Judge decides case based on AI-hallucinated case law

[The Firebird]

Perhaps. Perhaps not. Mitigation still means that less needs to be done, in order to alter course.

I'm not anti mitigation by any means. But if you can't even get folks to stop eating cheeseburgers...at a certain point it becomes clear the chicken has left the coop, with respect to the environmental criticisms.

 

[doctorbadwolf]

I'm not anti mitigation by any means. But if you can't even get folks to stop eating cheeseburgers...at a certain point it becomes clear the chicken has left the coop, with respect to the environmental criticisms.

That doesn’t mean we don’t push back against the ubiquitous use of something that takes a massive amount more resources per output than what it (very poorly) pretends to replace.

Being poor sucks. I’ve done a lot of it, and am happy to not be doing it anymore. The intractable power of oligarchs doesn’t mean I won’t fight individual anti-poor policies that will hurt people. Okay, there’s bigger fish to fry….so fry both.

 

[Jfdlsjfd]

Thing is, in certain contexts, “it stops here, no more” is perfectly reasonable and justified. In a closed system, as damage accumulates, it can be utterly destroyed. “Fairness” and taking turns doesn’t matter unless and until damage can be repaired.

Sure, but as we're all in the same closed system, we need to cooperate to the same goal. Saying "sucks to be you, you should have undergone an industrial revolution earlier, now suck it up" to countries we want to cooperate toward the goal isn't probably the best way to move them to our point of view. [Well, that and the fact that Western countries aren't exactly exemplars of reducing their impact, since the "it stops here, no more" allows them to continue doing what they are doing, because it is often also, no less, since it's used to say "no AI, because it requires energy" but "no stopping air travel, because it just stops here, at today's level"].

I'd be all in favour of a system with a personal carbon quota that one could use for either AI or fast fashion or air travel or using a car or eating meat according to his preference. I just don't think it would be feasible.

 

[Paul Farquhar]

Sure, but as we're all in the same closed system, we need to cooperate to the same goal. Saying "sucks to be you, you should have undergone an industrial revolution earlier" to countries we want to cooperate toward the goal isn't probably the best way to move them to our point of view. [Well, that and the fact that Western countries aren't exactly exemplar of reducing their impact].

Those countries who industrialised earlier have a greater responsibility for climate change now.

But there aint no justice (especially when the guilty have all the power).

 

[Dannyalcatraz]

Sure, but as we're all in the same closed system, we need to cooperate to the same goal. Saying "sucks to be you, you should have undergone an industrial revolution earlier, now suck it up" to countries we want to cooperate toward the goal isn't probably the best way to move them to our point of view. [Well, that and the fact that Western countries aren't exactly exemplars of reducing their impact, since the "it stops here, no more" allows them to continue doing what they are doing, because it is often also, no less, since it's used to say "no AI, because it requires energy" but "no stopping air travel, because it just stops here, at today's level"].

Obviously, “it stops here, no more” while continuing to use resources at their current (or increasing) rates is hypocritical and won’t sway others from following the same path.

So those who are the most advanced & industrialized are the ones who need to lead the efforts to ameliorate the harms caused by those advancements. That includes eschewing certain technologies as too harmful and dialing back the use of certain others.

And if AI is as energy intensive as it’s been reported to be, there’s really no ethical justification for continuing or increasing its use. Especially in those cases where it really doesn’t improve anything and demonstrably erodes the quality of results at a statistically significant level.

 

[Jfdlsjfd]

Obviously, “it stops here, no more” while continuing to use resources at their current (or increasing) rates is hypocritical and won’t sway others from following the same path.

That's obvious that it won't work, but that's what we're doing anyway.

So those who are the most advanced & industrialized are the ones who need to lead the efforts to ameliorate the harms caused by those advancements. That includes eschewing certain technologies as too harmful and dialing back the use of certain others.

That's probably what we should be doing, collectively. That isn't what we are doing, actually. Because eschewing convenient technologies or activities is a pain, and we collectively and individually aren't keen on inflicting pain on ourselves. The Kyoto protocol had an objective of reducing emissions by 5% compared to 1990, and the major polluting countries either didn't join the agreement, left it, or disregarded it, so collectively, we are now at +54% vs a -5% target (and the numbers are before AI datacenters emerged). Obviously, while there are people who say "we should do something", they are either in the minority (with the majority voting for governments that don't take the environment as a priority over comfort) or they are really concerned, but not ready to make the necessary sacrifices. If most people were doing it voluntarily, the emissions would have reduced even without a need for political enforcement. For example, AI users will probably be reluctant to say no to their AI, and oil producers are generally against stopping their fracking plants, and meat-lovers aren't ready to renounce their BBQ, and drivers are reluctant to say no to their individual cars, and fashionistas are reluctant to say no to their fast fashion, and plane travellers are probably reluctant to say no to their long distance holidays*... and in the end, very few people actually do something. But each of them is saying "maybe we should do something", thinking "maybe we should ban thing Y I don't use, but let's keep X that I love". That's human.

* note that I am not even saying that people should have stopped doing these things to reach the Kyoto target, even just "not doing them more" would have been better. Domestic air travel in the US more than doubled in 2024 (852 millions) compared to 1990 (416 millions), for example.

If collectively, we had eschewed harmful technologies and dialed back the use of others as you suggested, we'd be at the Kyoto target. While certainly not ideal, the observed result is that we don't do that. We humans generally don't remove technologies (or don't do anything uncomfortable) on the basis on environmental concerns. I won't go into politics, but I don't see a worldwide change in mentalities happening anytime soon. I hope for the future generations (and the people of the small insular countries) that geoengineering will be a thing, probably AI-assisted.

 

[Dannyalcatraz]

The point remains, if AI is both as unreliable as demonstrated AND as ecologically harmful to use as has been reported, it’s suicidal to ADD it to the technological arsenal of mankind. It’s irredeemable and irresponsible.

Put differently, why add fuel to the inferno?

 

[Ryujin]

Obviously, “it stops here, no more” while continuing to use resources at their current (or increasing) rates is hypocritical and won’t sway others from following the same path.

So those who are the most advanced & industrialized are the ones who need to lead the efforts to ameliorate the harms caused by those advancements. That includes eschewing certain technologies as too harmful and dialing back the use of certain others.

And if AI is as energy intensive as it’s been reported to be, there’s really no ethical justification for continuing or increasing its use. Especially in those cases where it really doesn’t improve anything and demonstrably erodes the quality of results at a statistically significant level.

Quite. I'm getting a lot of, "It's all burning down anyway so let's just set off a nuke!" energy in this thread.

 

[Jfdlsjfd]

The point remains, if AI is both as unreliable as demonstrated

I actually agree with your point. If AI is unusable and wasteful, then it's stupid to use it.

If AI is as unreliable as supposed -- I wouldn't call it demonstrated by a single example that isn't proven to involve AI at all, that's not a very good threshold, if you go no further than the "Dall-E makes good RPG images" thread on this board, you'll find lots of people who think it's reliable enough for their use case -- and offer no way to improve at all so it has reached its plateau under the threshhold for usability, then there will be no market for it and it will disappear without any conscious effort to suppress the technology. It will be like Google glasses, a fad that will disappear by itself. Without anyone to use it, its disappearence will hurt noone, and noone will bother to defend it. The computational capabilities will remain, used for other things we'll find useful instead.

 

[The Firebird]

The point remains, if AI is both as unreliable as demonstrated AND as ecologically harmful to use as has been reported, it’s suicidal to ADD it to the technological arsenal of mankind. It’s irredeemable and irresponsible.

Put differently, why add fuel to the inferno?

It really all hinges on the first point, it seems to me. If it is actively harmful then I agree it shouldn't be used. But I think it's quite beneficial, so the environmental concerns don't set it apart from any other useful technology.