Crystal ball: Are you concerned about the increasing likelihood of an AI going rogue and killing us all? It’s a depressing thought, but an AI security expert sees the positive: His concern that the technology will lead to mass extinctions in the next decade has led him to stop saving for retirement.
Nate Soares, president of the Machine Intelligence Research Institute (MIRI), revealed his pessimistic decision to The Atlantic CEO Nicholas Thompson in a recent interview for his series “The Most Interesting Thing in AI.”
“I’m not saving for retirement right now. I have a 401(k) from when I worked at Google, but I haven’t contributed in a decade,” Soares said.
Apocalyptic warnings about AI are a dime a dozen, but they hit even harder when they come from an expert in the field. Soares has worked on AI direction for 12 years. Before joining MIRI, a nonprofit focused on AI risks, he was an engineer at Google and Microsoft. He is also the author of a book with the worrying title “If Someone Builds It, Everyone Dies.”
The problem, in Soares’ words, is not that a superintelligent AI would hate us. He wouldn’t care if we lived or died. People could become obstacles to the goals it develops, especially if we try to shut it down or set up a competing system. It’s a warning that other AI experts have also expressed.
“From my perspective, the real question is not how does it kill us, but how does it become self-sufficient?”
At the moment, the AI still needs us to keep the lights on and the chips to come. Once it can run its own automated economy, this dependence will disappear. We could be deliberately eliminated or simply die as the machines reshape the planet to their liking. None of the results sound particularly retirement-friendly.
Soares is less certain about the exact path to disaster. He compares it to playing against world chess champion Magnus Carlsen: You don’t have to know which piece will checkmate you to suspect that you’re in trouble.
But he insists there is still an alternative to waiting for the end.
“I don’t think it’s inevitable. I think we could definitely still stop this race to the very dangerous kind of AI,” Soares said. But his decision not to save for retirement suggests he is expecting the worst.
The potential existential threat of AI has made a lot of headlines recently. After the many incidents in which AI agents hacked other companies without instructions, an Anthropic researcher resigned, warning that the technology could wipe out humanity.
In 2025, Geoffrey Hinton, one of the so-called “Godfathers of AI,” said there was a 10 to 20% chance that AI could take control, agreeing with a prediction first made by Elon Musk.
