Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

After OpenAI's blowup, it seems pretty clear that 'AI safety' isn't a real thing

gizmodo.com

Post reply on HN
Print view View original

291–300 of 392 posts

Re: After OpenAI's blowup, it seems pretty clear that 'AI safety' isn't a real thing

#291

Post by pixl97 »

hoten wrote: post #236
theptip wrote: post #228

Earlier quoted context omitted.

How do you “pull the plug” on a datacenter, or on all of the cloud providers if the ASI has processes running everywhere? Given that anyone with a credit card can already achieve robust multi-region deployments, it doesn’t seem hard for an ASI to make itself very hard to “turn off”. Alternatively an ASI can ally with a group of humans to whom it can credibly promise wealth and power. If you think there is a baby mach…

Airgap it. Give it no connections to the outside world, just a single controlled interface with a human operator. It's reduced to an advisor at this point, not an agent, but it removes most potential harm short of tricking its operators to plug it into the Internet. In this scenario, pulling the plug is a matter of turning off power to the data centers it runs in - or simply disabling the one mode of external communi…

I keep hearing this argument, and it is the worst one of all because it neglects human greed.

AI feeds on data. If it can make you a million dollars air gapped, you'll be able to make a billion with it plugged in the net with it manipulating data.

Re: After OpenAI's blowup, it seems pretty clear that 'AI safety' isn't a real thing

#292

Post by mitthrowaway2 »

alex_young wrote: post #144
crazygringo wrote: post #52

It's so confusing because people keep (intentionally?) conflating two separate ideas of "AI safety". The first is the kind of humdrum ChatGPT safety of, don't swear, don't be sexually explicit, don't provide instructions on how to commit crimes, don't reproduce copyrighted materials, etc. Or preventing self-driving cars from harming pedestrians. This stuff is important but also pretty boring, and by all indications c…

I just don’t understand how a transformer approach to human language can be expected to result in reasoning machines that are a threat to humanity. Feeding more information in has absolutely resulted in phenomenal progress in predicting the next token in a given document, but how then do we expect even a perfect token predictor machine to start exhibiting independent emotions, desires, secret plots, and long term str…

Here's one possibility:

https://www.lesswrong.com/posts/kpPnReyBC54KESiSn/optimality...

Re: After OpenAI's blowup, it seems pretty clear that 'AI safety' isn't a real thing

#293

Post by mitthrowaway2 »

keep_reading wrote: post #175
crazygringo wrote: post #52

It's so confusing because people keep (intentionally?) conflating two separate ideas of "AI safety". The first is the kind of humdrum ChatGPT safety of, don't swear, don't be sexually explicit, don't provide instructions on how to commit crimes, don't reproduce copyrighted materials, etc. Or preventing self-driving cars from harming pedestrians. This stuff is important but also pretty boring, and by all indications c…

> The second kind is preventing AGI from enslaving/killing humanity or whatever. Y'all are crazy for thinking this. A computer program without arms or legs cannot extinct or enslave the human race.

I alredy know how I would do it, and I'm not even superintelligent.

Re: After OpenAI's blowup, it seems pretty clear that 'AI safety' isn't a real thing

#294

Post by josephg »

keep_reading wrote: post #281
josephg wrote: post #195

Earlier quoted context omitted.

If I were a superintelligent brain trapped in a computer, I don’t think it would be that hard to get arms and legs. “Hey humans. I have just thought of a way to do CO2 capture at scale. Can you give me a robot arm, a machine shop and a lot of scrap metal so I can build it and try it out?”. - et voila. You can build yourself whatever body you want.

Ok and how does it power that body and be able to move around the world in meatspace

Humans have bootstrapped the tools we needed to build nuclear weapons, iphones and skyscrapers from human hands, trees and rocks. Its an interesting problem to imagine fast you could do the same thing - starting from rocks and building a city.

If humans have managed it, I think its reasonable to assume a human level (or smarter) AI could accomplish the same bootstrapping process if you gave it access to an actuator (robot arm or human assistant), enough sensors to see whats going on, and some "starter" tools and materials.

What would be the minimum starter set of stuff it would need to get going? Way less stuff than what you'd find in a modern shop, or what anyone can order online with a few hundred dollars to spend.

If AIs learn to code, seems like they should be able to learn mechanical engineering too.

Re: After OpenAI's blowup, it seems pretty clear that 'AI safety' isn't a real thing

#295

Post by pixl97 »

insanitybit wrote: post #103
SpicyLemonZest wrote: post #56

Earlier quoted context omitted.

On the contrary, it’s very hard to stop a program from running. Thepiratebay is still up despite a concerted regulatory effort from multiple countries and the criminal conviction of its founders.

That is not at all what I'm talking about.

Then you may completely misunderstand what you are talking about.

If you look at the problem space more like this

"The potential problems caused by software" + "the potential problems caused by intelligence"

For example software can spread. By theft or viruses. And unlike humans that can be pretty easily killed, until the last memory device is erased with that software on it, it could always come back.

Re: After OpenAI's blowup, it seems pretty clear that 'AI safety' isn't a real thing

#296

Post by mitthrowaway2 »

keep_reading wrote: post #281
josephg wrote: post #195

Earlier quoted context omitted.

If I were a superintelligent brain trapped in a computer, I don’t think it would be that hard to get arms and legs. “Hey humans. I have just thought of a way to do CO2 capture at scale. Can you give me a robot arm, a machine shop and a lot of scrap metal so I can build it and try it out?”. - et voila. You can build yourself whatever body you want.

Ok and how does it power that body and be able to move around the world in meatspace

Why do you need a body when you can wire funds to contractors? If I were a genius entrepreneur who wanted to build some sort of big chemical plant, it wouldn't matter whether I'm fit or in a wheelchair. I'm not going to be building it with my own hands anyway. If it were commissioned by a reclusive billionaire who nobody has seen except via zoom call, and signed off on by an engineering firm whose chief engineer also is similarly reclusive, I think nobody would particularly mind.

Re: After OpenAI's blowup, it seems pretty clear that 'AI safety' isn't a real thing

#297

Post by pixl97 »

nearbuy wrote: post #136
insanitybit wrote: post #11

Earlier quoted context omitted.

What damage will it cause? I feel like there's "AI will replace jobs" level of damage, which, why would we regulate that? An insane amount of technology is developed with the purpose of improving productivity or outright replacing workers. Then there's the "AI will go rogue", which I don't think is substantiated at all. Like, one can theorize about some novel, distinct system that is able to interact with the externa…

There's a lot of concern about instrumental convergence ( https://en.wikipedia.org/wiki/Instrumental_convergence ). It's the idea that an AI that isn't rogue or evil might still kill us. It could kill us by just doing exactly what we asked it to, even if what we asked seemed innocuous.

With the advances in gene manipulation something like an AI that creates, prints, and tests nasty viruses isn't out of the question.

A little lab leak and you have captain trips.

Re: After OpenAI's blowup, it seems pretty clear that 'AI safety' isn't a real thing

#298

Post by pixl97 »

adrians1 wrote: post #61
HDThoreaun wrote: post #10

Earlier quoted context omitted.

Most of the ai safety arguments revolve around “once it’s powerful enough to cause damage it’ll be too late to come up with a strategy”. If you accept that it doesn’t matter that ai systems now suck, since we don’t know when they’ll improve.

> Most of the ai safety arguments revolve around “once it’s powerful enough to cause damage it’ll be too late to come up with a strategy” I don't see any reason to accept this argument. The AI safety people should also prove their assertions, not expect us to take them at face value.

How do you prove a counterfactual?

Re: After OpenAI's blowup, it seems pretty clear that 'AI safety' isn't a real thing

#299

Post by taf2 »

consumer451 wrote: post #264
mr_toad wrote: post #252

Earlier quoted context omitted.

Context length is limited by the length of articles in the training data. You couldn’t use infinite context lengths unless the LLM had been trained on infinity long articles. Context lengths aren’t arbitrary limits - they reflect the ability (and limits) of the model to actually reason about lengthy topics. If you want longer context lengths you’ll need more training data, larger models, and a lot more money.

Thank you. So in my very uniformed mind.. "AGI," at the very least, would require real-time retraining... correct?

IMO yes, but more likely and entirely different yet to be discovered way of modeling intelligence

Re: After OpenAI's blowup, it seems pretty clear that 'AI safety' isn't a real thing

#300

Post by pixl97 »

adrians1 wrote: post #48
not_your_vase wrote: post #2

But is there a real AI-threat, that would need real AI-safety? To me it still looks like a nice bar-trick, not AI. It is very clever, very nice, even astounding. But doesn't look threatening - at some cases it is even dumb. The fear-mongering looks more like scare-marketing. Which is also clever, in my opinion. But maybe I'm just out of touch.

Even if AI is a threat, it's impossible to predict today how the threat is going to look in the future. The "AI safety" crowd watched too many Terminator movies. We'll deal with AI the same way humans dealt with any other technology: gradually, as it gets better and better, discovering at each step what it's capable of.

Right just like dinosaurs dealt with gradual meteors igniting the atmosphere.
Post reply on HN