Regulating AI 'not the right place to start' says Bailey
AI needs "rigorous" testing and safeguards to contain risk, Andrew Bailey says.
Regulating AI 'not the right place to start' says Bailey
The Governor of the Bank of England has said regulating artificial intelligence (AI) "is not the right place to start", instead urging "rigorous" testing first to uncover vulnerabilities and put safeguards in place to limit risk.
In his first-ever article for Substack, Andrew Bailey said the dangers surrounding AI were "real and increasingly significant".
Bailey said AI development should not be stopped or banned - "on the contrary, the benefits are immense" - but said there must be a system that allows intervention and sets the boundaries within which AI functions.
Over recent weeks, discussion of the possible risks linked to the fast pace of AI development and what it could mean for humanity has grown more intense.
Leaders of major AI companies including Anthropic and OpenAI have called for the technology's development to slow and for an internationally co-ordinated approach to assessing risks and putting safeguards in place.
OpenAI recently said it would not release its latest AI model because of safety concerns
.
The idea of slowing down has been rejected by figures including President Donald Trump, who has said the US is ahead of China in the AI race, adding: "And, frankly, I want to keep it that way because whoever wins AI, wins."
But on Tuesday, after hosting a summit with leading AI firms such as OpenAI and Anthropic, Trump said executives had signed a "morally binding" document that would act as a "form of protection" against AI's potential risks.
Under the agreement, the companies are responsible for making sure their own technology is safe.
In the UK, Bailey said "important work" was already under way on AI testing - the UK has established the AI Security Institute - which, he said, should help contribute to a set of standards.
"But the pace of progress must accelerate," the bank governor said, though he added: "We should proceed with a degree of humility."
He said that with testing there will be failures and "models will behave unexpectedly".
But, he said: "That is not evidence that testing has failed, rather it is evidence of why testing is necessary."
Bailey said testing should not be viewed as a substitute for future regulation - over time, a more formal regulatory framework may emerge. "But regulation is not, in my view, the right place to start."