Sunak Says DeepMind to Offer UK Priority Research Access

Prime Minister Rishi Sunak said part of his plan to make the UK a center of AI research and regulation was to ensure Britain got a first look at insights from labs such as Google DeepMind, OpenAI and Anthropic.

(Bloomberg) — Prime Minister Rishi Sunak said part of his plan to make the UK a center of AI research and regulation was to ensure Britain got a first look at insights from labs such as Google DeepMind, OpenAI and Anthropic. 

“I’m pleased to announce they’ve committed to give early or priority access to models for research and safety purposes to help build better evaluations and help us better understand the opportunities and risks of these systems,” Sunak told the London Tech Week conference on Monday.

He also described an AI taskforce modeled on the one that steered the UK’s procurement and rollout of the Covid-19 vaccine, during the talk with Demis Hassabis, founder and head of Alphabet Inc.’s Google DeepMind AI division.

The prime minister’s AI push comes as part of an effort to dispel doubts about the UK’s attractiveness as a place to do business, especially in technology, after the decision by Cambridge-based chip design giant Arm Ltd. to list in New York instead of London. 

Read more: Sunak Seeks Biden’s Backing on AI After UK Left Out of Key Talks

Last month, the prime minister met with leading AI bosses, such as OpenAI Chief Executive Officer Sam Altman, and earlier in June announced the UK would hold a global summit on AI safety later this year. He said guardrails were important because of concerns around people’s jobs and misuse of the technology.

“Right now, there is an opportunity for human progress that could surpass the industrial revolution in both speed and breadth,” Sunak said.

Hassabis spoke to Bloomberg Television following his on-stage discussion with the UK prime minister, and said investment into researching effective safeguards for AI should be a priority.

“The number one thing that needs to be done right now is to put more investment into AI safety research and understanding what these systems can do, analyzing them, interpreting them, and then coming up with things like evaluation benchmarks so that we can understand what capabilities we want,” he said.

–With assistance from Tom Mackenzie.

More stories like this are available on bloomberg.com

©2023 Bloomberg L.P.