🤖AI Summary
Google's Gemini 3.1 Pro Preview achieved a perfect score on IPhO 2025 theory problems across five runs, surpassing previous AI performance that fell behind top human contestants. However, the researchers acknowledge potential data contamination since the model was released after the competition.
Key Takeaways
- →Gemini 3.1 Pro Preview achieved perfect scores on International Physics Olympiad 2025 theory problems in all five test runs.
- →This marks the first time an AI model has surpassed the performance of the best human contestants in the prestigious physics competition.
- →Previous AI models had achieved gold medal performance but still lagged behind top human participants.
- →The results may be compromised by data contamination since the model was released after the competition took place.
- →The achievement demonstrates significant progress in AI's ability to handle complex physics reasoning and problem-solving.
Mentioned in AI
Models
GeminiGoogle
Read Original →via arXiv – CS AI
Act on this with AI
Stay ahead of the market.
Connect your wallet to an AI agent. It reads balances, proposes swaps and bridges across 15 chains — you keep full control of your keys.
Related Articles