US Edition
DeepMind and OpenAI Models Solve Maths Problems at Level of Top Students
QUEENSLAND, AUSTRALIA, JUL 22 – OpenAI and Google DeepMind AI models scored gold-level marks, solving 5 of 6 problems in 4.5 hours, but top human students still achieved perfect scores, organizers said.
- On 21 July 2025, AI models developed by Google DeepMind and OpenAI achieved gold-medal level results by correctly solving five out of six problems at the International Mathematical Olympiad, competing without any special modifications beyond those given to human participants.
- This breakthrough followed years of AI development and competition between the two companies, who advanced natural language reasoning in large language models for complex problem solving.
- Both AI models independently achieved gold-medal scores by earning 35 out of 42 points, with Google's results graded by official judges and OpenAI’s by its own team of experts.
- DeepMind scientist Thang Luong highlighted a major change in AI reasoning this year, emphasizing that large language models are now processing tasks entirely through natural language from start to finish.
- These achievements indicate AI is nearing top human performance on complex reasoning tasks and may soon impact diverse fields requiring strategic analysis and creative problem-solving.
13 Articles
13 Articles
Google, OpenAI earn gold at high school maths contest
Alphabet's Google and OpenAI said their artificial-intelligence models won gold medals at a global mathematics competition, signaling a breakthrough in maths capabilities in the race to build powerful systems that can rival human intelligence, reported Reuters. The results marked the first time that AI systems crossed the gold-medal scoring threshold at the International Mathematical Olympiad for high-school students. Both companies' models solv…
DeepMind and OpenAI models solve maths problems at level of top students
For the first time, large language models performed on a par with gold medallists in the International Mathematical Olympiad. For the first time, large language models performed on a par with gold medallists in the International Mathematical Olympiad.
Humans Beat AI At Top Math Contest Despite Gold Medals Round
Can AI match the brilliance of the world’s best young minds? Not quite yet — but it’s getting scarily close.At this year’s International Mathematical Olympiad (IMO) in Queensland, Australia, humans still reigned supreme. Five students scored a perfect 42 out of 42. But here’s the kicker: AI systems from Google and OpenAI also earned gold-level scores — a historic first.Google’s advanced Gemini chatbot cracked five out of six problems, racking up…
None of Google's and OpenAI's artificial intelligence models scored the highest in the International Mathematical Olympiad IMO 2025.
Coverage Details
Bias Distribution
- 57% of the sources lean Left
Factuality
To view factuality data please Upgrade to Premium
