Amazon's Security Research Reportedly Triggered White House Ban on Anthropic's Fable AI

A recent report by The Wall Street Journal indicates that security research conducted by Amazon, coupled with direct conversations between CEO Andy Jassy and White House officials, played a pivotal role in the government's decision to issue an export control directive. This directive effectively banned foreign nationals from accessing Anthropic's advanced AI models, Fable 5 and Mythos 5, marking a significant development in the ongoing discourse surrounding AI security and international policy. According to the Amazon paper cited in the report, researchers were allegedly able to exploit Fable 5 through a series of carefully crafted prompts, compelling the AI to generate information that could potentially be leveraged in cyberattacks. Shortly after Jassy presented these findings to the government, the White House moved to block the use of these models by non-US citizens. This decision had immediate and far-reaching implications, particularly for Anthropic itself, as many of its own researchers, being foreign-born, found themselves suddenly barred from accessing the very products they helped develop. Anthropic, however, has publicly disputed the government's characterization of the issue, specifically pushing back against the term "jailbreak." The company argues that the vulnerabilities identified by Amazon are not unique to its models and can, in fact, be replicated using other publicly available AI systems, including OpenAI's GPT 5.5. This stance has garnered support from some independent security researchers, with figures like Katie Moussouris, CEO of LutaSecurity, stating that the paper does not describe a "jailbreak" in the traditional sense. Adding another layer of complexity to the situation is the long-standing tension between Anthropic and the Trump administration. The company has previously resisted government demands to allow its AI to be used for mass surveillance of American citizens or to power lethal autonomous weapons. This friction culminated in February when Trump instructed federal agencies to cease using Anthropic's AI, followed by Secretary of Defense Pete Hegseth designating the company a supply chain risk. While there had been signs of a potential reconciliation, with the government and Anthropic reportedly working together to expand access to Mythos, this latest directive signals a renewed period of contention. The incident underscores the delicate balance between fostering AI innovation, ensuring national security, and navigating the ethical implications of advanced artificial intelligence, suggesting that the relationship between tech giants and government bodies will remain a critical area of focus.