Rishi Sunak will announce the next steps in Britain’s ambition to regulate artificial intelligence when he visits Washington next week, i understands.
The Government is seeking to be a global leader in AI regulation following growing concerns over the destructive potential of the fast-developing technology.
But the Prime Minister also wants to help Britain share the economic and social benefits which many believe will spring from AI and is seeking to promote the UK as a destination for tech investment.
Mr Sunak will continue his regulatory balancing act when he meets Joe Biden next week. He has already promised to raise AI in his conversations with the President, and it is understood he will provide more information about his plans for further regulation.
The Government will use London Tech Week in mid-June to make its case for the UK as an AI hub, according to Whitehall sources.
In the coming weeks it is also expected to reveal the identity of an expert taskforce which has secured £100m in funding to explore the possibility of a British rival to “foundation models” like ChatGPT which synthesise enormous volumes of data in order to train themselves.
Over the next year regulators in various sectors will be expected to lay out guidance to companies on how to make the most of AI without exacerbating its potential dangers. The Cabinet Office is considering whether to add the technology to its “national risk register” which records the primary threats to UK society and imposes obligations to plan for worst-case scenarios involving them.
In the spring the new Department for Science, Innovation and Technology published a detailed white paper on AI which focussed primarily on making the most of its benefits. Since then Mr Sunak has had a partial change of heart, choosing to place more emphasis on the risks.
This week a coalition of leading researchers and entrepreneurs working on AI published a warning that the risk of human extinction as a result of runaway artificial super-intelligence must be “a global priority alongside other societal-scale risks such as pandemics and nuclear war”.
Some other experts have dismissed that claim but all major Western allies have committed to working together to mitigate the risks and enforce a global set of norms which will stop AI from being developed in an uncontrolled way.