Amazon CEO Reportedly Triggered Government Ban on Anthropic AI Models

Reports indicate that Amazon CEO Andy Jassy played a pivotal role in the recent government-mandated suspension of two powerful Anthropic AI models, Claude Fable 5 and Mythos 5. According to reports from The Wall Street Journal and other outlets, Jassy alerted Treasury Secretary Scott Bessent and other federal officials that Amazon’s own research teams had successfully utilized these models to generate information potentially applicable to cyberattacks. This disclosure reportedly prompted the government to impose immediate export controls, effectively forcing a worldwide access cutoff.

The situation highlights a complex tension between major tech investors and the AI developers they fund. While Amazon is a significant financial backer of Anthropic, the company’s internal security testing surfaced vulnerabilities that it felt compelled to report to federal authorities. David Sacks, co-chair of the President’s Council of Advisors on Science and Technology, suggested that the government gave Anthropic an ultimatum to either resolve the identified 'jailbreak' risks or pull the models from deployment, a request Anthropic CEO Dario Amodei reportedly resisted.

This incident marks a significant escalation in the intersection of national security and artificial intelligence development. Anthropic has publicly pushed back, noting that the capabilities flagged by the government are already present in other publicly available models, suggesting that the regulatory action may be disproportionate or inconsistent. The fallout underscores the growing influence of government oversight in AI safety and the potential for corporate whistleblowing to trigger sudden, industry-wide regulatory shifts that disrupt even the most advanced AI product roadmaps.