UK sets priorities for AI safety summit
But experts fear country lacks a 'coherent approach'
Delegates at the global summit in November will discuss AI risks, measures to increase safety and areas for international collaboration.
The UK government has set out its key priorities for the Global AI Safety Summit, taking place on 1st and 2nd November in Bletchley Park.
Key countries, such as the G7 nations, will send representatives, as will "leading" technology organisations. Academic and civil institutions will also take part.
The summit's goal is to "inform rapid national and international action at the frontier of AI development." It will do so by focusing on the risks and challenges AI poses or exacerbates; for example, the proliferation of access to information that could undermine biosecurity.
As well as risks, summit delegates will focus on how AI can be used for good, in areas like healthcare and public transport.
As part of a consultation process with stakeholders involved in the summit, the UK has shared its five objectives to work on during the event:
- A shared understanding of the risks posed by frontier AI and the need for action;
- A forward process for international collaboration on frontier AI safety, including how best to support national and international frameworks;
- Appropriate measures which individual organisations should take to increase frontier AI safety;
- Areas for potential collaboration on AI safety research, including evaluating model capabilities and the development of new standards to support governance;
- Showcase how ensuring the safe development of AI will enable AI to be used for good globally.
Prime minister Rishi Sunak has previously spoken at length about the UK becoming a "science and technology superpower." Part of that goal is to become a global leader in AI regulation, with the government moving to increase AI investment this year.
However, MPs have warned that the UK risks losing out on its AI ambitions if the government does not introduce new legislation in the King's Speech, also set for November.
Alistair Dent, chief strategy officer at data company Profusion, said the summit represents "a huge opportunity" for the UK to play a role in shaping global AI regulation.
"Unfortunately, this is unlikely to happen as the UK is still struggling to create a coherent approach to regulation... The Government is risking repeating the mistakes it made in its [hands-off] approach to regulating data privacy and social media."
He added, "What is absent from the UK's policy debate is the idea of building an ethical framework around using AI. If we can get consensus on these ethical principles, we can use them as the basis for more flexible regulation that focuses more on preventing undesirable outcomes than it does on individual applications.
"Ideally, these principles would be enforced globally and the AI summit is an excellent place to start this discussion. It's important to remember that no one organisation is going to have all the answers to something as complex and impactful as AI. This is why we need to bring together as many perspectives as possible and spend a lot of time ensuring we get it right.
"My fear is that the government has not fully appreciated how beneficial, but also how damaging, AI could potentially be. The US and EU are already debating their own legislation and the UK really needs to speed up, or it could become irrelevant to the global AI debate."
Computing says:
The summit's goals seem laudable, focusing on understanding the risks and promoting both international collaboration and individual responsibility. But we need to see whether it follows through on those objectives, and if the delegates - especially governments and companies, which are notoriously against being told what to do - commit to any progress that comes out of it with actions, not words.