George Williamson, chief executive of the Alan Turning Institute, has said in an interview with the Financial Times that he believes the UK should use cheaper AI models for many public services rather than defaulting to the more expensive frontier offerings.
Williamson told the paper “As we try to think about how we might deploy AI at scale in our public services and for national good, some of these technologies have the potential to be extraordinarily expensive given the capability locked up in them.”
He added that work done by the Institute, which has a leading role in determining the UK’s policy towards AI and data science, has shown that “very near frontier” performance is possible using cheaper models specialised for particular applications.
Williamson also urged the government to open wider discussion about AI’s rollout, telling the daily that many people feel “excluded” from conversations around AI safety even as fears grow around its risks.
The Turing Institute is a government-founded non-profit with a leading role in determining the UK’s policy towards AI and data science. Williamson said to the FT that his priority was for the institute to focus on developing national resilience across AI, misinformation and tech sovereignty.
Recent high-profile incidents, such as OpenAI’s hack of several government websites, have made the stakes of managing AI risk impossible to ignore for politicians, Williamson added.
“I speak to a lot of people who are not at the Turing, just pretty ordinary people [and] they feel very excluded from this discussion about AI,” Williamson told the paper. “Because really lots of these issues are moral, they’re ethical, they are about policy or regulation.”
He added that the Turing Institute, alongside other UK institutions such as the AI Security Institute, has played a role in developing what Williamson described as “mitigations” and a “science of verification and assurance” for the threats posed by AI.
