. government has blacklisted the AI company Anthropic after the Pentagon labeled it a 'supply chain risk.' This decision follows Anthropic's refusal to remove safeguards limiting military use of its AI model, Claude, due to ethical concerns about mass surveillance and autonomous weapons. The Pentagon, led by Defense Secretary Pete Hegseth, reacted by canceling its $200 million contract with Anthropic and barring other defense partners from using Claude. Trump criticized the company for attempting to enforce its Terms of Service over national interests. Anthropic CEO Dario Amodei had rejected what the Pentagon described as its 'best and final offer.' The Pentagon argues that its demands are necessary for operational flexibility. Despite sharing similar AI safety standards, the Pentagon has turned to OpenAI and recently inked a deal with Elon Musk’s xAI to deploy its Grok model. Anthropic may challenge the blacklisting in court. This marks a pivotal confrontation between AI ethics and military control over advanced technology.
https://www.axios.com/2026/02/27/altman-openai-anthropic-pentagon - Axios reports that OpenAI CEO Sam Altman has stated that the company will adopt the same ethical 'red lines' as Anthropic in its dealings with the Pentagon—specifically, refusing to develop AI for mass surveillance or autonomous lethal weapons. This move aligns OpenAI with a growing tech industry stance and could complicate the Pentagon’s search for an AI partner following conflict with Anthropic over similar issues. While Altman emphasized his willingness to work with the military in sensitive contexts, he stressed the importance of upholding safety and ethical standards. ChatGPT is currently used in unclassified military systems, and discussions are underway to expand its role, although concerns remain about private companies wielding too much control in national security. The issue has sparked broader industry debate, with employees from OpenAI and Google signing a letter encouraging resistance to Pentagon pressure. Despite Defense Department assurances they will not engage in mass surveillance or rapid deployment of lethal AI, the debate continues over the ethical boundaries of military-AI collaboration.
https://www.axios.com/2026/02/27/pentagon-openai-safety-red-lines-anthropic - Axios reports that OpenAI has reached an agreement with the Pentagon to deploy its AI models under strict safety conditions. CEO Sam Altman announced the partnership, highlighting that the Defense Department agreed to key safety 'red lines' that align with those of rival AI firm Anthropic. These include a ban on domestic mass surveillance and maintaining human accountability in the use of force, especially in autonomous weapon systems. Altman emphasized the importance of compliance with legal standards, noting that mass surveillance is illegal. The agreement reflects OpenAI's commitment to ethical AI use, even under pressure from political figures. The decision also contrasts with the Pentagon's recent criticism of Anthropic, which it labeled as driven by 'woke' philosophy. OpenAI’s stance may help it maintain favorable political standing while promoting ethical standards across the AI industry.
https://www.axios.com/2026/02/28/trump-anthropic-supply-chain-risk-reaction - Axios reports that a former Trump administration AI official has sharply criticized the Pentagon's recent blacklisting of the AI company Anthropic, characterizing it as 'attempted corporate murder.' The controversy centers on Anthropic’s refusal to meet Pentagon demands to remove all safeguards from its AI model, Claude, out of ethical concerns about the use of AI in surveillance and autonomous weapons. The Biden administration views Anthropic as a 'woke' company, and its exclusion could force major investors—Amazon, Google, and Nvidia—to divest, negatively impacting both the company and broader tech sector. Dean Ball, the former AI adviser, expressed alarm over this move on social media. Tech industry stakeholders worry that this political pushback could mark the beginning of a more hostile regulatory environment for AI and tech companies, diverging sharply from Wall Street's preference for innovation-friendly, low-regulation policies.
https://www.axios.com/2026/02/27/senate-defense-anthropic-pentagon-ai - Axios reports that top leaders in the U.S. Senate's defense committees are stepping in to mediate a growing dispute between the Pentagon and AI company Anthropic over the use limitations of Anthropic's AI model, Claude, in classified defense settings. The Senate Armed Services Committee Chair Roger Wicker and Ranking Member Jack Reed, along with Defense Appropriations leaders Mitch McConnell and Chris Coons, have sent a private letter to both parties urging a resolution. The conflict arises from Anthropic's restrictions on its AI use—specifically prohibiting applications like mass surveillance and autonomous weapons—conditions that clash with Pentagon objectives. The Pentagon has given Anthropic a deadline of 5:01 pm EST on Friday to comply with its demands or risk losing their defense contract. The increasing pressure has prompted Congress to potentially play a more active role in reconciling AI ethics with national security needs.

Noah Fact Check Pro

The draft above was created using the information available at the time the story first emerged. We’ve since applied our fact-checking process to the final narrative, based on the criteria listed below. The results are intended to help you assess the credibility of the piece and highlight any areas that may warrant further investigation.

Freshness check

Score: 10

Notes: The article is current, published on February 27, 2026, and reports on recent events, including the Pentagon's ultimatum to Anthropic and CEO Dario Amodei's response. No evidence of recycled or outdated content was found.

Quotes check

Score: 8

Notes: Direct quotes from Dario Amodei and Pentagon spokesperson Sean Parnell are used. These quotes are consistent with statements from the original sources. However, the exact earliest usage of these quotes could not be independently verified, raising a slight concern about their originality.

Source reliability

Score: 9

Notes: The article is published by Fortune, a reputable news organisation known for its business and technology reporting. The sources cited, including the Associated Press and Axios, are also reputable. However, the article relies heavily on these secondary sources, which may introduce potential biases or inaccuracies.

Plausibility check

Score: 9

Notes: The events described align with recent developments in AI and military policy. The claims are plausible and supported by multiple reputable sources. However, the heavy reliance on secondary sources without direct access to primary statements or documents slightly reduces the confidence in the absolute accuracy of the details.

Overall assessment

Verdict (FAIL, OPEN, PASS): PASS

Confidence (LOW, MEDIUM, HIGH): MEDIUM

Summary: The article provides a timely and plausible account of the dispute between Anthropic and the Pentagon over AI usage. While the content is accessible and the sources cited are reputable, the heavy reliance on secondary sources without direct access to primary statements or documents introduces some uncertainty. Additionally, the exact earliest usage of the quotes could not be independently verified, raising a slight concern about their originality. These factors contribute to a medium level of confidence in the article's accuracy.

AI safety
Military technology
US defence policy