Once again, a warning of a "heavy" topic.

Since about two months ago of coming across the concepts of AI safety and s-risks, my life took a complete turn. Before this, I was just focusing on my survival. I always knew that we live in a world filled with suffering, but it's not hell. There seem to be some safeguards that this world's creator put in order.

a) The most obvious one being death, and suffering-causing trauma eventually leading to it, so that the suffering isn't prolonged into eternity. There's also suicide as a way out for extreme cases.

b) Many victims of violent deaths will even enter shock, drastically decreasing the amount of suffering that they will go through. Adrenaline also reduces suffering in these cases.

c) Shock is only present in mammals, which, along with other factors, makes me think that the more complex the brain, the higher the suffering. Suffering in insects is still debatable, and indeed it's hard to phantom how it would be in any order of magnitude compared to that of humans, since the same applies to their consciousness. Also, some of the most vulnerable and abundant beings like bacteria and plants can't suffer.

d) Certain emotions (or their absence) as a deterrent. Non-human animals can't be evil. Most humans will feel the suffering of others emotionally, and develop compassion.

e) The existence of medicines to be found in Nature, and more generally, the existence of solutions to most of our problems (until now, at least).

However, I see a very high probability that this balance (which is far from making this world an acceptable one, but again, it's not hell) being broken with the coming technologies of this century, most importantly advanced AI.

Since there seems at least to be some order in this world that prevents it from falling into hell/chaos, can we expect some laws of physics/biology to effectively stop the advent of hell/chaos? At least partially?

For instance, I hope that emulating consciousness in a computer is impossible (how would we sleep? how would we die? Consciousness seems to necessitate these things, among others). Or that at least, the emulation is not the original identity, aka death is certainly final.

I hope that immortality of any conscious being is impossible (here the wishful thinking of transhumanists comes into play... You WOULDN'T like to be immortal, believe me. Even 80 year olds are tired of life in many cases. Death is as essential as sleep. And don't get me wrong, I don't feel good about this fact (I don't wanna die) but it seems so nonetheless).

Maybe we can only be emulated in a computer of atoms (the universe), and not one contained in it and of different structure and far less complexity. As Michio Kaku says "the most efficient computer that can simulate the Universe, is the Universe" (regarding the Simulation Hypothesis).

I hope that advanced nanotechnology is physically impossible (though here my hopes aren't that high).

Other thing that gives me some hope is the Fermi Paradox. I mean, we already have people like Erik Lentz proving that warp drives are possible, they just need a ton of energy... If so, then any advanced civilization wouldn't take tens of millions of years to colonize the galaxy, but way less. But, like Fermi said, where is everyone? Where are the warp drives, or at least the Dyson Spheres and the cosmic AIs etc etc?

Or maybe we're just the first ones to get to technology and everything is indeed possible and chaos/hell is indeed possible. Ouch...

There's also the notion in this community that hell is rare among the possible worlds. Could someone elaborate?

This is all pure speculation obviously. My opinion is still that very bad things are possible, that not even 1% of us are aware (even in AI safety most just mention x-risk), and that we should start freaking out about it and taking action - because death/extinction is "acceptable" (they don't hurt (for long) and they're inevitable); hell on Earth (and beyond) is not.

AGI alignment will be impossible with the current expected timelines. The only hope, in my opinion, is convincing the world to stop all AI development (as well as nanotech) until we've made these technologies provably safe. It is a very difficult task but it's tractable, while the former is not.


New Comment
8 comments, sorted by Click to highlight new comments since: Today at 4:40 PM

You have arrived at a combination of ideas that I haven't seen before. I won't respond to everything, but just a few items. 

Starting at the end: aligning AI may or may not be achievable, but "convincing the world to stop all AI development" is hopeless, I think. For all we know, aligning AI requires just a few correct ideas. Stopping all AI development requires halting something that dozens of governments and hundreds of rich corporations are working on. Good luck halting even one-tenth of that activity! 

You're more worried about "s-risks" (immense suffering) rather than "x-risks" (extinction) arising from AI. As you have noticed, this is the opposite of how people normally think - they understand that the human race might simply perish, either through direct attack by machines (The Terminator) or from side effects of technology. But "s-risk" seems to require weird malevolence (a Matrix built to torture. or where humans are expendable NPCs) or a failed attempt at coerced happiness (e.g. wireheading that is actually an endless bad trip). Maybe the closest thing to an "s-risk" that sounds reasonable, is the idea of some endless dictatorship (1984). 

You also say we shouldn't want immortality, because many 80-year-olds are "tired of life". Well, life is tiring when your health is constantly breaking down. Rejuvenation doesn't seem to be part of your calculus. Human sensibility is greatly affected by biologically imposed factors, and we have very little idea of what's possible in a transhuman world. There's no honor in resigning ourselves to our current limitations; the only reason to be a luddite is fear that opening Pandora's box will bring about something worse than our current vale of tears. 

That there are palliative factors that limit suffering in our current world, does not strike me as good evidence that the world was designed to limit suffering; only as evidence that suffering is not the purpose of the world. As to whether the laws of nature "prevent hell", the main limiting factor seems to be, that pain is associated with damage to our organism, and so extreme pain usually means that something is happening, which unchecked will destroy the very possibility of being alive and aware. 

Physics is full of limits like the speed of light and the Bekenstein bound, and it's reasonable to think that there is some upper bound on how bad an experience can be in our world; but that ultimate bound may be a lot worse than anything that can happen to small, randomly evolved organisms. And detailed knowledge of body and brain already allows official torturers to cause pain of an unnatural degree and duration. 

So if your "hell" means something significantly worse than what we already have, I see no natural law that prevents it. The mitigating factor is just, probability. Is it likely that a posthuman world would be organized for the purpose of giving pain (thus being an s-risk)? Isn't it more likely that a posthuman world would be organized in support of some nonhuman values and form of life, thereby being unfriendly to human life as a side effect, and therefore being an x-risk for human beings. 

Very thoughtful reply, thanks. To each paragraph:

  1. If all governments and tech leaders could be effectively convinced that x-risk is not the only outcome, but also s-risk, then things would AT LEAST start changing. The biggest problem is that I think 99% are terribly deluded about this, both leaders and masses, and the biggest factor might well be self-delusion to preserve sanity/happiness. Is it a monumental task? Sure. But way more feasible than on-time alignment, specially in the sight of transformers and the scaling hypothesis. Also, things needn't be done in just a big whole step. If we could someway affect the production and availability of compute resources, that might very well give us decades (considering that the scaling hypothesis is what is creating the high probabilities on short timelines). Decades that could be used to implement other measures. It would be a gradual step.

  2. If you claim to be midly skeptical of s-risk and then give 3 quite plausible outcomes, you've just contradicted yourself. Most people worry only about x-risk because no one is talking about s-risk, aka social bias, and to preserve their sanity. But we must start realizing that physical suffering is way worse than any anxiety about it. If this happens, we might have a chance to do something.

  3. You're right, a self-rejuvenating 80 year old might not feel tired of life. It wasn't the best explanation. But maybe a 800 year old would. I can't see any way of sane immortality without partial memory deletion. But also partial personality deletion. So you'd still die in a way. Anyway this is all highly speculative. There's no way of knowing which side is right. That why I vehemently agree when you say

"the only reason to be a luddite is fear that opening Pandora's box will bring about something worse than our current vale of tears"

Precisely. Would you play wheel of fortune with 50% chance of heaven and 50% chance of hell? Or even 70-30, or even 90-10. Only someone living in a rare modern oasis of comfort would ever take that chance in my opinion, being removed from the worst possibilities on the hedonic scale.

  1. Agree that suffering is not the purpose of the world. It's merely a tool for survival. But it's a very outdated and cruel tool considering modern human life, which the more the technology, the more it can be hi-jacked (i.e. torture). That's how I see technology getting out of control and bringing up hell on Earth.

  2. That's what makes me a luddite, at least until some point.

  3. I hope so. Still, a very grim view, so that we should do everything to avoid it, aka doing everything to not develop AGI and nano before they are provably safe. The s-risk outcome is just too grim, and the probability doesn't seem small.

I think it's way too late to stop. Now the world knows what transformers can do, how are you going to stop it worldwide? Shut down every data center you can? Have trusted cyber-regulators overseeing every program that runs, in every remaining data center? 

I always promote https://metaethical.ai by @june-ku as the best concrete proposal we have. Understand it and promote it and you might be doing some good. :-) 

Basically we need a lot more Eliezers. We need a lot more AI advocates that tell it like it is, that make us shit our pants, that won't soften their message to appear reasonable. That are actually realistic about timelines and risk. As long as most of the popular advocates keep with the approach of "don't panic, don't be afraid, don't worry, it's doable, if only we remain positive and believe in ourselves" then there is no hope. As long as people will keep lying to themselves to avoid panic, there is no hope. Panic can be treated in many ways, even in the most extreme cases with benzodiazepines. Disaster once it settles has no solution, and it's a lot worse.

It would take a vast proportion of the world to shit their pants and form international organizations for regulation. As long as you can restrict global production and access to supercomputers, you can gain a few decades. Those decades will allow for more measures to be tried.

Formalizing ethics seems like a bad way. We need concrete priorities, not values. Value learning is dangerous. Anyway, like with most other alignment approaches, you'd need centuries for that. What's the probability you'll get there in 1-2 decades? I'd say less than 1%. Whereas my approach gives you time, time that can be used to try a multitude of approaches, yours included.

Shock is only present in mammals, which, along with other factors, makes me think that the more complex the brain, the higher the suffering. 

Where did you get that idea? The freeze response is evolutionarily older than the fight or flight response. It's from a time when predators frequently identified their prey by movement.

And by the way, shock is not freeze. Shock is going unconscious from the amount of pain, so that you suffer from say 5 seconds instead of 10 minutes. But like I say just one among many examples and the broader picture is what matters.

I read it somewhere while learning about these things. I might be wrong. It's not too relevant for the broader topic anyway.

New to LessWrong?