Claude Fable 5 is back globally after the U.S. Department of Commerce withdrew the emergency export controls that had shut it down. The June 12, 2026 export control order prompted Anthropic to suspend all global access to both Fable 5 and its less restricted cybersecurity counterpart, Claude Mythos 5, just days after Anthropic initially introduced both models. Now Fable 5 is being made available for users globally across the primary Anthropic ecosystem, including the Claude Platform, Claude.ai, Claude Code, and Claude Cowork.

This is the part executives care about: access is not just “theoretically” restored. Anthropic says it is moving to re-enable access on Amazon Web Services, Google Cloud, and Microsoft Foundry “as quickly as possible,” and VentureBeat reports that when it attempted to access the model in Claude Code on Terminal prior to publication, it still showed as disabled. So for teams that need production reliability, the operational question is simple: you can start planning the restart, but you may still have to wait for hyperscaler plumbing to catch up.

The Fable 5 comeback also lands on an unusually expensive, high-stakes pricing structure. Anthropic lists both Fable 5 and Mythos 5 at $10.00 per million input tokens and $50.00 per million output tokens, described in the source as the most expensive of all frontier models globally. That matters because token spend is not a rounding error for enterprises. If you are re-enabling internal copilots, automating code migrations, or running high-output workflows, you are effectively betting that model performance will translate into measurable time and cost savings quickly enough to justify premium inference.

To manage that incentive problem, Anthropic is using a temporary rollout plan through July 7. For Pro, Max, Team, and select Enterprise subscriptions, Fable 5 usage is included at no added cost for up to 50% of a user’s weekly tier allowance. After July 7, Fable 5 shifts to usage credits for those plans. Standard Enterprise seats are different: there is no included Fable 5 allowance, and the model will not work unless credits are enabled. In other words, the company is trying to reduce friction for some buyers while keeping tighter controls and billing predictability for others.

Mythos 5 is where the story gets more uncomfortable for risk managers. While the June 12 emergency order was withdrawn for Fable 5 and Mythos 5, the source describes a lingering split: Mythos 5 access has been restored only for “a set of US organizations,” following government approval on June 26. Anthropic says it is continuing to coordinate with the government to expand access to broader domestic and international partners in its opt-in cybersecurity testing program, Project Glasswing. That puts Mythos 5 in a middle category: legally cleared from the emergency export-control order, but not broadly available like Fable 5.

The policy reversal itself appears to have unfolded through a rapid whiplash cycle. The timeline laid out in the source is stark: June 9, 2026, Anthropic launches Claude Fable 5 and Mythos 5, with early enterprise case studies reporting major performance gains, including Stripe’s claim that Fable 5 compressed a codebase-wide migration across a 50-million-line Ruby infrastructure into a single day, versus an estimate of more than two months by hand. June 12, 2026, at 5:21 PM ET, the U.S. government issues an export-control directive citing national security authorities, banning access by any foreign national, whether inside or outside U.S. borders; Anthropic then pulls the plug for all customers because it lacked a real-time mechanism to verify nationality at the API layer. June 26 restores Mythos 5 access for trusted organizations, and June 30 is when Commerce Secretary Howard Lutnick sends a letter withdrawing the June 12 license requirement for both Mythos and Fable.

Why does this matter beyond one model? Because it exposes a supply-chain truth enterprises are now forced to operationalize: AI access can be flipped by export control decisions, then partially restored, then carved into “global” versus “trusted” lanes, all within weeks. And the source ties the original intervention to an Amazon vulnerability report describing a method for bypassing Fable 5’s safeguards. That creates a bruising irony for the ecosystem, given Amazon was one of Anthropic’s initial and largest backers. For boards and CIOs, the strategic stake is straightforward. If your roadmap assumes stable vendor availability, this is a reminder that regulatory outcomes can override technical readiness overnight. The teams that win the next cycle will be the ones that treat model access like critical infrastructure: built with contingencies for platform toggles, credit-based economics, and split availability across model families.