ChatGPT Lead: GPT-5 Still Faces "Hallucination" Issues, Users Advised to Verify Answers

BlockBeats News, August 18 — Although ChatGPT has already launched the new GPT-5 model, there is still a possibility of errors. An OpenAI senior executive reiterated this point again this week. Nick Telle, head of ChatGPT, stated in an interview: "When it comes to reliability, there is a significant gap between being reliable and being completely reliable."

He further explained: "Unless we can demonstrate that ChatGPT is more reliable than human experts in all areas, not just some, we will continue to recommend that you carefully verify the answers."