Since GPT-6 Astra, GPT refusals spike and feel clumsily Anthropic-like, says user
ivan_bezdomny · x · 2026-09-15
Replying to TheStalwart's GPT-6 Astra post, ivanbezdomny reports that GPT was "pretty chill" before Astra — no rejections — but since Astra has become much more Anthropic-like in its guardrails, though "a bit dumber since they are newer to this kind of censorship."
A user-side data point reinforcing reports that OpenAI's latest model ships with stricter, less polished safety behavior.
Related event: Users report GPT-6 Astra refuses more, resembling Claude's style(2 posts)→
More from Models
- ARC Prize 2026 closes in 3 days as Yi-Chia Chen tops ARC-AGI-3 at 75.56% — GregKamradt · 2026-10-31
- Fable 5.5 Max stuns with gorgeous SVG of a pelican riding a bicycle, headed to Claude Code — ChrisGPT · 2026-10-31
- OpenAI reportedly to train on users' Daybreak sessions with no opt-out — dyn___ · 2026-10-31
- NVIDIA Nemotron 3.5 scales LLM self-verification, +9-10 points on agentic benchmarks — sarahcat21 · 2026-10-31
- Mistral Large 4 scores 94% on 1,273 USMLE-style questions but flips 44 at temp 0 — MaziyarPanahi · 2026-10-31
- Researchers find 5 personality structures in LLM weight space, Big Five-like — luke_drago_ · 2026-10-31