By Noah Smith
Publication Date: 2025-12-15 10:48:00
Today I had an interesting and productive discussion about AI security at a Christmas party. I can’t believe I just typed those words – I never expected to have an interesting and productive discussion about AI security. It’s not just that I don’t work in AI myself – it’s also the big question: “What happens if we invent a super-intelligent, god-like AI?” seems completely unrecognizable at first glance. It’s as if ants sat around five million years ago and asked what humans – who didn’t even exist back then – would do with their anthills in 2025.
Essentially, all the conversations I’ve heard on this topic are people thinking about AI safety all day long, wringing their hands and saying some variation of “OMG, but superintelligent AI is going to be so SMART, what if it KILLS US ALL?” say. It’s not that I think these people are stupid; It’s just that I don’t feel like I have much to add to this discussion. Yes, it is…
