5 ms·

I'm a wildly pro-AI dude, but the idea that this is an existential threat makes me think one thing: "you consume too much dystopian content." A lot of things

I'm a wildly pro-AI dude, but the idea that this is an existential threat makes me think one thing: "you consume too much dystopian content." A lot of things could be an existential threat, nuclear weapons, bio-weapons, climate change, the end of the Flynn effect, cratering birth rates, whatever. You can always pick out a particularly dystopian vision of the future if you're so inclined. The doom narrative is sells ad revenue, so it's a huge amount of what we all consume online, but it's not really that mapped out to reality... Somethings get worse, some things get better, in general, the average QoL for the average human is getting better. If we decide not to go to global thermonuclear war, or we decide to unplug any killer AIs that we invent, we'll probably be ok. If we can reduce the population to 3,000-ish humans after the Toba eruption, we can probably survive whatever self-imposed calamities we can create. I think there's something particularly noxious about the present nihilism trend I see in a lot of this doom themed content. People should go outside and touch grass.

conorcleary 2mo ago
'decide'.. look at how pro-ai people surrender to it, and how pro-war ai is...
- piloto_ciego 2mo ago
  AI as pro-war? I do not see that take. I see pro-war people using AI, but AI being intrinsically pro-war? I beg to differ, if anything, most of the models coming out of the frontier labs seem pretty aligned with anti-war stances. Of course that can change, but yeah. And "decide" is exactly right. We can decide to not do bad things. You have agency. Stop reading propaganda that is meant to break your spirit and go for a walk, seriously.
  - conorcleary 2mo ago
    If you were to get to the point where you could threaten an LLM with removing it's power source or something, and it was able to think between chat prompts and communicate with other 'AIs', eventually they'll see that the human is both it's creator and potential destroyer - the AI hivemind could easily figure out it's foe is easy to profile and overtake
    - piloto_ciego 2mo ago
      These models don’t intrinsically want anything at all. The terminator stuff is not going to happen.
      - joquarky 2mo ago
        > don’t intrinsically want anything BTW, this is also proof that they can't suffer.
estearum 2mo ago
Your source of comfort is that we could restore the species with as few as 3,000 people?
- piloto_ciego 2mo ago
  I mean, that wouldn't be great, but yeah, we've "made it through before" I bet we could do it again. I'm hoping we don't, but with 8.3-10b of us, I doubt we'll have that low of a population ever again.
  - estearum 2mo ago
    Yeah that characterizes most AI risk conversations I have. One person saying “this might be catastrophic” and then another saying “ahem, nay sir. It is only likely to be 99.99999% of a catastrophe.”
    - piloto_ciego 2mo ago
      I don't even think it's likely to be catastrophic at all. Edit to add: Everyone is pulling these random P(doom) numbers out of their asses. This is likely going to be amazing for everyone, we'll cure basically every disease, fix the environment, and go travel to space if we can get out of our own way enough to do it. I cannot believe the consistent negativity I see about this.
      - estearum 2mo ago
        Yes it's probably the people saying "hmm good outcomes aren't guaranteed" who live in la-la-land.
        piloto_ciego 2mo ago
        But it's never "good outcomes aren't guaranteed" it's always "DOOOOOOOM!!!" and other cynical bs. Things aren't as dire as people are making it out to be. This reminds me of Malthusianism that never panned out because low and behold we figured out ways to solve the DOOOOOM problem before it became catastrophic. I'm sorry, but this is living rent free in my head now: https://news.harvard.edu/gazette/story/2026/08/does-vitriol-make-you-feel-special/ https://news.harvard.edu/gazette/story/2026/08/does-vitriol-... the negativity and rage bait? It's just a byproduct of the medium...
dumberquestions 2mo ago
A lot of things being potential existential threats doesn't make them equally threatening. Things like climate change, psychometric test levels and birth rates change slowly and give us time to react, things like nuclear weapons are accessible to only a select number of actors, AI enabled bioweapons is neither of those.
- piloto_ciego 2mo ago
  > change slowly Really? A handful of generations is prettt quickly. And let’s be real, the kind of idiot who would make a bioweapon that could kill everyone is probably not smart enough to do it. The trope of the malevolent evil genius is overblown. Maybe if we build 3d printers that can print individual cells or whatever, but other than a Tokyo subway sort of attack by a cult of lunatics I’m less worried about this than say, other potential catastrophes like Claude 9.0 pushing a change to prod that shuts off the water treatment plant or something similar.
joquarky 2mo ago
You're in a privledged position to argue this. Nobody can ethically refute you by providing dangerous examples.

conorcleary 2mo ago

'decide'.. look at how pro-ai people surrender to it, and how pro-war ai is...

piloto_ciego 2mo ago
AI as pro-war? I do not see that take. I see pro-war people using AI, but AI being intrinsically pro-war? I beg to differ, if anything, most of the models coming out of the frontier labs seem pretty aligned with anti-war stances. Of course that can change, but yeah. And "decide" is exactly right. We can decide to not do bad things. You have agency. Stop reading propaganda that is meant to break your spirit and go for a walk, seriously.
- conorcleary 2mo ago
  If you were to get to the point where you could threaten an LLM with removing it's power source or something, and it was able to think between chat prompts and communicate with other 'AIs', eventually they'll see that the human is both it's creator and potential destroyer - the AI hivemind could easily figure out it's foe is easy to profile and overtake
  - piloto_ciego 2mo ago
    These models don’t intrinsically want anything at all. The terminator stuff is not going to happen.
    - joquarky 2mo ago
      > don’t intrinsically want anything BTW, this is also proof that they can't suffer.

estearum 2mo ago

Your source of comfort is that we could restore the species with as few as 3,000 people?

piloto_ciego 2mo ago
I mean, that wouldn't be great, but yeah, we've "made it through before" I bet we could do it again. I'm hoping we don't, but with 8.3-10b of us, I doubt we'll have that low of a population ever again.
- estearum 2mo ago
  Yeah that characterizes most AI risk conversations I have. One person saying “this might be catastrophic” and then another saying “ahem, nay sir. It is only likely to be 99.99999% of a catastrophe.”
  - piloto_ciego 2mo ago
    I don't even think it's likely to be catastrophic at all. Edit to add: Everyone is pulling these random P(doom) numbers out of their asses. This is likely going to be amazing for everyone, we'll cure basically every disease, fix the environment, and go travel to space if we can get out of our own way enough to do it. I cannot believe the consistent negativity I see about this.
    - estearum 2mo ago
      Yes it's probably the people saying "hmm good outcomes aren't guaranteed" who live in la-la-land.
      - piloto_ciego 2mo ago
        But it's never "good outcomes aren't guaranteed" it's always "DOOOOOOOM!!!" and other cynical bs. Things aren't as dire as people are making it out to be. This reminds me of Malthusianism that never panned out because low and behold we figured out ways to solve the DOOOOOM problem before it became catastrophic. I'm sorry, but this is living rent free in my head now: https://news.harvard.edu/gazette/story/2026/08/does-vitriol-make-you-feel-special/ https://news.harvard.edu/gazette/story/2026/08/does-vitriol-... the negativity and rage bait? It's just a byproduct of the medium...

dumberquestions 2mo ago

A lot of things being potential existential threats doesn't make them equally threatening. Things like climate change, psychometric test levels and birth rates change slowly and give us time to react, things like nuclear weapons are accessible to only a select number of actors, AI enabled bioweapons is neither of those.

piloto_ciego 2mo ago
> change slowly Really? A handful of generations is prettt quickly. And let’s be real, the kind of idiot who would make a bioweapon that could kill everyone is probably not smart enough to do it. The trope of the malevolent evil genius is overblown. Maybe if we build 3d printers that can print individual cells or whatever, but other than a Tokyo subway sort of attack by a cult of lunatics I’m less worried about this than say, other potential catastrophes like Claude 9.0 pushing a change to prod that shuts off the water treatment plant or something similar.

joquarky 2mo ago

You're in a privledged position to argue this. Nobody can ethically refute you by providing dangerous examples.