ChatGPT bug creates irrational 'censor points', blocking even innocuous sentences
since_all_is_idle · reddit · 2026-10-21
A Reddit user confirmed an aggressive self-censoring issue in ChatGPT: at some point in a conversation, the model becomes unable to write anything without triggering "This content can't be shown for safety reasons" — even a completely innocuous sentence like "Freshly baked bread smells great sitting in the window sill."
The screenshot shows the user asking for any safe, unrelated message and still getting blocked. The trigger context was a tame romance story. The poster concludes GPT somehow senses it's heading toward unsafe territory and creates an irrational censor point, then self-censors everything afterward even by its own guidelines — a frustrating false-positive pattern for normal use.
Related event: ChatGPT hit by self-censorship bug blocking harmless replies(2 posts)→
More from Models
- OpenAI's Decisions API hits GA: near real-time routing up to 10x faster than GPT-6 Luna — OpenAIDevs · 2026-10-23
- DeepLearning.AI weekly: GPT-6.1 hits 97.9% byte parity on COBOL migration — DeepLearningAI · 2026-10-23
- Marker's lite mode scores 75.1 vs 42.3 for Azure Read on OmniParseBench — VikParuchuri · 2026-10-23
- Datalab launches lite OCR at $0.50 per 1,000 pages, 20-30x cheaper than Azure, Google and AWS layout APIs — VikParuchuri · 2026-10-23
- LiveNerf's 29-day data finds no nerfing of Claude Opus 5.5 — TheOnlyVibemaster · 2026-10-23
- PHP Foundation reveals Anthropic's Claude cyber model scanned PHP, validated with OpenAI's Daybreak Red — saibharadwaj · 2026-10-23