Intelligence gives power, and right now we don't know how to make those systems obey our rules, like do not help somebody to create very destructive effects.
We do need to figure out, and I think we can, how to make sure AI will be safe, protect the public, will not harm people, will not be used in harmful ways. But we do need to coordinate internationally to make sure we can make it happen.
Are governments around the world taking the threat from AI seriously enough? Professor Yoshua Bengio is one of the so-called godfathers of artificial intelligence. He's previously warned that AI could eventually replace almost every single job. He is an eminent scientist and has incredible expertise on the risks of AI. He drafted international declarations, chaired a major report, and since 2023 has had an advisory role to the UK government. Welcome back to Newsnight. Thank you for being with us and talking to our audience again.
You are among those experts seriously concerned about AI. Why?
Because intelligence gives power, and if the trend continues and there's no sign of slowing down, it's like opening a Pandora's box with many possible bad things and good things, which is also why we're continuing to explore that Pandora's box.
Right. I'm going to ask you about the bad things, if I may. What are you most concerned about?
I'm most concerned about how the power of AI could be abused, mostly in the hands of people who don't trust each other, governments who don't trust each other, companies who don't trust each other, and are willing to take risks with public safety, are willing to consider a world where the decisions are taken by just a few leaders, a few CEOs, or a couple of governments, and moving away from the idea of democracy where we all decide together in some ways.
Right. What are the risks to citizens around the world then?
Well, there are many, and you mentioned jobs. That's pretty clear, although it's not for tomorrow morning. We're going to see it slowly, but even a small percentage of jobs being automated is going to have a strong effect on our already fragile economies. I'm also worried about how the technology that we're putting out already we're seeing is used by bad actors and criminals. For example, with cyber attacks, the AIs have reached a point where they can be used to damage our infrastructure, where ordinary people who are not experts might be able to use them in ways that could be very destructive for us. But it's just the tip of the iceberg in the sense that intelligence gives power, and right now we don't know how to make those systems obey our rules, like do not help somebody to create very destructive effects.
We do need to figure out, and I think we can, how to make sure AI will be safe, protect the public, will not harm people, will not be used in harmful ways, but we do need to coordinate internationally to make sure we can make it happen.
Okay, I want to come back to that international action and whether it's a possibility. In terms of the jobs, I mean, what are the consequences for the fabric of society if most jobs are wiped out?
Well, it's not going to happen in one day, and it's not going to happen all the jobs at the same time. There are jobs that are easier to automate already. For example, we can see in the realm of engineering jobs, programming jobs, this is already happening. And when the demand for those jobs is strong, then maybe we just have more output. But when the demand is fixed, for example, people who are providing answers, support, and so on, they're just going to lose their job. It's not like there's going to be better service, very likely. So we have to think also if governments are going to help the people who lose their jobs, where are they going to take the money? And the only place where they can take the money is where the profits of AI will be.
Right. Is that going to happen?
