Yoshua Bengio

And as a consequence, our immune system would not recognize those pathogens, which means those pathogens could go through us and eat us alive.

And in fact, eat alive most of living things on the planet.

And biologists now know that it's plausible this could be developed in the next few years or the next decade if we don't put a stop to this.

So I'm giving this example because science is progressing sometimes in directions where the knowledge in the hands of somebody who's

you know, malicious or simply misguided, could be completely catastrophic for all of us.

And AI, like superintelligence, is in that category.

Mirror life is in that category.

We need to manage those risks.

And we can't do it, like, alone in our company.

We can't do it alone in our country.

It has to be something we coordinate globally.

I would say there is a risk that we haven't spoken about and doesn't get to be discussed enough, and it could happen pretty quickly.

And that is the use of advanced AI to acquire more power.

So you could imagine a corporation dominating economically the rest of the world because they have more advanced AI.

You could imagine a country dominating the rest of the world politically, militarily, because they have more advanced AI.

And when the power is concentrated in a few hands,

Well, it's a toss, right?

If the people in charge are benevolent, that's good.

If they just want to hold on to their power, which is the opposite of what democracy is about, then we're all in very bad shape.

And I don't think we pay enough attention to that kind of risk.