The UK says it will establish the world’s first AI safety institute to prevent the technology escaping human control.
The UK’s prime minister has said the country will tackle the "dangers and fears" of artificial intelligence (AI), including making it easier to build biological weapons and spin out of human control, as the UK prepares to host the world’s first AI global safety summit next week.
Rishi Sunak made the comments in a speech in London on Thursday when he also announced he would establish the world’s first AI safety institute in the UK.
He said it would look at "all the risks - from social harms like bias and misinformation, through to the most extreme risks," and that the findings would be shared worldwide.
The British prime minister has continuously spoken about the importance and dangers of AI since he took office last year.
Sunak went on to pledge an additional £100 million (€114 million) of investment into AI for diseases, such as cancer and dementia.
Britain is striving to be a global leader in the technology and carve out its place among the United States, China and the European Union.
AI is ‘one of the greatest tests of leadership we face’
“I believe nothing of our foreseeable future will be more transformative for our economy, our society, and all our lives than this technology. But in this moment, it is also one of the greatest tests of leadership we face”.
The UK will host the global AI safety summit next week (November 1- 2) in southern England at Bletchley Park, which was once the secret base of Britain’s Second World War codebreakers.
Around 100 participants, including political leaders such as US vice president Kamala Harris, AI experts and companies, will gather to discuss how AI should be regulated, its impact on cybersecurity and other threats such as the ability to develop bioweapons and its impact on elections.
"Get this wrong, and AI could make it easier to build chemical or biological weapons," Sunak said.
"And in the most unlikely but extreme cases, there is even the risk that humanity could lose control of AI completely," he added.