Google DeepMind was formed in April 2023 by merging two of the world’s top AI research groups: DeepMind (founded in London in 2010 by Demis Hassabis, Shane Legg, and Mustafa Suleyman, acquired by Google in 2014 for around $500 million) and Google Brain (Google’s internal AI research division). Demis Hassabis leads the combined organization.
DeepMind made global headlines in 2016 when AlphaGo defeated world Go champion Lee Sedol — a feat many experts thought was a decade away. They followed that with AlphaFold, which solved the decades-old protein folding problem and made its predictions freely available for 200 million proteins. That work won the 2024 Nobel Prize in Chemistry for Hassabis and John Jumper.
As Google DeepMind, the lab develops the Gemini family of multimodal AI models that power products across Google’s ecosystem, including Search, Workspace, and Android. Gemini Ultra, Pro, and Nano serve different use cases from cloud APIs to on-device applications.
The research output is staggering — Google DeepMind publishes hundreds of papers per year across areas like reinforcement learning, neuroscience-inspired AI, climate modeling, and mathematics. Their work on AI for weather forecasting (GraphCast) and materials discovery has demonstrated real-world scientific impact. With thousands of researchers in London, Mountain View, and other offices worldwide, Google DeepMind is one of the largest and most productive AI labs on Earth.
