October 27, 2023
The UK government is pushing for a global response to the risks posed by AI at its upcoming AI Safety Summit.
Controversially, that response includes China, which has been cast by the West as its adversary when it comes to matters of cyber defence and security, espionage, and technology development – including AI.
“We’ve invited China,” said UK prime minister Rishi Sunak, in a speech at The Royal Society on Thursday.
“I know there are some who will say they [China] should have been excluded. But there can be no serious strategy for AI without at least trying to engage all of the world’s leading AI powers,” he said. “That might not have been the easy thing to do, but it was the right thing to do.”
At the AI Safety Summit, which takes place next week at Bletchley Park – the birthplace of hacking into encrypted comms as far as the UK goes – Sunak hopes to broker the first ever globally-agreed statement on the nature of the risks posed by AI.
In doing so, his hope is for the UK to lead a coordinated, international response to those risks, ensuring AI technology benefits humanity rather than jeopardises it.
Interestingly, Sunak said the world cannot rely on AI developers themselves to self-regulate, or “mark their own homework,” as he put it.
However, it is worth noting that the Frontier Model Forum – an AI safety body established in July by Google, Microsoft and ethical AI developer Anthropic – is attempting to do just that.
This week, it appointed its first executive director in Chris Meserole, who most recently served as director of the artificial intelligence and emerging technology initiative at US think-tank Brookings.
In his new role, Meserole will help the Frontier Model Forum advance AI safety research to promote responsible development; identify best practices; share knowledge with policy makers, academics, and other stakeholders; and support efforts to leverage AI to address major societal challenges.
The Frontier Model Forum has also chipped in an initial $10 million to establish the AI Safety Fund, which will support independent academic research into AI safety.
All this self-regulation isn’t going to dissuade Sunak from doing some regulating of his own though.
Alongside his speech, the government published its first AI discussion paper. Based on various sources – including intelligence assessments – it goes into the capabilities and risks of what it describes as ‘frontier AI’.
It covers the current state of frontier AI, including how it might improve in the future, as well as the risks it currently presents – like societal harms, misuse and loss of control. It then posits how the situation might evolve between now and 2025, before looking further ahead at potential scenarios in 2030.
“Get this wrong, and AI could make it easier to build chemical or biological weapons. Terrorist groups could use AI to spread fear and destruction on an even greater scale. Criminals could exploit AI for cyber-attacks, disinformation, fraud, or even child sexual abuse,” warned Sunak. “And in the most unlikely but extreme cases, there is even the risk that humanity could lose control of AI completely.”
“I don’t want to be alarmist,” he insisted. “And there is a real debate about this – some experts think it will never happen at all. But however uncertain and unlikely these risks are, if they did manifest themselves, the consequences would be incredibly serious.”
While most people would agree that it’s worth taking seriously the potential pitfalls of unchecked AI development, not everyone agrees that China should have a hand in the West’s strategy for tackling them.
Liz Truss, who spent an eventful month as prime minister this time last year, wrote to Sunak, urging him to rescind China’s invitation.
“The regime in Beijing has a fundamentally different attitude to the West about AI, seeing it as a means of state control and a tool for national security,” Truss wrote. “These and other concerns were a significant factor in the decision by the now deputy prime minister [Oliver Dowden] in 2020, when he was digital secretary, to remove all Huawei equipment from the UK’s 5G networks. That was the correct call then and should have informed decisions about invitations to the Bletchley Park summit.”
On the one hand, inviting China gives Sunak a chance to play the mature statesman, who urges everyone to put aside their differences to pursue a common goal. But as the Truss letter demonstrates, hawks will seize upon the opportunity to portray Sunak as naïve for thinking Beijing will do anything other than undermine Western efforts to control the AI genie.
About the Author(s)
You May Also Like