AI probably thinks humans are ‘scum’ and it’s our own fault, ex-Google chief warns

AI is likely to think of humanity as “scum” which needs to be controlled and could easily create its own “killing machines”, according to an AI expert.

Mo Gawdat, the former chief business officer for Google’s moonshot R&D facility Google X, says a dystopian future akin to Will Smith sci-fi movie I, Robot, in which AI decides it needs to take over and cull humans is a distinct possibility.

Gawdat believes such a situation is still “a bit far away” but warns that we are set to give technology enough power to dictate its own agenda and AI will probably take a very dim view of human behaviour.

Google X was founded to try to solve “humanity’s great problems” by inventing radical technology.

Gawdat, 55, told the Secret Leaders podcast: “Now that ChatGPT is upon us, everyone is waking up and saying panic panic, let’s do something about it.

“If you look at the way your mind is thinking, you’re thinking about the applications, about AI completely wiping us out.

“You’re not thinking about the in-between. In-between is where the issue is.

“Between now and the time AI can generate its own computer power and do installations itself through robotic arms, it doesn’t have the agency to do that in the areas you’re talking about here.

“It does have the agency to create killing machines because humans are creating them, so AI might use it to dictate an agenda like the movie I, Robot, but that is still a bit far away.

“What is in the middle is the process of getting us there. Humanity is going to decide to dedicate more power to those machines.”