The model was never the thing that needed governing, the switch was
16 June 2026 (Transformer). Raffi Krikorian, CTO of Mozilla, writing in Transformer on June 16, 2026, about the US government's abrupt order to cut off Anthropic's Mythos and Fable models from foreign access, which led Anthropic to disable them for all customers.
Mythos and Fable can make us all safer. Shutting them down is reckless
Who else?
What it refers to: The power to switch off a deployed frontier AI model, used here through an export-control order with no published threshold, written reason or independent review.
Other voices
- “We must assume a model is compromised and contain it from the start” Satya Nadella, Builder,
- “Disagreeing with the government is the most american thing in the world” Dario Amodei, Builder,
- “Risks from AI went up 10-fold after Mythos” Jamie Dimon,
- “It is critical that alignment and security mature faster than capabilities advance” Anthropic alignment report, Sep 2026,
- “A master class in arrogance and betrayal” Pete Hegseth,










