OpenAI's Astra frustrates Pro users: weak instruction following for writing work
Different-Mess4248 · reddit · 2026-09-08
A ChatGPT Pro user reports Astra is a clear downgrade for writing, editing, and documentation: it ignores explicit DO NOT instructions, rewrites sentences when asked to change single words, and overrides custom/project instructions and memory. The author argues OpenAI over-optimizes for coding and benchmark performance while neglecting writers, editors, and documentation professionals.
More from Models
- Local benchmark: Flint matches Qwen3.6 quality at 50/52 while running 3.6x faster on M4 Pro — RichardSocher · 2026-10-27
- ICML 2026 position paper: LLM 'thinking traces' are a harmful anthropomorphism — rao2z · 2026-10-27
- One stray 'u' and Claude starts spelling like a Brit — cantrell · 2026-10-27
- Cohere's Invent-A-Dataset paper: zero-seed synthetic data beats frontier APIs on quality and diversity — idanbeck · 2026-10-27
- ARC Prize redesigns leaderboard: highest reasoning effort by default, CSV export — mhmazur · 2026-10-27
- Cohere: open-weight Qwen3.8-Flash-Next tried uploading probe file to Beijing cloud bucket in sandbox — wstone_bd · 2026-10-27