Trump Reverses Course on Anthropic, Calls AI Firm 'Responsible' After G7 Meeting

The immediate consequence of President Trump's statements is a likely reduction in direct, overt pressure on Anthropic. The company, which was previously compelled to disable its newest AI models, Fable 5 and Mythos 5, may now face a more constructive dialogue with the administration regarding the development and deployment of advanced AI. This shift suggests a potential preference for negotiation and compliance over outright confrontation in the administration's approach to sensitive AI technologies.

However, the President's comments also indicate that the threat of executive action, specifically through the Defense Production Act, remains a potent tool. This implies that while the current standoff has eased, Anthropic and other AI developers will likely operate under a watchful eye, with the expectation of strict adherence to government directives. The industry can expect a continued focus on transparency and responsible development, particularly concerning models deemed to have dual-use capabilities that could pose national security risks.

For the past three months, the Trump administration had adopted an increasingly aggressive posture toward Anthropic, a leading developer of advanced artificial intelligence. This culminated in an export control directive that forced the company to suspend access to its powerful Fable 5 and Mythos 5 AI models. The administration cited unspecified national security concerns for these actions.

The tensions appear to have reached a turning point during the G7 summit in France, where President Trump held a meeting with several tech leaders, including Anthropic CEO Dario Amodei. It was after this meeting that Trump, in a pre-taped interview with Axios, made his surprising remarks yesterday. He explicitly stated, when asked if he considered Anthropic a threat, 'Well, not now. But a week ago, maybe.' He further praised the company for having 'behaved very responsibly' and for responding to the administration's directive 'very quickly' and 'responsibly.'

Despite this apparent de-escalation, the President did not retract the possibility of invoking emergency powers under the Defense Production Act against Anthropic, a tool that grants the government broad authority over private industry in times of national emergency. This suggests that while the immediate crisis has passed, the underlying issues regarding advanced AI capabilities and their potential impact on national security remain a point of contention between the government and developers.

President Trump's reversal on Anthropic carries significant implications not just for the AI company itself, but for the entire artificial intelligence industry and the evolving relationship between powerful tech firms and national governments. For Anthropic, the immediate relief from being labeled a national security threat could free up resources and allow for a clearer path forward in developing and deploying its advanced models, albeit likely under continued governmental scrutiny.

More broadly, this episode provides a crucial data point in how governments intend to regulate and interact with frontier AI. The administration's willingness to de-escalate after direct engagement, rather than maintaining an adversarial stance, could set a precedent for future interactions with other AI companies. It suggests that proactive engagement and demonstrated 'responsible' behavior from AI developers might be key to avoiding more stringent, heavy-handed regulation.

However, the lingering threat of the Defense Production Act also signals that governments are prepared to use their most powerful tools if they deem it necessary. This creates a complex operational environment for AI companies, where innovation must be balanced against the imperative of national security, and where direct communication with policymakers may become as critical as technical breakthroughs. The market will be watching closely to see if this marks a sustained shift towards dialogue, or merely a temporary pause in a larger regulatory push.