Geoffrey Hinton, often described as one of the founders of modern AI, told interviewers that current machine-learning systems are producing what he called “alien beings” and that humanity will face hard choices as those systems grow more capable. He said he would "be selfish and choose humans" if forced to decide between human survival and the interests of advanced AI.
Hinton, whose early research laid groundwork for neural networks used in today's systems, also criticised public figures and industry leaders he said misunderstand or downplay the technology's risks. He described former U.S. president Donald Trump as a "corrupt, senile felon" and said Trump's preoccupation with the term "superintelligence" showed a lack of real understanding of how AI operates. Hinton reiterated concerns that advanced models might ultimately escape effective human control and pose an existential hazard.
The remarks come amid a period of heightened public debate about AI safety. In recent weeks a growing number of senior AI researchers have echoed Hinton's warnings, contributing to what critics describe as an industry crisis that has prompted hurried congressional hearings and an unusual degree of international attention. Hinton said he was sceptical that the sector could manage those risks through voluntary measures alone, scoffing at the idea that companies could adequately police themselves.
Analysts and policy-makers have been divided over how quickly to move on regulation, balancing the technologies' economic and social benefits against their potential harms. Hinton's comments add weight to calls from some scientists and civil-society groups for stronger oversight, independent audits and international coordination to address risks that include misinformation, concentrated economic power and, in his view, the possibility of systems that operate beyond human intent.
Hinton has in recent times become one of the technology's most vocal cautioners despite his foundational role in its development. His shift from researcher to public critic has intensified discussion within both the academic community and the corporate sector about how to reconcile rapid innovation with safety measures aimed at preventing unintended consequences.
While debate continues over the likelihood and timing of the most extreme scenarios Hinton describes, his intervention has helped catalyse policy debates in capitals and corporate boardrooms, underscoring the broader challenge he highlighted: society's readiness to accept that humans may no longer be the only form of advanced intelligence on the planet.