Gemini 3.1 Pro Improves AI Reasoning Across Google’s Ecosystem, Targeting Complex Enterprise and Developer Workflows

Google’s newest Gemini model focuses on deeper reasoning — not just faster answers — as AI competition shifts toward intelligence quality.

Google and OpenAI Models Outperform Humans at ICPC Coding Finals

OpenAI’s GPT-5 achieved a perfect score of 12/12, while Google’s Gemini 2.5 Deep Think solved 10/12, ranking second overall at the 2025 ICPC World Finals.

OpenAI’s BrowseComp Tests AI Browsing Skills on Hard-to-Find Questions

Amazon to Launch ‘Reasoning’ AI Model to Rival OpenAI & Anthropic