Geoffrey Hinton - the researcher whose work on neural networks underpins much of modern AI - said on September 14 that Dario Amodei's call to slow development was 'very sensible,' and that he backs companies pausing the pace of capability gains. The endorsement is significant less for its substance than for its source: the field's most famous warning voice has joined the labs' own plea.
Hinton, speaking on ABC Radio National Breakfast, agreed with Amodei's core argument that capability progress is outrunning safety and control research. His concern, like Amodei's, is that advanced systems could escape human control before the mechanisms for controlling them exist. It is a remarkably unified position coming from a man who has spent years warning that the risks of what he helped build were being underestimated.
The timing is what makes it notable. Hinton's endorsement lands days after Amodei published his essay calling for the industry to 'pace the frontier,' and hours after Sam Altman, Elon Musk, and Demis Hassabis echoed the call. The godfather joining in gives the argument a kind of historical weight that the lab CEOs cannot provide on their own - because Hinton is not asking for anything for himself.
The Brakes Find Their Witness
Hinton has been the conscience of the field for years, resigning from Google in 2023 to speak freely about the risks of the technology he helped create. His endorsement of a slowdown is not a business calculation. It is a judgment from someone who has repeatedly said the risk of losing control is real, and who has now agreed that the pace should be adjusted to account for it.
For the arc this paper has tracked since No. 73, Hinton is the missing voice. Amodei asked for brakes, Altman and Musk agreed, and the agreement became an antitrust defendant. Hinton's endorsement does not change the legal position. But it does change the moral one: the man who built the field's foundational tools, and then spent years warning about them, now says slowing down is the sensible thing to do.
The question is whether the field listens. The labs have argued among themselves. The regulators are divided. The godfather has spoken. If the weight of that endorsement moves the industry at all, it will be the first time a cautionary voice from outside the executive suite has shaped the pace of the frontier - which may be exactly the point.
