In an unprecedented milestone, AI systems developed by Google’s DeepMind and OpenAI have outperformed expectations by securing gold medal-level scores at the International Mathematical Olympiad (IMO), a competition historically reserved for the world’s brightest high school mathematicians.
For the first time ever, these AI models solved five of the six notoriously difficult problems using natural language reasoning, mimicking human thought processes rather than relying on rigid programming languages or brute computational force.
Google’s model, Gemini Deep Think, completed all tasks within the official exam time, while OpenAI’s experimental AI massively scaled computing power to “think” deeply and explore multiple solution paths simultaneously.
This approach, though “very expensive,” allowed the AI to demonstrate advanced problem-solving that many experts believe could soon assist mathematicians in cracking unsolved puzzles.
Junehyuk Jung, a Brown University math professor involved with Google DeepMind, said this breakthrough “opens the door for real collaboration between AI and human mathematicians,” marking a new era where artificial intelligence is no longer just a tool but a partner in discovery.
While only 11% of the 630 student competitors earned gold medals, these AI systems matched that elite performance level, signaling that artificial intelligence may soon revolutionize scientific research far beyond math.
With official results set to be published soon, this achievement has ignited excitement and debate about how AI could reshape education, research, and problem-solving worldwide.
No comments:
Post a Comment