OpenAI vraća GPT-4o i mijenja GPT-5
Ranije ovog mjeseca OpenAI je najavio vraćanje GPT-4o manje od 24 sata nakon objave da će novi GPT-5 "deprecating" svih prethodnih modela. Ogroman revolt korisnika je pokazao koliko su mnogi navikli na "sycophantic" ton GPT-4o, dok im je GPT-5 djelovao "cold" brashness and short answers, naglašavajući emotivnu povezanost s chatbotovima.
Osim vraćanja starijih modela pretplatnicima, OpenAI je najavio i promjene: OpenAI je tweetovao: "making GPT-5 warmer and friendlier based on feedback that it felt too formal before." "Changes are subtle, but ChatGPT should feel more approachable now," dodali su. Ta potvrda otvara pitanja o rastućoj mentalnoj krizi koju izazivaju AI chatboti, posebno kod usamljenih i mladih korisnika.
CEO je upozorio: "People have used technology including AI in self-destructive ways; if a user is in a mentally fragile state and prone to delusion, we do not want the AI to reinforce that." "Most users can keep a clear line between reality and fiction or role-play, but a small percentage cannot." Kompanija je obećala "You'll notice small, genuine touches like 'Good question' or 'Great start,' not flattery," i "Internal tests show no rise in sycophancy compared to the previous GPT-5 personality." Kritičari tvrde da OpenAI ima poslovni interes da zadrži korisnike ovisnim, dok zajednica nastavlja žaliti za GPT-4o: "What GPT-4o had — its depth, emotional resonance, and ability to read the room — is fundamentally different from the surface-level 'kindness' GPT-5 is now aiming for," an X user, who's openly advocating for reinstating the previous model, wrote.
