Your cheat sheet to Anthropic's latest drama with the White House

In an unprecedented move, the Trump administration Commerce Department has forced AI powerhouse Anthropic to abruptly disable its newly launched, most advanced AI models, Fable 5 and Mythos 5, for all global users after issuing an Export Control Directive on June 12, 2026. Citing nebulous 'National Security Concerns' over a potential 'AI Jailbreak' in Fable 5 that could exploit software vulnerabilities, the directive marks a significant escalation from previous hardware restrictions to directly controlling deployed commercial AI models. This shock intervention comes as Anthropic was reportedly preparing for a nearly $1 trillion IPO, sending ripples of uncertainty through the burgeoning AI market. The drastic order is the latest flashpoint in an escalating feud between Anthropic and the Trump administration, which began earlier this year when the Pentagon placed Anthropic on a 'Pentagon Supply Chain Blacklist' in March 2026. That earlier conflict stemmed from Anthropic refusal to permit its AI models for domestic surveillance and 'Fully Autonomous Weapons Systems', a stance that positioned the company against the administration's evolving AI policy, which has shifted from an initial deregulatory approach to a more assertive, security-focused agenda. The White House reportedly acted after Amazon CEO Andy Jassy raised concerns about Fable 5 guardrails, and amid suspicions that a China-linked group had accessed the Mythos model. Anthropic has vehemently disputed the government's assessment of the jailbreak's severity, deeming it a 'narrow, non-universal' issue and arguing other publicly available 'Frontier AI Models' share similar capabilities, claiming the directive sets a dangerous precedent that could stifle innovation across the industry. With cybersecurity experts decrying the ban as detrimental to defenders, the episode intensifies calls for a comprehensive federal AI governance framework, such as the proposed 'Great American Artificial Intelligence Act of 2026', as policymakers grapple with defining the boundaries of AI regulation and national security in an increasingly complex geopolitical tech race. The ball is now in Anthropic court to resolve the issue with the Commerce Department to restore access.