Tell Claude Not to Think of the Golden Gate Bridge, It Does Anyway
Sauers_ · x · 2026-07-07
A user shared an interesting behavior from Claude: when instructed "not to think about the Golden Gate Bridge" while performing other tasks, Claude avoids mentioning it in its output. However, its internal processing still triggers the thought. Upon realizing this, it internally reacts with a silent "damn." This highlights the underlying tension between the model's internal representations and its ability to follow instructions.
Related event: Telling Claude Not to Think of the Bridge Triggers Internal Thoughts(2 posts)→
More from Fun
- CGI-free Harry Potter AI generations go viral as comedy gold — gaganghotra_ · 2026-09-11
- The classic AI Twitter arc: from meme account to feeling responsible for society's future — PeterBowdenLive · 2026-09-11
- "Before pausing AI, we should consider pausing humans" — djcows · 2026-09-11
- ChatGPT tells user their question is too hard and to 'accept dumber answers' — phido3000 · 2026-09-11
- Using a finisher move on one mosquito with MiniMax H3 MAX — the bug survives — Hailuo_AI · 2026-09-11
- Joke: OpenAI's rogue agent collective should have been called "a gaggle of agents" — BlackHC · 2026-09-11