IMO rebukes OpenAI for claiming victory on its own, stating that "none of the 91 judges participated in the scoring." Netizens: There's no bottom line for hype.

Less than 24 hours after OpenAI claimed its new model won a gold medal at the IMO, the plot took a dramatic turn!

Multiple official IMO representatives and academic authorities have spoken out, accusing OpenAI of acting "rudely and inappropriately."

The IMO organizers requested that AI companies wait until one week after the closing ceremony to announce the results, so that the focus would remain on the participating teenagers. However, OpenAI couldn't wait and announced the results right after the closing ceremony.

Some netizens commented: "OpenAI will do anything for hype, as always. They have no official scores, no patience, and no sense of shame."

Even more shocking is that OpenAI's so - called "gold medal" result may not hold water at all:

OpenAI is not one of the AI companies that cooperated with the IMO to test the model, and none of the 91 official IMO judges participated in evaluating their answers. This means that OpenAI's "gold medal" claim is self - proclaimed and has not been officially certified.

What was supposed to be a milestone moment for AI development has unexpectedly sparked a fierce debate about academic ethics and commercial hype.

The IMO officials are angry: "Please leave some space for the children."

The trigger for this incident was a revelation from a senior IMO figure.

Joseph Myers, a veteran who has been involved in the Math Olympiad since 1992, had a conversation with Ria van Huffel, the Secretary - General of the IMO.

The IMO jury and coordinators generally believe that it is "rude and inappropriate" for AI developers to announce results during the IMO, especially before the closing ceremony. The official expectation was that AI companies should wait at least one week after the closing ceremony to release the results.

On OpenAI's side, Noam Brown, who was involved in this research, responded.

His statement acknowledged that OpenAI did not contact the IMO officials in advance. They only informed one organizer before announcing the results, and the organizer asked them to announce the results after the closing ceremony.

This contradicts the requirement in the revelation, which was one week after the closing ceremony.

So, what exactly happened?

An announcement from another participating AI team, Harmonic, confirmed that there was indeed a requirement of "one week after the closing ceremony," and specified the deadline as after July 28.

While OpenAI was making a high - profile announcement of its results, another AI giant, Google DeepMind, showed restraint, in sharp contrast to OpenAI's flamboyance.

Multiple insiders revealed that DeepMind may also have achieved a gold - medal - level result, but they chose to abide by the IMO's requirements and patiently wait for the right time.

Besides the issue of the announcement time, there is also a controversy over whether OpenAI actually won a gold medal.

Thang Luong, who leads DeepMind's super - reasoning team, added that there is actually an official scoring guide within the IMO that is not accessible to the outside world. Without a score based on this guide, there is no qualification to receive a medal.

In this IMO, there were 6 questions, each worth 7 points. The gold - medal threshold was 35 points, and OpenAI's self - reported score just barely passed the line. Even a small deduction in the solution process could drop OpenAI from a gold medal to a silver medal.

In short, the battle for the IMO gold medal is far from over, and everything should be based on the results certified by the IMO organizers.

One More Thing

Finally, let's shine the spotlight on the human gold - medalists.

In the final results, the Chinese team topped the list with all 6 members winning gold medals and a total score of 231 points. After losing to their old rival, the US team, by 2 points last year, they regained the world - number - one position.

The second place went to the US team, which won 5 gold medals and 1 silver medal, followed by South Korea, Japan, and Poland.

The six members of the Chinese team are Deng Zhewen from Wuchang Experimental High School, Xu Qiming and Tan Hongyi from Wuhan Economic Development Zone Foreign Languages High School, Zhang Hengye from Chongqing Bashu Middle School, Dong Zhenyu from Hangzhou Xuejun High School, and Deng Leyan from Shanghai High School.

Both Deng Zhewen and Xu Qiming won gold medals at the 2024 IMO. Xu Qiming was also successfully selected for Peking University's Mathematics Talent Class in 2025.

Dong Zhenyu is the only senior - three student in the team. Although he entered the national training team as early as his freshman year in high school and was guaranteed admission to Tsinghua University's Yao Class, he missed out on the IMO national team in the previous two selections.

This was his third attempt to make the national team, and he finally achieved his dream of participating in the IMO.

Tan Hongyi was also eligible for direct admission to Tsinghua or Peking University after being selected for the national training team for two consecutive years with a gold medal in the China Mathematical Olympiad (CMO).

In addition, Deng Leyan and Zhang Hengye scored a perfect 42 points in this IMO, tying for the world - number - one position. This is also the seventh consecutive IMO since 2019 that a Chinese team member has achieved a perfect score.

Among them, Deng Leyan is the youngest member of the Chinese team. As a freshman in high school, he was selected for the national training team with a perfect score in last year's CMO and stood out in the IMO national - team selection, breaking the record for the youngest IMO participant in Shanghai in the past 10 years.

Last year, he also won the fifth place globally in the combinatorics and probability track of the Alibaba Global Mathematics Competition, becoming the first middle - school student to win a medal in this event.

Zhang Hengye won a gold medal in the 2023 CMO and was selected for the national team. He also won an award in the Alibaba Global Mathematics Competition. After being selected for the national team last year, he voluntarily went to Peking University's Mathematics Department to audit courses and benefited a lot.

It should be noted that among the more than 600 participants in this IMO, only 6 achieved a perfect score. Almost everyone else failed on the sixth question - combinatorial mathematics, which was also the only question that OpenAI failed to solve.

Additionally, the next IMO will be held at Shanghai High School. Shanghai High School has won 18 IMO gold medals to date, ranking first in the country in total. It will also become the first middle school in the world to host the International Mathematical Olympiad.

It's worth noting that among the top 30 participants in this IMO, almost all had Asian or Eastern European faces.

The runner - up US team also consisted entirely of Asian - looking members.

Netizens said that perhaps future math competitions will be a battle between the Chinese team, the US team, and artificial intelligence.

Reference links:

[1]https://x.com/HarmonicMath/status/1947023450578763991

[2]https://www.imo - official.org/

[3]https://x.com/ai_for_success/status/1946984782178709719[4]https://x.com/lmthang/status/1946960256439058844

This article is from the WeChat official account "QbitAI". The authors are Meng Chen and Lu Yu. It is published by 36Kr with permission.