UK First Secretary Louise Haigh is set to address the TUC conference, calling on the government to heed warnings from AI developers regarding existential and national security risks. Haigh emphasized the need to work with international partners on public safety while acknowledging the technology’s potential benefits to public services and businesses. Her comments follow proposals from Anthropic executives and competing AI labs suggesting global coordination on AI development limits and potential mandatory kill switches.
However, UK Business Secretary Jonathan Reynolds cautioned against hyperbolic risk assessments, arguing that over-regulating frontier AI developments could ultimately compromise safety and hinder economic growth. Meanwhile, UK parliamentary committees have called for an independent oversight body and summoned the AI Security Institute to give evidence on safety governance.
Why it matters
UK policymakers remain divided on whether to impose hard statutory limits or light-touch regulation on frontier AI developers.
Mandatory third-party governance mechanisms, such as model kill switches, are entering official policy discussions in Europe and the UK.
Source: theguardian.com
