In a significant moment for artificial intelligence (AI), humans outperformed generative AI models from Google and OpenAI at the 2023 International Mathematical Olympiad (IMO). While both AI systems achieved gold-level scores—35 out of 42 points—they did not secure perfect marks like five young competitors. Google’s Gemini chatbot solved five out of six problems, with IMO president Gregor Dolinar praising its clarity and precision. OpenAI’s experimental model also scored 35 points, marking a significant milestone in AI development. The latest results indicate a notable advancement in AI’s mathematical capabilities, as both models improved markedly from last year’s performance. However, the IMO cautioned that it could not verify the computing power used by AI or any potential human involvement. With predictions of imminent breakthroughs in AI reaching human-level cognitive abilities, industry leaders emphasize the rapid pace of innovation, raising intriguing questions about the future of AI and its implications for human expertise.
Source link
Share
Read more