How the bad science of AI doomerism is good for big business
thebulletin.org
thebulletin.org
Human beings are addicted to it. And, unfortunately, it sells. It sells not just in politics, where in my country there is a surge of blaming the "other" (Russian disinfo ops, immigrants) for the slow decay of our lifestyles, it also sells in tech to otherwise smart people.
We'd all benefit from a little less fear, which shuts off our ability to reason. I'd so value more rational discourse that doesn't immediately veer into this extreme or that extreme. How can it be that we across the globe are so polarized, so incapable of having reasonable discourse that doesn't immediately result in wild-eyed proclamations of the total extinction of the human race or turning the "other side" into slavering devils who may as well be non-humans?
There are lots of juicy parallels between climate risk and AI x risk. Both claim non trivial chance of human extinction and pause economic activity for it.
Extinction is much less interesting to discuss because it's so binary. Global catastrophe is much more interesting to discuss because the topic then becomes about failure modes, cascading failures, and the impact of those failures.
For example significant climate change activists argue that collapse is imminent if we don’t do something about emissions.
Yes I do think tail risks are mostly not true for climate change. And I think it’s the same for AI risks.
They had the time to find and interview Gebru and Bender to say that AI will never be smart enough to matter (which, in 2026: cmon), but not to engage with a single actual paper on the subject. Instead they just tossed some faces on a gradient, said that their reasoning is “not clear”, and moved on. Maybe, idk… maybe the books and papers and stuff might help? Or you could interview them?
Sources for the claim that AI scientists are “bad scientists”? Well, let’s run em down:
- some people speculated they are, without specifically citing an argument, figure, or even vague group.
- there’s a plausible motive for them to lie, as they want their field to exist and be important.
At least the climate change doubters had some data! These folks only remembered to bring the spurious allegations part. Kids these days…
The people who got predictions right on the other hand are people like Yud and Nate Soares who think x risks are real.
So here you have people with a good track record, who know how things work and they seem to suggest x risks are real. What do we take from this?
>There is no clear path showing that today’s AI systems will lead to superintelligence
was iffy. I mean two or three years ago leading models got about 100 on IQ tests, recently 135. Just keep doing what we are doing and let the trend play out.