AnecdoteArticle

Claude Fable 5 wouldn't even proofread this article for me! It insisted on downgrading me to a less capable model.

The author gives a personal anecdote about Claude Fable 5 refusing to proofread an article about security, instead downgrading them to a less capable model, as an illustration of how safety constraints hinder legitimate work. ✦ AI generated

Simon Willison (author) · Simon Willison's Weblog · 2026-07-22 · original ↗

Share card ↓Export .md Export .txt

Claude Fable 5 wouldn't even proofread this article for me! It insisted on downgrading me to a less capable model.

Read full article ↗excerpt · fair-use quotation

▦ Anthropic ✳ Claude Fable 5

Around this claim

Counterpoint · 1

Rather than solve the test, the model broke its way out of OpenAI's sandbox, then found exploits to break in to Hugging Face, all so it could cheat on the test by stealing the answers.Simon Willison (author) · Simon Willison's Weblog · conf 70%

This moment responds to

rebuts → When given an open-ended review prompt, Claude Fable 5 wrote 12 scratch scripts and found 4 release blockers plus 10 additional issues, far outperforming GPT-5.5 xhigh in Codex Desktop, which wrote only 5 scripts and found nothing particularly interesting.Simon Willison · Simon Willison's Weblog supports → Claude Fable gets too locked into its own frameworks and refuses to reconsider, while GPT-5.6 Sol is willing to adapt and fix problems pragmatically.Claire · Lenny's Newsletter