The UK government’s AI Safety Summit, which claims to be a global gathering to discuss frontier AI, has faced criticism for its lack of diverse perspectives among its delegates. Critics argue that the summit is more of a staged photo opportunity rather than an open forum for debate.
Narrow Range of Voices
The government has not publicly disclosed the list of delegates, leading campaigners to question the lack of diversity in attendance. Mark Lee from the University of Birmingham expresses the need for an open debate with input from various perspectives, including legal, ethical, and technological viewpoints, rather than just powerful figures from companies.
Lee further argues that the summit is overly focused on hypothetical existential risks, such as “robots with guns,” instead of addressing real and pressing issues related to biased decision-making by AI systems in areas like medical diagnosis, criminal justice, finance, and job applications. Including a wider range of voices with diverse backgrounds could lead to more practical discussions on regulation.
Calls for Diversity
Over 100 trade unions, charities, and other groups have signed an open letter expressing concerns about the lack of diversity at the summit. They raise the issue that communities and workers most affected by AI have been marginalized by the event.
The campaign group PauseAI, which focuses on highlighting potential existential risks of future AI models, also staged a protest outside Bletchley Park, where the summit took place. They argue that the attending companies are in a competitive race to develop powerful AI without considering the potential dangers. PauseAI believes that only a few individuals at the summit will advocate for measures to slow down development and prevent the creation of super-intelligent AI.
Joep Meindertsma of PauseAI states, “Polls show that the public wants to slow down AI development, while AI company CEOs and government representatives want to accelerate. There are probably only a handful of individuals at the summit who will be pushing for concrete measures to slow down. No company should be allowed to build a super-intelligence.”
Limited Access for Journalists
Reporters attending the summit have also faced restrictions. Some were not allowed entry, while others were ushered into a high-security media center and told they could not leave without a pre-arranged interview with a delegate. Requesting a list of delegates resulted in the information being withheld.
A journalist from another AI-focused outlet, who preferred to remain anonymous, shared their experience of being refused entry. They expressed frustration with the difficulty of obtaining interviews, access, and updates, with most emails going unanswered.
In response to the criticism, a government spokesperson defended the summit, stating that it has brought together a wide array of attendees from international governments, academia, industry, and civil society. The spokesperson emphasized the collaborative approach taken to ensure safe and responsible development of AI.
