Report from the New York Times
In Brief – The Trump administration lifted export restrictions that it had imposed on June 12th prohibiting Anthropic from allowing any foreign nationals from accessing it’s top AI models, allowing the company to restore access to Claude Mythos and Claude Fable for its top clients. Commerce Secretary Howard Lutnick said the move followed two weeks of negotiations during which Anthropic agreed to implement safeguards to address the US Administration’s national security concerns, including proactively identifying security risks, coordinating with the US government on protocols and standards for current and future models, and reporting malicious activity. The reversal eases tensions in a contentious relationship between the government and Anthropic, which was designated a supply-chain risk by the Pentagon earlier this year after disagreements over military applications of AI. Anthropic, which challenged that designation in court, explained its latest security policies. The Commerce Department reserved the right to revisit its latest export control decision if Anthropic fails to meet its obligations.
Context – The Trump Administration’s AI policy has been aggressively anti-regulatory, arguing that innovation and investment is critical to competing against China, and has pulled policy internationally in that direction. However, cracks are building at home based on increasing public anxiety, a vocal cadre of anti-Big Tech conservatives, and a culture of sci-fi AI dystopia that offers scary stories when AI developers and business leaders cannot predict how the technology will evolve. The Administration’s ad hoc security interventions are internally divisive. Anthropic’s progressive leadership is clearly impacting the ongoing cyber and national security policy machinations involving the company because opposition to “Woke AI” is one of the few things that unites the pro-tech and anti-tech conservatives. Foreign access shutdowns could become the Administration’s default policy tool to push AI developers to be cooperative on security concerns because their export control authority is broad and flexible, which is further boosting international concerns about US tech and AI kill switches and the EU’s tech sovereignty drive.