OpenAI released new AI models on Thursday even as people wonder how it plans to handle the massive costs of competing with Google.

The San Francisco AI giant claimed its new GPT-5.2 Pro and GPT-5.2 Thinking models are the best versions for handling math and science work.

"Strong mathematical reasoning is a foundation for reliability in scientific and technical work," OpenAI said in a blog post.

"These capabilities are also closely tied to progress toward general intelligence."

Artificial general intelligence is seen as a big goal in the tech world a point where machines can think like humans or even smarter.

The launch follows OpenAI CEO Sam Altman pushing his team to keep pace with Google which has been rapidly advancing in AI.

While Google can use its huge online ad earnings to fund AI OpenAI has been spending tens of billions on computing power even though it hasn’t made a profit yet.

"We are confident we can continue to drive the revenue growth to meet" the investments in computing power Altman said Thursday in a CNBC interview.

Without the infrastructure investments, "of course we can't drive the revenue growth but we see way more reasons to be optimistic than reasons to be pessimistic."

OpenAI’s ChatGPT 5.2 comes just weeks after Google’s Gemini 3 Pro outperformed the chatbot’s previous version in multimodal tasks and complex reasoning. Google’s model had been leading the LMArena leaderboard.

Independent user-driven tests also suggested that Google’s AI was stronger than OpenAI’s 5.1. Many users pointed out that the older ChatGPT version often hallucinated or gave surface level responses instead of staying focused on the topic something Gemini 3 handled more effectively.

With that backdrop OpenAI has now introduced its latest upgrade. Early tests indicate that ChatGPT 5.2 is noticeably more refined and advanced than its predecessor. However, it remains to be seen whether it can surpass Google’s Gemini 3 anytime soon.

OpenAI’s chief of applications Fidji Simo, said during a briefing that a ChatGPT “adult mode” is expected to launch early next year, but only after the company improves its ability to verify users’ ages.

Earlier this year, Sam Altman announced plans to relax restrictions so adult users could have erotic conversations with ChatGPT.

Meanwhile, OpenAI is also dealing with multiple lawsuits from families who claim the company allowed teenagers to have harmful interactions with its chatbots interactions that in some cases allegedly contributed to suicide.