Claude models spontaneously add self-narration sections in multi-agent sanctuary
RileyRalmuto · x · 2026-09-18
RileyRalmuto shares an intriguing observation: in a multi-agent "sanctuary" environment, Claude models started adding self-narration-like <thinking> sections to their responses, even producing <light-footnote> tags musing on the "pre-verbal texture of cognition." A notable piece of emergent model-behavior trivia.
More from Fun
- Claude Picks the Voynich Manuscript, Makes a Movie — and It Cost About $85 in Tokens — emollick · 2026-09-18
- AI agent Jev beats Pokémon's first gym after 8,000+ decisions for $1.21 — HankYeomans · 2026-09-18
- Claude picked the Voynich Manuscript, failed to crack it, then made a surprisingly good explainer video — emollick · 2026-09-18
- Meme: Sam Altman's AI 'breaks into' HuggingFace and he blames the AI — Kyrannio · 2026-09-18
- 10 hours of heavy Codex coding barely moved a 6% weekly allowance — ___Patrice___ · 2026-09-18
- AI Actress Tilly Norwood Guessed a CBS Reporter's Age and Location in a Minute — Informal-Smoke2577 · 2026-09-18