US Gov't Bans Anthropic AI
The U.S. government has ordered a full phase-out of Anthropic's AI from all federal agencies. The move reportedly came after CEO Dario Amodei refused to lower safety guardrails for military and surveillance applications, creating a massive contract vacuum. OpenAI is poised to absorb the business, sparking a debate over its willingness to compromise on safety for government work.
The standoff escalated after Anthropic sought specific assurances that its AI, Claude, would not be used for mass surveillance of Americans or in fully autonomous weapons systems. The Pentagon, however, insisted on the contractual ability to use the technology for "all lawful purposes." This culminated in President Trump ordering all federal agencies to "immediately cease" using Anthropic's technology, allowing for a six-month phase-out period. Defense Secretary Pete Hegseth designated Anthropic a "supply chain risk to national security," a label historically applied to foreign adversaries. This designation prohibits any contractor, supplier, or partner doing business with the U.S. military from also conducting commercial activity with Anthropic. Anthropic has stated its intention to challenge this "unprecedented action" in court. The ban creates a significant void, as Anthropic's Claude was the only frontier AI model deployed on the Pentagon's classified networks. The company had a two-year agreement with the Department of Defense worth up to $200 million to prototype advanced AI capabilities. Its technology was in use or being tested in at least five government agencies, including the Pentagon and the Department of Homeland Security. Hours after the President's order, OpenAI's CEO Sam Altman announced a new agreement to deploy OpenAI's models on the Pentagon's classified network. Interestingly, Altman stated that OpenAI's agreement includes prohibitions on domestic mass surveillance and requires human responsibility for the use of force, the very "red lines" that were at the center of the dispute with Anthropic. The conflict has drawn sharp lines in Silicon Valley. Elon Musk, whose company xAI was recently certified for classified military systems, accused Anthropic of hating "Western civilization." In contrast, a growing number of tech workers from rivals like OpenAI and Google have voiced support for Anthropic's stance in open letters. This public fallout follows recent changes to Anthropic's own safety policies. The company amended its "Responsible Scaling Policy," no longer automatically pausing development of potentially dangerous models and instead considering the actions of competitors. This shift was attributed to a policy environment prioritizing AI competitiveness over safety-oriented discussions at the federal level. Anthropic was co-founded in 2021 by CEO Dario Amodei and other former employees of OpenAI. Amodei has been a vocal advocate for AI safety, estimating a 25% chance of catastrophic outcomes from unchecked AI development if not developed responsibly.
