Some of the more apocalyptic warnings about the dangers posed by artificial intelligence have come from those who stand to profit most from its breakneck development. Executives from tech giant Google and OpenAI, creator of ChatGPT, are among those who have recently raised the potential societal threats posed by intelligent machines. Demis Hassabis, who heads Google’s UK-based Deepmind AI unit, compared the challenge to that of climate change.
Apple co-founder Steve Wozniak and X’s Elon Musk meanwhile signed an open letter earlier this year in which they and other tech executives warned that AI tools present “profound risks to society and humanity” and called for a pause in their development. Less surprisingly, most sector leaders appear to agree with the OpenAI boss Sam Altman that, if suitably regulated, the benefits of AI will outweigh the risks. This same mix of doomsterism and boosterism characterises this week’s AI Safety Summit at Bletchley Park, once home to Britain’s wartime codebreakers, at which Prime Minister Rishi Sunak is seeking to put the UK at the centre of the AI debate.
Sunak echoed warnings that AI could be abused by criminals or terrorist groups
Ahead of the two-day meeting, Sunak echoed warnings that AI could be abused by criminals or terrorist groups, while the risk of cyberattacks, disinformation and fraud posed a real threat to society. “Some experts think there is even the risk that humanity could lose control of AI completely, through the kind of AI sometimes referred to as ‘super intelligence’.” Given this dystopian vision, his addendum - “We should not be alarmist about this… and some experts think it will never happen” - was scarcely reassuring.
Among Sunak’s self-declared summit objectives are to create a framework for international collaboration on AI safety and to develop shared standards to support governance of the developing technology. Another underlying objective is that the UK should emerge as a leader in the field, both as a developer and a regulator. A government paper boasted this year that the UK was a world leader in AI, with a national sector that already employs more than 50,000 people and contributed £3.7 billion to the national economy last year. That still leaves the UK as something of an AI minnow when compared with the tech giants in the US and China that currently dominate the sector.
The expectations for the Bletchley summit are perhaps more modest than the government might have hoped for
The expectations for the Bletchley summit, the first such international gathering on AI safety, are perhaps more modest than the government might have hoped for. Apart from Sunak, heads of government will be thin on the ground. The US, currently the world’s AI superpower, will be represented by Vice President Kamala Harris. A decision to invite China has been largely supported by politicians and experts in the tech sector who would like to see the US’s main AI rival inside the regulatory tent.
Sunak’s view that no serious strategy on managing AI risk can be achieved without China’s input has, however, been criticised by some in his own Conservative party. Politicians aside, international tech bosses figure prominently in the 100-plus summit guest list, no doubt eager that their own interests should be well-represented in the AI safety debate. “It’s like asking the tobacco industry to regulate smoking,” according to Dame Wendy Hall, a leading British computer scientist and co-author of a government-commissioned independent AI review. Hall told the BBC that the big companies were competitors whose aim was profit, “so I don’t think they should be running the show”.
She lamented what she saw as the lack of diversity in the Bletchley forum and stressed the need to “keep the human in the loop”.
Representatives of the public whose lives will be most affected by advancing AI technology are notable for their absence
Despite the UK sponsors’ nod to civil society, representatives of the public whose lives will be most affected by advancing AI technology are notable for their absence. Perhaps it is more convenient for tech bosses and governments to focus on the more extreme risks posed by AI, from cyber-terrorism to societal collapse, than to address the seemingly more mundane concerns of ordinary citizens. The AI narrative has tended to focus on the promise of more tech-related jobs rather than on the inevitable loss of a swathe of jobs in currently human-operated sectors.
A public increasingly wanting an answer to diminishing access to speedy health care is being offered instead a golden future of improved AI-assisted diagnosis. In an open letter to Sunak, shared with the Financial Times on the eve of the Bletchley gathering, more than 100 individuals and organisations said that “communities and workers most affected by AI have been marginalised by the summit”. In it, bodies that included the UK’s Trades Union Congress (TUC), the International Trade Union Confederation, Amnesty International and Liberty, asserted: “For many millions of people in the UK and across the world, the risks and harms of AI are not distant - they are felt in the here and now.” The letter underlines fears that AI is already underming job security, assisting rampant fraud and disinformation and curbing civil liberties by fostering a surveillance culture.
The signatories might have added that the belatedly regulated explosion of the internet and social media in the last quarter century serves as an object lesson. A revolutionary technological innovation that could have brought people together has been marked by the commodification of users and their personal information at the service of tech monopolies and their investors. Consumers have been transformed into the raw material of profit. That lesson should not be lost amid the sector-driven rhetoric about how to stop robots taking over the world. It is a debate in which the wider public should be allowed to have its say. As Kate Bell, TUC assistant general secretary, put it: “It shouldn’t just be tech bros and politicians who get to shape the future of AI.”
