OpenAI has launched GPT-5.2, its most advanced AI model, enhancing writing, coding, and reasoning capabilities. This release follows CEO Sam Altman’s internal “code red” initiative aimed at prioritizing improvements to ChatGPT amid competition from rivals like Google’s Gemini 3, which boasts over 650 million monthly users. OpenAI’s VP Fidji Simo emphasized that the launch timing was not influenced by the code red but that increased resources directed towards ChatGPT have been beneficial. GPT-5.2 features three tiers: Instant for speed, Thinking for coding and planning, and Pro, delivering maximum accuracy. It excels in real-world tasks, outperforming human professionals in over 70% of cases and completing tasks 11 times faster. The model also significantly reduces hallucinations, reporting 38% less compared to GPT-5.1. OpenAI aims for GPT-5.2 to provide substantial improvements for both everyday and advanced users through ChatGPT and its API offerings.
Source link
