Alarming Revelation: OpenAI’s New Reasoning AI Models Face Worsening Hallucinations

Increased AI Hallucinations: OpenAI's new reasoning models, o3 and o4-mini, are experiencing higher rates of AI hallucinations compared to previous models, raising concerns about their reliability in critical sectors like finance and cryptocurrency.

Implications for Trust and Usability: The elevated hallucination rates could undermine trust in AI systems and pose challenges for businesses, while also highlighting the need for improved accuracy through strategies such as integrating web search capabilities.