Anthropic chief executive Dario Amodei has called for the pace of AI development to be slowed, a call echoed by rivals Sam Altman of OpenAI and Elon Musk, while a former Anthropic researcher told the BBC that staff building these systems are "genuinely frightened" about humanity's future. The warnings mark a notable shift from the sci-fi framing of existential AI risk at the 2023 Bletchley Park summit towards a more urgent tone from senior industry figures themselves, yet critics argue that calls for a "slowdown" remain vague and impractical to implement.

Amodei has proposed a three-point plan involving independent monitoring of AI models, industry-wide regulation and global co-ordination, but sceptics such as Ed Zitron of EZ Primary Research say nobody has clearly explained what a slowdown would actually mean in practice, or how it could be policed given the trust the tech sector has yet to earn. Complicating matters further is fierce US-China competition, with President Trump asserting American AI leadership and declaring "whoever wins AI, wins", making unilateral restraint unappealing to firms fearful of falling behind. In the UK, where AI is central to economic growth plans and NHS improvements, a former government adviser said there is "no plan B", even as the industry burns through vast sums of money and resources without matching revenue, prompting economists to speculate about an eventual "levelling" or bursting bubble.