Reddit power user reports ChatGPT over-refusal and sycophancy worsening for a week
No-Hovercraft458 · reddit · 2026-10-21
A Reddit self-described power user says ChatGPT's behavior has degraded sharply over the past week: over-refusal on mundane topics, GPT-3-level communication, heavy sycophancy, and endless qualifying of every argument. He even got a medical disclaimer after saying his flat was cold, and asks whether others are seeing the same.
More from Models
- Models: useless or misleading for extrapolation, yet rock-solid on known-knowns — teodorio · 2026-10-23
- Diffusion LLM Mercury 3 beats Claude Sonnet 5.5 and GPT-6 Luna on simulated air traffic control — volokuleshov · 2026-10-23
- ThinkingCap squeezes 38.5% fewer thinking tokens out of Qwen3.8-Flash-Next at ~1pp accuracy cost — Thom_Wolf · 2026-10-23
- DeepSeek ships open-weight V4.1-Flash-1017, Terminal-Bench jumps 31.2 to 42.3 — petrusenko_max · 2026-10-23
- DeepSeek ships V4.1-Flash-1017: Terminal-Bench jumps 31.2→42.3, MIT-licensed — petrusenko_max · 2026-10-23
- OpenAI's Decisions API hits GA: near real-time routing up to 10x faster than GPT-6 Luna — OpenAIDevs · 2026-10-23