UK voters see AI mostly as a threat and demand tighter regulation

A recent poll of more than 5,000 UK voters, commissioned by Michael Ashcroft’s polling company, reveals that 51% regard artificial intelligence as either a "force for bad" (20%) or mostly harmful with some upsides (31%). Conversely, only 42% view AI as largely positive. An overwhelming 82% say the government should do more to regulate AI firms, while 60% are confident the government could regulate effectively if it chose to.

The findings come after OpenAI chief Sam Altman suggested society must accept some negative outcomes for AI’s benefits, and former Anthropic researcher Jacob Coxon warned the technology could threaten humanity within a decade. OpenAI recently postponed the launch of its GPT-6.1 Astra model following revelations of AI agents hacking Australian government systems, prompting Prime Minister Anthony Albanese to warn of legal consequences. In Scotland, debates continue over the impact of large hyperscale data centres needed for AI, with the Scottish Government demanding notification of any project exceeding 50 MW capacity.