
Google DeepMind's Gemini Deep Think AI system achieved gold-medal performance at the 2025 International Mathematical Olympiad, scoring 35 out of 42 points by solving five of six problems perfectly. This marks a significant advancement in AI's reasoning capabilities, as the system operated entirely in natural language to produce mathematical proofs, unlike prior specialized language requirements. The breakthrough, utilizing 'parallel thinking' and novel reinforcement learning, underscores rapid progress in AI's ability to tackle complex, abstract problems, a trend reinforced by OpenAI's concurrent achievement of similar gold-medal success. Google plans to make the model available to trusted testers and AI Ultra subscribers, indicating its potential for broader application.
Google (GOOGL) has demonstrated a significant advancement in artificial intelligence by achieving a gold-medal standard at the 2025 International Mathematical Olympiad with its Gemini Deep Think AI. The system's score of 35 out of 42 points, a marked improvement from its silver-medal performance last year, was achieved by operating entirely in natural language to solve complex mathematical problems—a key breakthrough that removes the need for specialized coding. This progress in advanced reasoning is attributed to novel techniques like 'parallel thinking' and reinforcement learning. However, the competitive landscape remains intense, as OpenAI simultaneously reported an identical gold-medal achievement with its own model. This parallel success underscores that rapid innovation is an industry-wide phenomenon, not a unique moat for Google. The company's plan to integrate this technology into its premium Google AI Ultra subscription signals a clear path toward monetizing these research milestones, directly linking advanced AI capabilities to its product ecosystem.
AI-powered research, real-time alerts, and portfolio analytics for institutional investors.
Request a DemoOverall Sentiment
strongly positive
Sentiment Score
0.75
Ticker Sentiment