Anthropic will begin restoring public access to its Claude Fable 5 model on Wednesday, after the U.S. Department of Commerce lifted the export controls it had imposed just over two weeks earlier.
The reversal ends a tense standoff that briefly locked one of the world’s most valuable AI companies out of serving its own users. Both Fable 5 and its more powerful sibling, Mythos 5, are cleared to return.
Key Takeaways
- Commerce lifted export controls on Fable 5 and Mythos 5 on Tuesday
- Access returns to customers Wednesday, roughly 18 days after the ban
- Amazon’s jailbreak warning reportedly triggered the June 12 order
- New safeguard blocks the flagged jailbreak 99% of the time, Anthropic says
- The clash landed weeks after Anthropic’s confidential IPO filing
What The Government Just Reversed
The story starts on June 12, 2026. Anthropic had launched Fable 5 and Mythos 5 only three days earlier. Then Commerce Secretary Howard Lutnick sent a directive to CEO Dario Amodei, ordering the company to block all foreign nationals from using both models.
The order was broad. It barred any foreign national, whether inside or outside the United States, including Anthropic’s own foreign-born employees. The company couldn’t filter those users out in real time, so it did the only thing it could: it pulled both models offline for everyone, worldwide.
The trigger was a reported jailbreak. According to multiple reports, Amazon researchers used a series of prompts to get Fable 5 to hand over information useful for cyberattacks, and CEO Andy Jassy relayed those findings to Treasury Secretary Scott Bessent. Amazon is Anthropic’s largest investor, with roughly $13 billion committed.
Anthropic disputed the severity. The company argued that a “narrow potential jailbreak” should not be cause for pulling a model, warning that the same standard applied industry-wide “would essentially halt all new model deployments for all frontier model providers.” A source close to the company said it was given just 90 minutes to pull Fable, with no prior warning of a national security threat.
Inside The Two Week Standoff
Access came back in stages, not all at once.
The first thaw arrived on June 26. Commerce cleared Mythos 5 for roughly 100 vetted U.S. organizations — Anthropic’s strongest cybersecurity model. That letter, though, left the public-facing Fable 5 untouched.
Talks continued through the following weekend. Then, on June 30, the full reversal came. Lutnick informed Anthropic co-founder Tom Brown by letter that the restrictions were gone entirely.
| Date | What Happened |
|---|---|
| June 9 | Fable 5 and Mythos 5 launched publicly |
| June 12 | Commerce issues export control order; models pulled |
| June 26 | Mythos 5 cleared for ~100 trusted organizations |
| June 30 | Export controls lifted on both models |
| July 1 | Access restoration begins for customers |
Lutnick framed the outcome as a win for U.S. competitiveness. Writing on X, he said his office had “worked closely with Anthropic to analyze and approve Fable 5 to ensure alignment across the US Government and strengthen America’s leadership in AI.”
Anthropic kept its own statement short. “We’re grateful to our users for their patience, and to everyone who worked with us on redeploying the models,” the company wrote.
The New Safeguards Anthropic Agreed To
This isn’t a simple return to how things were before June 12. The deal came with conditions.
The centerpiece is a fix for the exact jailbreak that set off the crisis. Anthropic says the new safeguard blocks the jailbreak officials worried about 99% of the time. In the rare cases it slips through, the company says the output only surfaced already-known or already-patched security flaws.
There’s a tradeoff, though. The tighter filter can catch harmless requests too. Anthropic acknowledged the new safeguard could result in Fable flagging more benign requests. Commerce’s own Center for AI Standards and Innovation tested both the old and new guardrails before signing off.
Anthropic also made broader commitments. Per Lutnick, the company agreed to proactively detect and address security risks, to work with the government on protocols and release standards for Mythos, Fable, and future models, and to inform officials of any malicious activity. Notably, Anthropic had already pledged much of this voluntarily, months before the export rule existed.
Why This Matters Beyond Anthropic
The episode set a precedent that rattled the whole industry. Before June 12, few expected the U.S. government to use export controls on a domestic company’s consumer product. Now it has happened once, which means it can happen again.
The ripple effects were immediate. Rival OpenAI limited the rollout of its new GPT-5.6 model to a small set of government-approved partners after pressure from the administration. That points to a bigger problem critics flagged from the start.
The government’s role in evaluating frontier models before release remains unsettled, creating an ad hoc regulatory environment for AI companies. Policy advocates on both the right and left had blasted the original ban.
As Brad Carson, president of Americans for Responsible Innovation, put it: “When ad hoc executive actions replace clear standards, America risks surrendering its lead in AI and allowing genuinely dangerous technology to be deployed.”
Meanwhile, competition abroad keeps building. China is moving closer to producing its own Mythos competitor while continuing to release open-weight models that rival Anthropic’s Claude Opus 4.8.
What Comes Next For Anthropic
The timing of this whole saga was awful for Anthropic. It unfolded right in the company’s IPO window.
Anthropic filed confidentially for an initial public offering at a reported valuation topping $900 billion. A prolonged ban on its flagship models threatened to complicate that timeline and raise the regulatory risk premium hanging over frontier AI firms. With access restored, that overhang eases — though the precedent doesn’t vanish.
More rules are coming, too. The Trump administration faces an August deadline under a recent executive order to create standardized benchmarks for evaluating the security risks of new AI models. Those standards could reshape how every lab ships its next release.
And not every fight is over. Anthropic remains locked in separate litigation with the Defense Department, which labeled the company a supply chain risk earlier this year after contract talks collapsed. For now, though, the immediate crisis has passed, and Fable is coming back online.
Digital Trendings is your trusted source for AI news and updates — stay tuned for more.
