U.S. Bans Anthropic, OpenAI Wins Pentagon Deal
The U.S. government has designated AI firm Anthropic a “supply chain risk,” ordering all federal agencies to stop using its technology amid a clash over AI safety. Hours later, competitor OpenAI announced it had secured a landmark contract to provide AI models to the Pentagon. The move effectively sidelines a major rival and solidifies OpenAI's position as the U.S. government's dominant AI partner.
The dispute escalated after months of private negotiations failed to produce an agreement. Anthropic CEO Dario Amodei stated the company could not "in good conscience" allow its AI model, Claude, to be used for mass domestic surveillance or in fully autonomous weapons systems, calling these "red lines." The Pentagon reportedly insisted on contract language that would permit "all lawful uses" of the technology, which Anthropic believed contained legalese that would nullify its safety safeguards. In an unprecedented move against a U.S. tech firm, Defense Secretary Pete Hegseth designated Anthropic a "supply chain risk," a label historically applied to foreign adversaries like China's Huawei. This designation prohibits military contractors from using Anthropic's technology. President Trump amplified the decision on Truth Social, accusing the company of being "Leftwing nut jobs" trying to "strong-arm the Pentagon." The OpenAI contract, reportedly valued at up to $200 million, allows for the deployment of its models on the Pentagon's classified networks. While the deal includes the "all lawful purposes" language that Anthropic rejected, OpenAI CEO Sam Altman announced that the agreement also explicitly prohibits the use of their technology for mass surveillance and autonomous weapons. Despite securing the lucrative contract, OpenAI publicly defended its rival. The company stated it did not believe Anthropic should be designated a supply chain risk and had communicated this to the government. Sam Altman called the government's action against Anthropic an "extremely scary precedent" and said OpenAI moved quickly to "de-escalate the situation." The administration's actions triggered a divided response in Silicon Valley. Hundreds of employees from Google and OpenAI signed an open letter supporting Anthropic's stance. Conversely, tech figures like Elon Musk sided with the administration, with Musk posting on X that "Anthropic hates Western Civilization." The move also drew criticism from some lawmakers. Senator Mark Warner, a Democrat from Virginia, expressed "serious concerns about whether national security decisions are being driven by careful analysis or political considerations." In San Francisco, two city supervisors announced plans to introduce a resolution affirming the right of AI companies to "stand by their values against coercion from the federal government." Anthropic has vowed to challenge the "legally unsound" supply chain risk designation in court. CEO Dario Amodei pointed out a contradiction in the government's position, noting one threat labels Anthropic a national security risk while another insists its AI is essential to national defense. OpenAI's deal includes deploying specialized technical professionals to work with the Pentagon and ensuring the AI models are only used on cloud networks, not on edge devices like drones. Sam Altman stated that his company negotiated for similar terms to be offered to all other AI labs.
