Summary
Wes Roth covers the sudden US government-ordered suspension of Anthropic’s Fable 5 and Mythos 5 in a video recorded while traveling. The directive, issued to Anthropic at 5:21 PM Eastern, required the company to suspend all access for any foreign national — whether inside or outside the United States, including Anthropic’s own employees — citing national security concerns over a reported jailbreak. Amazon researchers are identified as the party that flagged the vulnerability to authorities, having demonstrated a technique that prompted the models to reveal information about known security vulnerabilities.
Anthropy’s official statement, quoted in the video, pushes back: the company argues the identified vulnerabilities were minor, already discoverable using other public models, and that the government’s action does not meet its own stated standard of a transparent, fair, and technically grounded statutory process. The creator notes the irony that Anthropic CEO Dario Amodei has publicly advocated for stronger government oversight of AI — only to now be on the receiving end of what the company views as an overreaching, opaque directive.
The second half of the video pivots to pre-recorded Fable 5 capability demos, including a procedurally generated interactive space station with real-time shadow rendering, a Balatro-style coin-flip roguelike game, and a self-aware snake game — all generated without external image assets, using only mathematical rendering through Claude’s code generation.
📺 Source: Wes Roth · Published June 13, 2026
🏷️ Format: News Analysis
