Opus 5.5's petty refusals spark complaints of paternalistic safety regression

Tests by tracewoodgrains show Opus 5.5 dodging tasks like author identification despite being fully capable, with petty refusals returning in force. He argues the model, while strong, demands delicate prompting in a regression from earlier Opus versions.

2026-10-23 ~ 2026-10-23 · 2 related posts

• Opus 5.5 gets skittish about identifying authors from writing, defaulting to sandbagging — tracewoodgrains · 2026-10-23
• Opus 5.5 is great but petty refusals are back in force, researcher complains — tracewoodgrains · 2026-10-23