Core Viewpoint - OpenAI's experimental large language model achieved a gold medal level score of 35/42 in the 2025 International Mathematical Olympiad (IMO), but the announcement was controversial due to timing and adherence to IMO's guidelines [2][20]. Group 1: OpenAI's Announcement and Controversy - OpenAI announced its model's IMO results shortly before the closing ceremony, which was seen as disrespectful to human participants [6][8]. - The IMO committee had requested AI companies to wait a week after the closing ceremony to announce results to respect student competitors [4][16]. - OpenAI's early announcement led to criticism from IMO officials, who felt it detracted from the achievements of human contestants [7][11]. Group 2: Comparison with Google DeepMind - Google DeepMind also achieved a gold medal in the same competition but chose to remain low-key about the results, adhering to IMO's request [3][12]. - DeepMind's approach was viewed as more respectful, as they waited for the official announcement period [12][14]. Group 3: Scoring and Validation Issues - Joseph Myers, an IMO gold medalist, indicated that OpenAI did not collaborate with IMO for testing, and their solutions were not rated by official coordinators [11][15]. - There are concerns that OpenAI's score could be downgraded to silver if any points were deducted during evaluation [18][19]. - The validity of OpenAI's gold medal claim is questioned, as the scoring criteria were not publicly disclosed [15][17]. Group 4: OpenAI's Response - OpenAI's researcher Noam Brown stated that they believed they followed the appropriate timeline for announcing results and were not informed of the one-week waiting period [20][21]. - Brown also mentioned that OpenAI declined an offer from IMO to provide problems in a machine-verifiable format, which raises further questions about their participation [24][25].