AI honesty #1: by 2027 will we have AI that doesn't hallucinate random nonsense?
One-OffManifold MarketsFerméSans KYCDonnées de résolution vérifiéesAssez bien calibréCe marché s'est résolu: Yes (44.4%)
Résolu: 6 août 2026, 03:14
Résolution vérifiée par le fournisseur
CoinRithm dispose d'éléments du fournisseur pour le gagnant et l'heure de règlement de ce marché.
Prévision communautaire actuelleNon 55.6%
Prévisionnistes
46
Type de question
binary
Méthodologie
Play-money forecasting platform
Type de source
Prévision
Données du marché
Mis à jour il y a 14 jours
Obsolète13 oct. 22, 15:526 août 26, 3:14
Chronologie des Probabilités
Yes44.4%
No55.6%
55.6%
44.4%
Règles
By "hallucinate random nonsense" I mean the thing where language models will occasionally just say totally ridiculous out of left field stuff, or blatantly be wrong about the current conversational context (e.g. tell you you said a thing you very definitely did not say).
- Resolves YES if my subjective impression is that language models basically don't do this anymore, or if we end up getting a reasonable benchmark for this and very good performance on that benchmark.
- Update 2026-01-27 (PST) (AI summary of creator comment): The creator indicates the resolution is "pretty close" to being determined.
- The key factor will be where the line is drawn between "out of left field stuff" and "normal amounts of factual incorrectness" when making the final subjective determination.