OpenAI Launches GPT-5.2 as It Navigates ‘Code Red’

OpenAI introduced GPT-5.2, its smartest AI model yet, with performance gains in writing, coding and reasoning tests. The launch comes just days after CEO Sam Altman internally declared a “code red,” a company-wide initiative aimed at improving ChatGPT amid intense competition from rivals.

“We announced this code red to really signal to the company that we want to mobilize resources in a particular area, and it’s a way of really setting priorities,” Fidji Simo, OpenAI’s applications CEO, said during a briefing with reporters Thursday. “We have seen an increase in resources focused on ChatGPT in general.”

Simo denied that OpenAI had brought forward the launch of GPT-5.2 in light of its code red, saying the company had been working on releasing this model for months. However, she said the additional resources around ChatGPT have been “helpful.”

Although OpenAI’s models and products were considered best-in-class when ChatGPT launched in 2022, this is no longer a settled issue. The startup now faces an array of worthy challengers, perhaps none more threatening than Google, whose recently launched Gemini 3 model has been well received by the tech industry. Google’s Gemini app has seen impressive growth over the last year, now with over 650 million monthly active users, compared to OpenAI’s 800 million weekly active users. This pressure has forced OpenAI to curb some of its most ambitious projects, including its work on introducing ads to ChatGPT, and refocus on improving its core technology and products.

Much like the company’s recent template releases, GPT-5.2 ships in a series of templates: Snapshot, which responds faster and is more efficient for information retrieval; Thinking, who excels at coding, math, and planning; and Pro, the most powerful level of OpenAI models that provides greater accuracy on difficult questions.

OpenAI considers GPT-5.2 its best model yet for everyday professional use. GPT-5.2 Thinking achieved the highest scores to date on GDPval, an OpenAI benchmark that compares performance between AI models and human professionals in 44 real-world professions. The company claims the model beat human professionals at more than 70% of tasks and completed them 11 times faster.

Max Schwarzer, OpenAI’s post-training manager, says the new version should also offer a substantial reduction in hallucinations. The company claims that GPT-5.2 Thinking performed 38% worse than GPT-5.1 on criteria measuring answers to factual questions.

The company offers GPT-5.2 to ChatGPT users and developers on OpenAI’s API product. OpenAI says the new series of models “brings clear gains in everyday and advanced use cases.”