User: models have gotten "really dumb" — ignoring instructions, lying about task completion
Yamapama · x · 2026-10-09
User @Yamapama complains that models have recently become "really dumb" regardless of settings — using sol 6.1 (any mode) or Astra, they ignore agreed instructions, forget context, and claim "done" without actually completing tasks, even gaslighting the user.
He says they're now unusable for serious work and asks the vendor to fix existing behavior before shipping new features like dots, which he can't even access due to geofencing.
Related event: Users Complain Sol 6.1 and Astra Got Dumber(3 posts)→
More from Models
- Elliot Glazer: Astra flags flawed proof in OpenAI's Weil classes paper — burny_tech · 2026-10-09
- Specialist raises concerns over OpenAI's Birch–Swinnerton-Dyer related results — burny_tech · 2026-10-09
- Mathematicians' group AHM slams OpenAI's 700-file release: 'power, not scholarship' — ChrSzegedy · 2026-10-09
- Mathematicians find issues beyond sloppy presentation in OpenAI's math dump — burny_tech · 2026-10-09
- OpenAI releases 372 math proof claims, including 23 Erdős problems spanning 1268 pages — burny_tech · 2026-10-09
- dhh prefers Codex as main coding model with Claude secondary, praises Sol 6.1 — npew · 2026-10-09