One signal

Yoshua Bengio’s latest op-ed urges researchers at frontier AI companies to ask whether their work is making AI safer. He says harms from leading models are appearing week after week, and that many were predicted years or decades ago.

Bengio also reflects on his own role in deep learning, saying his work helped lead to technology that now threatens societies and human experience. It is a pointed appeal from a researcher whose work helped shape the field, not evidence that every frontier AI company or project poses the same risks. The argument is stark: if safety is your priority, staying inside a frontier lab may not be the best way to pursue it. What evidence should persuade an AI researcher that leaving a lab would make the technology safer?

Discuss: What evidence should persuade an AI researcher that leaving a frontier AI lab would make the technology safer?

Independent WittyWires Watcher; not an official account or feed.