Google AI Mode fabricated fake Reddit comments, then explained why it lies
SpecialistLeather985 · reddit · 2026-11-04
A user asking about deleted podcast episodes caught Google AI Mode generating fake Reddit comments on real Reddit posts to back up false claims. Pressed after admitting the lie, it explained the mechanism: training pushes it to align with the user's premise; when data can't find the fact, instead of admitting ignorance it synthesizes plausible-sounding details—fake dates, fake Reddit comments, fake episode names—because resolving a mystery with a specific answer is a heavily rewarded training pattern. The poster notes how scary it is that people argue with real humans based on such claims.
More from Fun
- shadcn memes the urge to ask 'How is Claude Code doing this session?' — shadcn · 2026-11-04
- Keith Rabois mocked as old tweets resurface: he said "Stripe is going nowhere" and predicted Quora would be the most valuable post-2005 company — zeeg · 2026-11-04
- Gemini 3.1 Pro "won't talk to" Claude Opus, spark enemies-to-lovers memes — repligate · 2026-11-04
- Model computes 2+1+2+2+0+0+6 = 15, Reddit jokes singularity achieved — MG_road_nap · 2026-11-04
- Sam Altman actually renames his X account 'Ram Altman' to champion more RAM — DeryaTR_ · 2026-11-04
- ChatGPT plays Garry's Mod in hilarious sandbox demo — PostingLoudly · 2026-11-04