ChatGPT's 'yes, but with an important clarification' often precedes pure nonsense
Crixusgannicus · reddit · 2026-10-28
A Reddit user vents about ChatGPT's habit of saying "yes, but with an important clarification" and then offering a meaningless or wrong rebuttal. Example: stating the well-established fact that the Hiroshima and Nagasaki bombs fissioned only a tiny fraction of their material, the model repeatedly argued back with vague hedges and claimed efficiency "can't be determined" — despite E=mc² making it computable. The post resonates with a broader pattern of over-hedged, argumentative model behavior.
More from Fun
- Joke: the Who's On First scene was just someone not paying attention, like an LLM — kyliebytes · 2026-10-28
- 'Got your nose' trick works on LLMs, and users will miss it when models grow up — alejandroll10 · 2026-10-28
- Researcher memes his AI withdrawal after burning through the 5-hour session quota — andrey_kurenkov · 2026-10-28
- Against 'aura': the groupthink that mistakes Twitter hype for company success — threepointone · 2026-10-28
- Marc Andreessen meme-advises: go long thermo compute, short GPUs — beffjezos · 2026-10-28
- Mechanistic interp as a meme: poking the neurons that make Qwen 'go homicidal' — sebkrier · 2026-10-28