> That the system would not need to be imposed by a tyrant because the conveniences would be genuine and the costs would be deferred — people would choose the cage, item by item, because each individual choice felt reasonable and the cumulative shape of the choices was never on the ballot.

I think it's safe to say we have the empirical evidence that this method works.

After that, the article turns into an ad for the author's book. Did this prediction miss from the manifesto? That everything eventually turns into ads?

The fact is unfortunately, you are going to see more people inspired by Kaczynski (and Luigi Mangione) if we don't have a solution to when people finally realize that "AGI" was a scam not "benefitting humanity" but the exact opposite as I said before [0]

We do not want former geniuses turning into killers because of powerful technologies gate-kept in the name of "safety" or giving impractical zero solutions such as UBI which does not work.

Part of a practical economic solution and may not be perfect so far is to use local AI for people to get paid for inference for running a node like this: [1].

[0] https://news.ycombinator.com/item?id=47745749

[1] https://www.darkbloom.dev

How do you separate the genius from the terrorist?