British AI companies are pushing back against what one of their own founders called a “Silicon Valley AI doomer bubble”, as pressure mounts on Westminster to legislate before the technology matures.
The intervention comes from Synthesia, the London-based AI video firm, whose UK director Alexander Voica told the ABC that the industry’s central problem is unpredictability. “We know that these systems are getting more powerful, but we don’t know where and how they’re going to be used, and we haven’t figured out essentially a way of taking full advantage of their potential,” he said. “My only concern in rushing to regulate now, where there’s still a lot of open questions, is that it could actually backfire.”
The plea lands in the middle of a genuine panic cycle. OpenAI chief executive Sam Altman has publicly urged Britain to pass binding AI safety laws while the political window is open, Anthropic boss Dario Amodei has called for a managed slowdown of frontier development, and a former Anthropic researcher who warned that the industry is “racing straight to self-improving superintelligence” has been invited to give evidence to parliament. Labour MPs and crossbench peers are circulating drafts of an AI bill.
The industry’s counter-position, and Synthesia is its most prominent UK voice, is that the panic is calibrated to the wrong risk. These companies are not asking for no rules. Their argument is that the frontier-lab doomsday scenario is a theoretical risk being used to justify regulation that will bind practical ones. Their product, an avatar-video platform used by banks, retailers and training departments for explainer videos and onboarding, is not the system anyone is debating extinction over.
Cohere, the Canadian model firm, has made a similar case for regulation based on demonstrated harms rather than rules written by a small circle of leading developers. The split matters for Britain because the government has staked its tech pitch on being the place where AI is built safely, hosting the global safety summit and the institute, while also wanting to be the jurisdiction that does not smother the companies it just courted.
Which is the bind in miniature. If ministers legislate tightly, the Synthesias of the world argue innovation migrates to lighter-touch markets. If they do nothing, the next bad AI headline lands on a minister who passed. Both sides now agree the argument will be settled by whichever story about AI defines the news cycle in October, when the Budget and the next tranche of AI announcements collide.
