AI seems to be akin to the space or the arms race of the last century — even if participants want, they won’t necessarily be able to slow down progress unless their adversaries decide to slow down as well.
Google DeepMind CEO Demis Hassabis has made a striking admission: he would support pausing AI research under the right conditions. In a conversation with Emily Chang, the AI pioneer outlined his vision for a collaborative approach to developing artificial general intelligence (AGI).
When asked whether he would advocate for a pause if every other company and country would do the same, Hassabis responded in the affirmative. “I think so. I mean, I’ve been on record saying what I’d like to see happen. This was always my dream, the kind of roadmap at least I had when I started out DeepMind 15 years ago and started working with AI 25 years ago now, was that as we got close to this moment, this threshold moment of AGI arriving, we would maybe collaborate in a scientific way.”
Hassabis envisions something ambitious: “I sometimes talk about setting up an international CERN equivalent for AI, where all the best minds in the world would collaborate together and do the final steps in a very rigorous, scientific way involving all of society. Maybe philosophers and social scientists and economists, as well as technologists, to kind of figure out what we want from this technology and how to utilize it in a way that benefits all of humanity. And I think that’s what’s at stake.”
However, the DeepMind CEO acknowledges the fundamental challenge with this approach. “Unfortunately it kind of needs international collaboration though because even if one company, or even one nation or even the West decided to do that, it has no use unless the whole world agrees, at least on some kind of minimum standards. And international cooperation’s a little bit tricky at the moment. So that’s gonna have to change if we want to have that kind of rigorous scientific approach to the final steps to AGI.”
Hassabis didn’t name China directly, but it was clear what he was referring to. Even if the west got together to decided to slow down research, China was unlikely to participate, and would end up taking the lead in this critical new technology. Hassabis isn’t the only AI leader who’s said this — Elon Musk has said he’d slow down AI and robotics progress if he could, but admitted he couldn’t do it even if he tried. The paradox is clear: those building the most powerful AI systems recognize the risks, yet feel compelled to continue racing forward because unilateral restraint would simply cede ground to competitors. Without international coordination—something Hassabis admits is “a little bit tricky at the moment”—the AI development race will likely continue at breakneck speed, regardless of individual leaders’ reservations about the wisdom of doing so.