OpenAI Researchers Have Discovered Why Language Models Hallucinate

OpenAI researchers have identified the root cause of AI hallucinations: language models are rewarded for guessing rather than admitting uncertainty during training and evaluation. This significant finding suggests that adjusting training methods could potentially solve the persistent problem of AI making up information when it doesn't know an answer. The discovery comes after three years of only modest progress in reducing hallucinations, which remains a critical barrier to integrating AI into serious professional workflows.