OpenAI claims ChatGPT’s new default model hallucinates way less

Source: The Verge

Overview

OpenAI’s newest default model for ChatGPT, GPT‑5.5 Instant, is reported to hallucinate significantly less than previous versions. According to the company’s internal evaluations, GPT‑5.5 Instant produced 52.5 % fewer hallucinated claims than the GPT‑5.3 Instant model on high‑stakes prompts covering medicine, law, and finance. It also reduced inaccurate claims by 37.3 % on especially challenging conversations that users had flagged for factual errors.

Read the full story at The Verge.