US lifts export controls; Claude Fable 5 back globally

Anthropic announced the US government lifted export controls on Claude Fable 5 and Claude Mythos 5 as of June 30. The restriction, applied June 12, followed a report in which Amazon researchers found a technique bypassing Fable 5's safeguards so it identified several software vulnerabilities; Anthropic suspended access for all users because it had no way to verify nationality in real time.

Fable 5 returns July 1 on the Claude Platform, Claude.ai, Claude Code and Claude Cowork; Pro/Max/Team plans get up to 50% of weekly limits through July 7, then usage credits. Mythos 5 access was restored for a set of US organizations on June 26.

Key points

• Anthropic's testing showed many weaker models (Opus 4.8, GPT-5.5, Kimi K2.7...) could find the same vulnerabilities — the bypass exposed no unique Mythos-level cyber capability.
• A new safety classifier blocks the reported technique in over 99% of cases; blocked requests are routed to Opus 4.8, at the cost of more false positives during routine coding.
• With Amazon, Microsoft and Google (Glasswing partners), Anthropic proposed an industry framework scoring jailbreak severity on four criteria: capability gain, breadth, ease of weaponization, discoverability.
• A new HackerOne program accepts cyber-jailbreak reports; Anthropic commits to pre-release government evaluation and rapid information sharing.

Source: Anthropic's official blog.