Analysis by Keith Rankin.
Role: Economic historian.
Keith Rankin: September 11, 2026 – Yesterday, much of the corporate and government media ran a story about AI [Artificial ‘Intelligence’; not AW, Artificial Wisdom]. (In some contexts, ‘intelligence’ is another word for ‘stupidity’.) The story is about how industry insiders are predicting that there is a ten percent chance that AI will kill all humans within a decade (or, even worse, “by the end of the decade”).
Here is the RNZ version of the CNN story: ‘Gambling with our lives’: Another AI employee quits over safety concerns. The story on ThreeNow News showed armies of robots looking anything but intelligent, but with humanicide on their minds.
It is not clear how AIs are going to kill us, but its unlikely to be through mindless humanoids going on a shooting spree. They will be more ‘clever’ than that.
What is strange is that, for anyone following the war between Russia and the E3 proxy that is Ukraine, the chances of an apocalyptic nuclear exchange in the next ten years would appear to be somewhat higher than 10%. (The latest news seems to be that Canada is having a ‘me-too’ moment, seeking to create an A4 [Atlantic Four], volunteering itself as a nuclear target. See, among many other stories largely absent from the New Zealand media, Canada pledges support for Ukraine air defense as Zelensky visits, France24, 10 September 2026.) The way that ‘The West’ is presently taunting the world’s equal-greatest nuclear power, Russia, daring it to deploy its nuclear arsenal, seems to me far more scary than any story about AI. Just as bad, if not worse, is our mute denial of this threat.
On this basis, there would appear to be a race going on to see which of these can first commit species suicide: Nuclear Weapons or AI. (There is of course a dreadful possibility that the answer is ‘both’; that AI does the deed through nuclear weapons; but I’ll ignore that.)
My key point is that a nuclear apocalypse would be more than a single-species suicide: it could be a murder-suicide that would destroy all life on Earth; indeed all life in the Universe if Earth is the only place with life.
So, could an AI holocaust be a good thing? The answer has to be ‘yes’, if AI gets there first, killing only all humans and not all life.
About the writer:
Keith Rankin (keith at rankin dot nz), trained as an economic historian, is a retired lecturer in Economics and Statistics. He lives in Auckland, New Zealand.