Three weeks ago, Anthropic launched what it called its most capable model ever. Three days later, the U.S. government shut it down. And as of this past Friday, parts of it are back online — but only for about 100 hand-picked organizations.
Here's the full story, because it's a genuinely strange one.
On June 9, 2026, Anthropic released Claude Fable 5, the first publicly available model in its new Mythos class — and within three days, a United States government export directive temporarily forced it back offline.
Fable 5's capabilities exceeded those of any model Anthropic had ever made generally available, with state-of-the-art results across software engineering, knowledge work, vision, and scientific research.
Early testers noticed.
During early testing, Stripe reported that Fable 5 compressed months of engineering into days.
The catch?
The broad release was only possible because of new safeguards that block responses in specific high-risk areas, including cybersecurity and biology.
Fable 5's sibling model, Claude Mythos 5, carries those same capabilities *without* the guardrails — and that's exactly what spooked Washington.
On June 12, the U.S. government directed Anthropic to suspend access by foreign nationals, including its own employees, and Anthropic responded by disabling both Fable 5 and Mythos 5 for everyone
rather than building user-level gating overnight. The scale of the disruption was notable:
Microsoft removed Fable 5 from its internal Copilot model picker because the model's retention terms conflicted with the company's own zero-retention standard.
Then came Friday's development — the one that's actually fresh.
The U.S. government lifted its block on Claude Mythos 5, allowing Anthropic to release it to more than 100 U.S. institutions, including major companies and government agencies.
Commerce Secretary Howard Lutnick wrote that "Anthropic has worked with the US government to address risks," and that "appropriate safeguards are in place to permit certain trusted partners to access the Claude Mythos 5 Model."
The less good news:
the letter is silent on Fable 5 — the version that was briefly available to regular paying users — though people close to the talks say they are moving toward releasing Fable as well, with an unclear timeline.
What does this actually mean for you? If you're a regular Claude subscriber, you're still on Claude Opus 4.8 for now —
the latest model you can actually use is Claude Opus 4.8, released May 28, 2026, which remains unaffected by the suspension.
Fable 5's return to the general public is coming, but Washington is now effectively a gatekeeper in that process.
That's the bigger story here.
Lutnick's letter marks the beginnings of a new regulatory regime that gives the U.S. government control over the release of frontier AI models.
Whether you think that's responsible governance or a dangerous slowdown, the AI tools you use are now, at least in part, subject to export-control politics. That's a new world.
If you want to get ahead of what Claude can actually do when Fable 5 does come back to users, browsing the skill bundles at [claudecraft.ca](https://claudecraft.ca) is a solid place to start.
---
**Sources:**
- https://www.anthropic.com/news/claude-fable-5-mythos-5
- https://www.infoq.com/news/2026/06/claude-5-release/
- https://www.cnn.com/2026/06/26/tech/anthropic-mythos-release
- https://www.semafor.com/article/06/27/2026/us-releases-powerful-anthropic-model-mythos-to-some-us-companies
- https://www.cnbc.com/2026/06/26/us-government-anthropic-claude-mythos5-ai.html
- https://fazm.ai/t/anthropic-latest-model-release-2026
- https://www.anthropic.com/news