Developers using Fable seem to often get downgraded because of false positives on cybersecurity or biology questions, but I’ve never seen that on GPT-5.6 Sol until today. Trying to find vulnerabilities in my own code. Fascinating and annoying.
@manton I asked Fabel to review some personal health stuff (same questions I ask Opus and Sonnet). It refused with a biology constraint. Very odd.
@manton It’s not exactly predictable, either. This really, truly happened to me: honeypot.net/2026/06/1…