ChatGPT’s new default model reduces hallucinations by over 50%
OpenAI’s latest default model for ChatGPT significantly cuts down on false claims, addressing a long-standing AI reliability issue. Internal tests show dramatic improvements in factual accuracy across critical domains.