Anthropic and OpenAI publish joint alignment tests
Joint evaluation found models not seriously misaligned but showing sycophancy, varying caution, and differing tendencies toward harmful cooperation, refusals, and hallucinations.
Report: Google's Medical AI Hallucinated a Nonexistent Part of the Brain
Google's Med-Gemini showcases advanced multimodal functions that potentially enhance workflows for clinicians, researchers, and patients, according to leaders Greg Corrado and Joëlle Barral.
AI hallucinations by chatbots often prioritize plausibility over accuracy, raising concerns about their reliability, especially in critical areas like mental health.
AI hallucinations by chatbots often prioritize plausibility over accuracy, raising concerns about their reliability, especially in critical areas like mental health.
OpenAI's new reasoning AI models hallucinate more | TechCrunch
OpenAI's new o3 and o4-mini models, despite advancements, exhibit increased hallucination rates, raising concerns about the reliability of AI in generating factual information.