Openai and Google surpass the Mathletes but not each other

AI models from Openai and Google Deepmind achieved gold medal scores in 2025 International Math Olympiad (IMO), one of the world’s oldest and most challenging maths competitions in high school, announced companies independently in recent days.

The results emphasize how fast AI systems go ahead, and yet how evenly matched Google and Openai appear to be in the AI race. AI companies compete hard for the public perception of being ahead of the AI race: an intangible battle of “Vibber” that could have major consequences for ensuring Top AI talent. A lot of AI researchers come from backgrounds in competitive mathematics, so benchmarks like IMO mean more than others.

Last year, Google scored a silver medal on the IMO using a “formal” system, which means it required people to translate problems into a machine -readable format. This year, both Openai and Google entered into “informal” systems in the competition that were able to take questions and generate evidence -based answers to natural language. Both companies claim that their AI models correctly answered five out of six questions on IMO’s tests and scored higher than most high school students and Google’s AI model from last year without requiring any translation of people machine.

In interviews with Techcrunch, researchers behind Openai and Google’s IMO efforts claimed that these gold medal performance represents breakthroughs around AI-Reasoning models in non-verifiable domains. While AI resonance models tend to do well with questions with straightforward answers, such as simple math or coding tasks, these systems are fighting on tasks with more ambiguous solutions, such as buying a good chair or helping with complex research.

However, Google raises questions about how Openai performed and announced his gold medal IMO performance. After all, if you get to join AI models in a math competition for colleges, you might as well argue as teens.

Shortly after Openai announced his business on Saturday morning, Google Deepmind’s CEO and researchers took to social media to slam Openai to announce his gold medal too early – shortly after IMO announced which colleges had won the competition on Friday night – and not to have their model test officially evaluated by IMO.

Thang Luong, a Google Deepmind Senior Scientist and head of the IMO project, Techcrunch told Google waiting to advertise his IMO results to respect the students who participated in the competition.

Luong said Google has been working with IMO’s organizers since last year in preparation for the test and wanted the IMO president’s blessing and official classification before announcing his official results, as it did Monday morning.

“The IMO organizers have their classification direction line,” Luong said. “So any evaluation not based on this guideline could not make any requirement for gold medal level [performance]. “

Noam Brown, a senior Openai researcher who worked on the IMO model, told Techcrunch that IMO reached out to Openai a few months ago to participate in a formal math competition, but the chatgpt maker rejected because it worked on natural language systems as it thought was more worthwhile. Brown says Openai did not know that IMO performed an informal test with Google.

Openai says it hired third-party evaluers-three former IMO medals that understood the classification system-to classify its AI model. After Openai was told about his gold medal score, Brown said the company reached the IMO, who then asked the company to wait to advertise until after IMO’s ceremony on Friday night.

IMO did not respond to Techcrunch’s request for comment.

Google is not necessarily wrong here-it went through a more official, rigorous process to achieve its gold medal score-but debate can miss the bigger image: AI models from multiple leading AI-laboratories are improving rapidly. Countries from around the world sent their brightest students to compete at IMO this year, and only a few percent of them scored as well as Openai, and Google’s AI models did.

While Openai used to have a significant lead over the industry, it certainly feels like the race is closer matched than any company would like to admit. Openai is expected to release GPT-5 in the coming months, and the company certainly hopes to give the impression that it is still leading the AI industry.