Stay informed with free updates
Simply sign up to the Artificial intelligence myFT Digest — delivered directly to your inbox.
Rishi Sunak said he would not “rush to regulate” artificial intelligence as he announced the creation of an AI safety body in the UK to evaluate and test new technologies.
Speaking ahead of hosting a summit next week to look at the threats posed by the fast-evolving technology, the prime minister said the UK would “push hard” for a “first ever” joint statement by global leaders about the nature of the risks it poses.
But that did not mean there should be a “rush to regulate” the sector. “This is a point of principle, we believe in innovation,” he said, adding: “How can we write laws that make sense for something that we don’t yet fully understand?”
He said the new institute would “advance the world’s knowledge of AI safety . . . so that we understand what each new model is capable of, exploring all the risks from social harms like bias and misinformation through to the most extreme risks”. He said its work would be available to everyone.
He would also propose the creation of a global panel of experts, nominated by the countries and organisations in attendance at the summit, to publish a report on the “state of AI”, modelled on the UN’s Intergovernmental Panel on Climate Change.
To coincide with Sunak’s speech the government released a series of papers outlining potential risks posed by the rapid rollout of AI.
One such scenario outlined how automation could lead to increased unemployment and poverty by 2030, amid “fierce public debate about the future of education and work”.
Other scenarios included the emergence of open-source models capable of competing with humans on a wide range of tasks “with startling autonomy and agency”, and which can be used to instigate malicious attacks.
Speaking at the Royal Society, the UK’s academy of sciences, in London, Sunak warned that if we “get this wrong” the fast-evolving technology could make it easier to build chemical and biological weapons, perpetrate cyber attacks, spread disinformation and commit child sexual abuse. In the most extreme cases “humanity could lose control of AI completely,” he warned.
Sunak sought to strike a reassuring note, adding that “this is not a risk people need to be losing sleep over right now” in large part because “the UK is doing far more than any other country to keep you safe”.
“Even if there is uncertainty the risks could be significant so it’s right that we take all the measures possible to protect people,” he said. “The British people should have peace of mind.”
One of the accompanying papers further sought to dispel fears the technology could run amok: “Many experts see this as highly unlikely” as the AI would need to be able to gain control and “be able to avoid being switched off,” it said.
Sunak also outlined some of the potential benefits of AI across health and social care, enabling specialists to “solve some of the greatest challenges of our time”, including world hunger.
The government is hoping to host a number of world leaders at next week’s summit but there are concerns attendance will be affected by the war between Israel and Hamas.
Ursula von der Leyen, head of the European Commission, is still scheduled to come to the summit, along with US vice-president Kamala Harris and commerce secretary Gina Raimondo, as well as representatives from China.
Read the full article here