descriptive claim
Anthropic trains models to refuse some requests, which is distinct from a probe-triggered fallback.
Thariq Shihipar · Claude Code’s Next Era — Thariq Shihipar, Anthropic · 29 Sept 2026
In their words
The transcript segments this claim was taken from
“to train this into the model and we still do this as well the model will refuse a request that's not a fallback right so like it's not a probe that's activating and falling back it's just refusing to do it and we we do this training um but it's like”Play from 1:18:50Report a misattribution
95%how strongly it was put, not whether it is true
80%importance to the conversation
96%confidence the claim was read correctly