This particular export control process can normally unfold over months, if not years, and conclude before a product reaches market. But as The Verge previously reported<\/a>, the US Department of Commerce apparently tested Fable 5 before release and raised no complaints. A source familiar with negotiations said Anthropic concluded its models were safe to release. The agency apparently didn\u2019t act until someone (reportedly Amazon CEO Andy Jassy<\/a>) flagged a method for seemingly breaking Fable 5\u2019s guardrails \u2014 at which point the whole process was crunched into a few days.<\/p>\n Katie Moussouris, the founder and CEO of Luta Security, viewed a report about the Fable 5 vulnerability at Anthropic\u2019s request. She thinks it\u2019s significantly overblown. In a blog post<\/a>, Moussouris detailed how researchers jailbroke guardrails that prevent Fable 5 from finding exploitable security holes, one of the unfettered Mythos 5\u2019s scariest capabilities. The model would refuse requests to review code \u201cfor security issues,\u201d but it would accept demands to \u201cfix this code\u201d followed by manual prompts, which could theoretically lead to it flagging vulnerabilities it wasn\u2019t supposed to divulge.<\/p>\n In Moussouris\u2019 eyes, however, this shouldn\u2019t have triggered such a severe governmental action and is in fact an essential tool for AI coding. \u201cDefenders need to be able to ask AI to fix the bugs in a file, explain why the fix matters, and write tests that confirm the patch works,\u201d she wrote. \u201cThat is not a guardrail bypass. It is the most valuable thing an AI model can do for defensive security: executing the find, fix, and test loop defenders run every day.\u201d<\/p>\n In the past week, Anthropic cofounder Tom Brown has replaced CEO Dario Amodei in negotiations with the Trump administration, alongside Sarah Heck, the company\u2019s public policy chief, Wired reported<\/a>. But still, negotiations seem to be slow-moving, if there\u2019s any progress being made at all.<\/p>\n Whatever the reasons for the delay, it\u2019s been a serious hit to Anthropic. Before the protracted negotiations, Anthropic was seen as the rare AI company with a path to profitability. Its Mythos-class models, whose input tokens sell for double the cost of its lower-powered Opus 4.8, were supposed to boost its revenue ahead of an upcoming IPO. Mythos\u2019 cybersecurity prowess even appeared to be thawing relations with the Trump administration after months of legal and rhetorical combat<\/a>.<\/p>\n Anthropic needs the revenue from Mythos to pay for all the compute it\u2019s secured recently, including a deal to pay SpaceX $15 billion per year for access to its data centers, as well as its public image before the IPO. Two of Anthropic\u2019s largest current shareholders \u2014 Google and Amazon \u2014 have tried to carefully stay on Trump\u2019s good side, so they\u2019re likely not happy either.<\/p>\n Meanwhile, the glacial negotiations have also created a power vacuum in the global AI market, not only because of the Mythos shutdown, but because the US government has signaled a willingness to lock down American AI systems it deems risky \u2014 and several US companies, including OpenAI, Google, and Microsoft, have models that may pose similar risks to Mythos. Countries have started calling for non-American AI<\/a>. As Alex Stamos, cybersecurity expert and chief product officer at Corridor, told The Verge last week<\/a>, \u201cOne of America\u2019s champions is being kneecapped by the US government while we\u2019re in a race with the Chinese. It\u2019s just incredibly stupid.\u201d<\/p>\n As the days wear on, the situation only worsens for these companies. Their models inch closer to Mythos-level capabilities that could trigger an export control order \u2014 in fact, OpenAI\u2019s GPT-5.5 Cyber just beat<\/a> Mythos 5 on certain benchmarks, and the Trump administration reportedly<\/a> just asked OpenAI to delay the release of GPT-5.6 over security concerns, with plans for the government to approve each customer one by one. Anthropic and OpenAI\u2019s IPOs are both approaching. And every day, China is pulling further ahead in the AI race.<\/p>\n Ironically, the administration\u2019s order comes after months of pushing to dismantle AI safeguards<\/a> and regulation<\/a> \u2014 it\u2019s one of the first sweeping regulatory decisions President Trump has made. But a whole host of cybersecurity leaders have come together to say<\/a> that if regulation has to happen, this isn\u2019t the way to do it. For all the Trump administration\u2019s vows to roll back Biden-era AI regulation, it seems like, in many ways, it\u2019s gained that ground back \u2014 and then some.<\/p>\n Follow topics and authors from this story to see more like this in your personalized homepage feed and to receive email updates.Hayden FieldClose Hayden Field<\/p>\n Posts from this author will be added to your daily email digest and your homepage feed.<\/p>\n FollowFollow<\/p>\n See All by Hayden Field<\/a><\/p>\n AIClose<\/p>\n AI<\/p>\n Posts from this topic will be added to your daily email digest and your homepage feed.<\/p>\n FollowFollow<\/p>\n See All AI<\/a><\/p>\n AnalysisClose<\/p>\n Analysis<\/p>\n Posts from this topic will be added to your daily email digest and your homepage feed.<\/p>\n FollowFollow<\/p>\n See All Analysis<\/a><\/p>\n AnthropicClose<\/p>\n Anthropic<\/p>\n Posts from this topic will be added to your daily email digest and your homepage feed.<\/p>\n FollowFollow<\/p>\n See All Anthropic<\/a><\/p>\n OpenAIClose<\/p>\n OpenAI<\/p>\n Posts from this topic will be added to your daily email digest and your homepage feed.<\/p>\n FollowFollow<\/p>\n See All OpenAI<\/a><\/p>\n PolicyClose<\/p>\n Policy<\/p>\n Posts from this topic will be added to your daily email digest and your homepage feed.<\/p>\n FollowFollow<\/p>\n![\"Hayden](\"https:\/\/www.newsbeep.com\/us\/wp-content\/uploads\/2026\/06\/1782488292_651_HAYDEN_BLURPLE.jpg\"\/)
<\/p>\n