DeepMind | Artificial Intelligence

Artificial intelligence may be one of the most useful inventions for humans

We are a team composed of scientists, engineers, ethicists, and others dedicated to solving intelligent problems, promoting scientific progress, and benefiting humanity.

We have always been fascinated by human intelligence - it has shaped the modern world we live in today.

Intelligence enables us to learn, imagine, collaborate, create, communicate, and more. By better understanding the different aspects of intelligence, we can use this knowledge as inspiration to build novel computer systems that learn to find their own solutions to problems.

Just like the Hubble Telescope, which has helped us delve into space, these tools are already expanding human knowledge and having a positive global impact. Our long-term goal is to solve intelligent problems and develop more universal and capable problem-solving systems, known as General Artificial Intelligence (AGI).

Guided by safety and ethics, this invention can help society find answers to some of the world's most urgent and fundamental scientific challenges.

When we founded DeepMind in 2010, people's interest in the AI field was far less than it is today.

To accelerate the development of this field, we have adopted an interdisciplinary approach that combines new ideas and advancements in machine learning, neuroscience, engineering, mathematics, simulation, and computing infrastructure, as well as new methods for organizing scientific work.

We have achieved early success in computer games, and researchers often use it to test artificial intelligence. Our program learns to play 49 different Atari games from scratch, simply by looking at the pixels and scores on the screen. Our AlphaGo program was also the first project to defeat a professional Go player, a feat described as leading the era by ten years.

In 2014, we teamed up with Google to accelerate our work while continuing to develop our own research agenda. Our project has learned to diagnose eye diseases as effectively as the world's top doctors, saving 30% of energy used to keep data centers cool, and predicting complex 3D shapes of proteins - which may one day change the way drugs are invented.