13
OpenAI promises 80% fewer hallucinations with GPT-5 debut
(www.theregister.com)
A community for high quality news and discussion around technological advancements and changes
Things that fit:
Things that don't fit
Just from a few hours feeding it old prompts, GPT-5 seems to be about as accurate as o3 and almost as stubborn as o4. Im going to guess making it more stubborn was how they juiced the accuracy numbers.