hallucinations

OpenAI's new GPT-6 Astra model shows major improvements in hallucination rates and jailbreak resistance over predecessor GPT-5.6 Sol, according to OpenAI's system card. However, indirect prompt injection attacks hidden in documents still succeed 8.5% of the time in external testing by Gray Swan, down from 27% but still above rival Claude Opus 5's 4.8% rate.

OpenAI released GPT-5.5 Instant as the new default ChatGPT model on May 5, 2026. The company claims the update produces 52.5% fewer hallucinations on high-stakes prompts and 37.3% fewer inaccurate claims on challenging conversations compared to GPT-5.3 Instant.

Accepted papers at major AI conferences contain fabricated citations—references to publications that don't exist. A new open-source tool called CiteAudit is the first systematic attempt to detect and eliminate hallucinated references from peer-reviewed research.