DeepMind's AI technology, which became famous after beating a human player at the chess-like game Go, has already been put to work for Google, reducing the energy needed for cooling its data centres by 40 per cent last year and increasing efficiency by 15 per cent. And co-founder Mustafa Suleyman outlined last year his hopes that this same technique could be applied to the National Grid and other large scale infrastructure. Read more: Here's how Google's DeepMind is using blockchain-like technology Now that has developed into early-stage talks taking place more recently between DeepMind – named City A.M's most innovative company of the year at the City A.M. Awards – and the National Grid, although there is no guarantee of anything being agreed. “We are in the very early stages of looking at the potential of working with DeepMind and exploring what opportunities they could offer for us," said National Grid.
"One really interesting possibility is whether we could help the National Grid maximise the use of renewables through using machine learning to predict peaks in demand and supply."
“We are always excited to look at how the latest advances in technology can bring improvements in our performance, ensure we are making the best use of renewable energy, and help save money for bill payers.”Energy is the latest area which DeepMind is looking at in the UK after inking several partnerships to work with the NHS over the past year to work on health apps, though largely without the use of AI. It this week unveiled new technology it is working on to allay fears about its access to patient health records. Privacy of data and the reach of a company ultimately owned by Google parent Alphabet has concerned some.
Read more: Three major problems of the modern web keeping Tim Berners-Lee up at night The talks with National Grid are understood to be the first between the AI firm and a FTSE 100 company. Suleyman said at a conference in November that applying the technology used to help Google to such national energy infrastructure "essentially all have the same characteristics".
"All of our algorithms that we develop are inherently general and so given some data set, we should be able to train an algorithm based on some inputs, develop a model, predict some outputs, and then provided we have access to the controls, we should be able to deliver similar sorts of performance."