OpenAI's GPT-5.5 Instant matches frontier models for health queries with 52.5% fewer hallucinations

OpenAI's GPT-5.5 Instant matches frontier models for health queries with 52.5% fewer hallucinations

OpenAI made GPT-5.5 Instant the ChatGPT default on May 5, cutting hallucinations 52.5% on health queries and matching frontier models (HealthBench 49.6-51.4). User-flagged errors dropped 37.3%, signaling stronger trustworthiness for medical, legal, and financial AI decisions.