President Trump has issued a directive through a Truth Social post, ordering federal agencies to immediately cease all use of products from AI company Anthropic. This decision follows a public dispute between Anthropic and the Department of Defense (DoD). While a six-month phase-out period was granted for departments currently utilizing Anthropic's offerings, the president unequivocally stated that Anthropic would no longer be considered a federal contractor, declaring, “We don’t need it, we don’t want it, and will not do business with them again.”

Initially, President Trump's announcement did not include any mention of designating Anthropic as a supply chain risk, a consequence that had been previously discussed. However, this changed swiftly when Secretary of Defense Pete Hegseth followed up with a tweet. Secretary Hegseth's post confirmed the president's directive and further instructed the Department of War to "designate Anthropic a Supply-Chain Risk to National Security." This designation has immediate implications, as it prohibits any contractor, supplier, or partner working with the United States military from conducting commercial activity with Anthropic.

At the heart of the Pentagon's dispute with Anthropic was the AI company's staunch refusal to permit its artificial intelligence models to be employed for either mass domestic surveillance or the development and deployment of fully autonomous weapons. Secretary Hegseth had previously deemed these restrictions to be unduly restrictive from the DoD's perspective. Anthropic CEO Dario Amodei publicly reiterated the company's unyielding position, emphasizing their preference to continue serving the Department and warfighters, but only with these two fundamental safeguards in place. Amodei had also stated that should the Department choose to offboard Anthropic, the company would facilitate a smooth transition to an alternative provider to prevent disruptions to military operations and critical missions.

Intriguingly, initially, OpenAI, another prominent AI firm, appeared to support Anthropic's principled stand. According to the BBC, OpenAI CEO Sam Altman circulated an internal memo on Thursday, asserting that he shared similar "red lines" and that any defense contracts involving OpenAI would likewise reject uses deemed "unlawful or unsuited to cloud deployments, such as domestic surveillance and autonomous offensive weapons." Even OpenAI co-founder Ilya Sutskever, who had a public disagreement with Altman in late 2023 before founding his own AI company, weighed in on X (formerly Twitter), commending Anthropic's resolve and the significance of OpenAI taking a similar stance.

However, the situation took a dramatic turn within hours of the Trump administration's directive against Anthropic. OpenAI moved quickly to capitalize on the vacuum created, announcing a new deal with the Pentagon. CEO Sam Altman claimed that this new agreement preserved the very same core principles — prohibitions on domestic surveillance and autonomous weapons — that Anthropic had been fighting to uphold. Reports from the New York Times indicate that OpenAI and the government had already commenced discussions regarding a potential partnership as early as Wednesday of the same week, suggesting a strategic move to fill the void left by Anthropic's forced departure.

It is worth noting that Anthropic, OpenAI, and Google had all received contract awards from the U.S. Defense Department just last July. While some employees at Google have voiced support for Anthropic's position, Google and its parent company have yet to issue an official comment on these unfolding events. The rapid developments suggest that this complex saga of AI ethics, national security, and corporate competition is far from over, with "more twists to come."