Read up on the long history of wrong predictions of existential threats. When I was a kid in school we practiced for nuclear bombs by crawling under our desks. Of course anything might happen, and so-called AI will have unknown effects, but we’re probably more likely to see more fraud and advertising than the collapse of democratic civilization.

I’ll watch Doomsday Preppers for the entertainment of seeing people center their lives and the lives of their families around unlikely events and unpredictable effects.

Good to see Henry Kissinger talking about existential threats, something he has personal experience with. I’m writing this from Saigon, a place with first-hand experience of Kissinger’s expertise.

Maybe invest in some AI companies, if you want.

Technology-driven mass calamity, as you put it, follows no rule or law that causes it to play out slow, and frankly, I can think of half a dozen examples where it in fact did. Most notably, nuclear weapon proliferation became a crisis within a few years of the invention of nuclear weapons (and required enormous capital inputs to reproduce, whereas I already have a copy of LLaMA.) Other examples include: the arrival of firearms; the invention of crack cocaine; and the radio, which, we must be at pains to recall, was one of the key tools used by the National Socialists to weaponize the latent hatreds of 1920s Germany.

Given that with minimal effort I've thought of several plausible counterexamples, and you self-report as beginning from your feels about alarmism, it seems to me like you're operating from general dislike of alarms -- "just unalarmism", I could say. A good, solid pragmatic stance that is simple, straightforward, and, given the technology in question, probably wrong.

I'd love to keep the conversation focused on facts, rather than feelings, and predictable, inferrable consequences, rather than on historical rules-of-thumb. (Unnamed and uncited, at that.)

It's a sad(?) reality of life our agency is _so small_. We do what we can. But even there, the best we can do is align ourselves to whatever forces of history our out there that correspond to our values. Even "great people" have more often than not been thrust into whatever role they had, rather than single handedly willing something to happen.

Agency is limited, sure, but futility is not a foregone conclusion.

Especially at the scale I'm thinking and working. Like, I'm interested in resources for my loved ones, protecting what I can.

There is stuff like trad prepping, sure, but we could also begin to talk about flexible governance structures, recruits, that sort of thing. Organizing.

Survival will require oodles of trust.

I have no idea what we could do, personally, or as societies, to usefully prepare. Even if there were things that could be done on a society level, I'm pretty sure that most societies wouldn't bother - or see the point - until to late. As the last few years have demonstrated, collective action is not our strong suit, in the west.

I doubt there any other option besides abstinence. Since that won’t happen, I plan to enjoy the ride.

The outcome for most of these is usually unknown. Sure, PCs led to spreadsheets. But they created tons of other things. The internet brought about an era of misinformation and porn, but also spawned Facebook and cheap books in your pocket. Apps led to ride sharing and the ability to report that asshole who ran the red light.

AI will do a lot of obvious things, but I'd say 80% of the things that emerge is unexpected. I mean, LLMs are basically text autocomplete machines and as of today, they answer physics questions better than most trained humans. There's an emergent factor.

Whatever AI will become, it won't be Skynet, because it's literally trained to not become that.

Humans are the best creatures on Earth at adapting. Heck, that's probably the only skill we have. My opposable thumbs are literally talking to someone on the opposite side of the planet now. We'll figure it out.