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

The good delusion: has effective altruism broken bad?

economist.com

Post reply on HN
Print view View original

361–370 of 619 posts

Re: The good delusion: has effective altruism broken bad?

#361

Post by ethanbond »

coldtea wrote: post #354
ethanbond wrote: post #319

Earlier quoted context omitted.

1) Why do we have a reason to believe there’s a cap to intelligence? 2) No, you don’t need to assume it happens via adding hardware 3) No, you don’t need to assume it can expand by itself 4) We’ve already seen people connecting AI systems to open internet and open internet to command and control of physical systems (cars, utilities, soon weapons systems) - turns out there was no convincing necessary 5) Nanobots also…

> 1) Why do we have a reason to believe there’s a cap to intelligence? There's a cap to all things. Diminishing returns is a thing. Both physical substrates have limits, and there's a utility limit to thinking or having more data. Limitless intelligence is fairy-tale magical thinking - especially in the way it's portrayed as some magical ability to build and control everything. Like, invent working cold-fusion - as i…

1) Obviously intelligence plays a role in how fast science develops. The quality of experiments and the quality of inference from those experiments are (very obviously) modulated by intelligence. No reason to believe the absolute pinnacle of intelligence happens to sit inside a species of mostly hairless apes here on earth. Lots of reasons to believe fundamental understanding of the universe is more accessible to intelligent beings than to unintelligent ones.

2) Nope. Could be different hardware. The brain consumes far less power than any human-designed computer that could hope to match its computational power.

3) Sure some do, but it’s not a necessary assumption to take AGI risk seriously.

4) Meh, this seems to me just the next concession in the line of “we can’t do that, we wouldn’t do that, we do it but it’s not that bad”

5) Okay so it’s unrelated to this conversation except I suppose some sort of ad hominem?

6) So we agree charisma is at best not necessary to manipulate the world

Re: The good delusion: has effective altruism broken bad?

#362

Post by FeepingCreature »

imgabe wrote: post #360
FeepingCreature wrote: post #347

Earlier quoted context omitted.

Aren't you making a general point here against public arguments that certain things lead to better outcomes than others? Do you reserve the same rancor for these little warning labels printed on packs of cigarettes about cancer risk? (Look at these non-smokers looking down on others...) Public debate is, to a first approximation, nothing but arguing about which actions lead to better outcomes. I could see an argument…

I draw a distinction between a particular charitable organization being effective, i.e. the amount spent on overhead vs going to cause they claim to support and a particular charitable cause itself being effective (mosquito nets vs. community theater). For the former I agree, seek out charities that most effectively support what you want to support. For the latter, I don't think it's worthwhile to deride people for s…

I still think there is a fundamental internal inconsistency here. I agree that people will, for instance, hold the moral belief that their community theater is worth more than five foreign lives. However, explicitly espousing this belief is politically unviable. The "common knowledge" moral view is that all humans are of equal moral worth. EA, as I understand it, is largely pursued by people who internalize this view; who recognize the mismatch and as a result, will prefer to sacrifice their community theater rather than admit inconsistency.

Re: The good delusion: has effective altruism broken bad?

#363

Post by kasey_junk »

rhn_mk1 wrote: post #238
tptacek wrote: post #137

My "Effective Altruism Is Not A Cult" t-shirt is prompting a lot of questions already answered by my shirt. I've dug into this stuff over the last couple days, because I had previously understood Effective Altruism to be something akin to like a pledge drive for Charity Navigator's best-reviewed charities or something, and it turns out it's nothing of the sort. In fact, if you read some of the commentary from EA insi…

> Either way: when your logic has established that you can obtain moral superiority by literally buying indulgences, it has officially proven too much, and if you can't recognize that, something has gone terribly wrong in your thinking. I see it as a natural outcome of consequentialism. Your ability to do good things is tied to your circumstances, which are not the same for everyone. Are you saying that you're certai…

The issue with the morality of indulgence isn’t that it recognizes that different circumstances allow for different outcomes, it’s that it allows for immoral behavior so long as the calculus works out.

Evade your taxes, no problem if you gave more away to charities than you owed!

Do a little light billion dollars fraud? So long as you saved the trillions of future humanity from an evil AI it’s fine.

Sacrifice your children to the fires of Kronos? Well if you hadn’t the Romans would have killed us all.

You can argue that literally anything is moral that way.

Re: The good delusion: has effective altruism broken bad?

#364

Post by peteradio »

Devasta wrote: post #344

As an Effective Altruist I want to cure cancer, so I maximize my income by selling cigarettes to children outside playgrounds. The money will eventually be donated to the cause in my will, and in the meantime the prevalence of lung cancer in teens will incentivise the Free Market™ to find a cure. No need to thank me, really.

Wow! Bless this man. Now minimize harm by targeting demographics of children who are less likely to grow up to be effectively altruistic!

Re: The good delusion: has effective altruism broken bad?

#365

Post by nonameiguess »

navane wrote: post #345

Their quality-of-life points combined with long-term-ism reminds me of that simcity build that maximized the population to absurd levels. If you get 'points' for each life, past and future, than maximizing the number of future life is the easiest way to maximize your high score. > Bostrom calculates that 10E29 potential lives are lost for every second that we fail to colonise the supercluster of galaxies containing t…

Seems like Bostrom isn't thinking big enough. If we could invent time travel and seed life earlier in the universe, figure out a way to slow down entropy and star death and make the current universe last longer, or create new universes wholesale, we can get far bigger total life numbers than that. Alternatively, since what we really care about is QALYs, not lives, making even just one person truly immortal gives us a literal infinity that outweighs the value of all mortal sentient creatures that can ever exist.

This also implies every fictional supervillain who ever tried to achieve immortality was the morally best possible person, no matter what evil they performed in the attempt, even if they failed. Since no probability is ever 0, and any small number multiplied by infinity is still infinity, trying to achieve immortality is the maximally good thing any person can do.

Re: The good delusion: has effective altruism broken bad?

#366

Post by medstrom »

Devasta wrote: post #344

As an Effective Altruist I want to cure cancer, so I maximize my income by selling cigarettes to children outside playgrounds. The money will eventually be donated to the cause in my will, and in the meantime the prevalence of lung cancer in teens will incentivise the Free Market™ to find a cure. No need to thank me, really.

You jest, but that would absolutely make sense if you were a psychopath with the singular goal of finding a cure for cancer. Now use this same mindset but change the goal for something saner along the lines of "maximize healthy+happy lifespan for everyone", and then you won't sell cigarettes in playgrounds, since that harms the goal.

Re: The good delusion: has effective altruism broken bad?

#367

Post by coldtea »

ethanbond wrote: post #350
notahacker wrote: post #343

Earlier quoted context omitted.

> Also, incentives, not charisma, are what steer human behavior. The number of people with incentives to work towards destroying the whole of humanity is rather small, and I'd rather we deal with them first since they need not wait for unfriendly AI to be dangerous...

And yet here we all are consuming and pumping oil because… incentives.

That's assuming

(1) "consuming and pumping oil" is a bad thing, and not a complex win-some/lose-some thing, which gave us huge progress and benefits in the 200 years we've used fossil fuels to pump the industrial revolution,

(2) and that we can just absurtly stop consuming and pumping it right now, and there won't be huge disastrous 2nd order effects (in economy, society, ability for people to heat themselves etc) until better replacement technologies are not just available, but have similar EROI AND (even more important) capacity to scale up production, building, and deployment fast enough to cover our needs (which we're nowhere near with any "sustainable" technology, including nuclear).

Re: The good delusion: has effective altruism broken bad?

#368

Post by ineedasername »

Tenoke wrote: post #316
AbrahamParangi wrote: post #312

Earlier quoted context omitted.

You describe all religions. Read some scripture, perform some simple devotions, and call it a day. I don’t mean to overstate the religiosity of EA but it’s not structurally different.

You can reduce everything to being religion if you do it this way to the point of the comparison being useless. HN? Religion. Weight lifting? Religion. Dental hygiene? Religion. Voting? Religion. Functional Programming? Religion. Punk? Religion.

I’m not sure that you know how the “thought leaders” in the upper social tiers of EA are operating. I mean that honestly and not as a slight, because if you do know we can (to the extent that I have the attention span) discuss it if we disagree and productively explore the topic:

There appears to be an upper eschelon, perhaps more of a social group, who are guiding the resources of this movement along philosophical/ideological lines about which they are not being transparent. And when someone outside of that social group verges on exploring those aspects of their philosophical thinking there is an effort to silence or soften the tone of their discourse so it appears less coldly calculating, maybe (per anecdotal reports of people closely tied there) less absurdly extreme and ridiculous.

Re: The good delusion: has effective altruism broken bad?

#369

Post by coldtea »

ethanbond wrote: post #361
coldtea wrote: post #354

Earlier quoted context omitted.

> 1) Why do we have a reason to believe there’s a cap to intelligence? There's a cap to all things. Diminishing returns is a thing. Both physical substrates have limits, and there's a utility limit to thinking or having more data. Limitless intelligence is fairy-tale magical thinking - especially in the way it's portrayed as some magical ability to build and control everything. Like, invent working cold-fusion - as i…

1) Obviously intelligence plays a role in how fast science develops. The quality of experiments and the quality of inference from those experiments are (very obviously) modulated by intelligence. No reason to believe the absolute pinnacle of intelligence happens to sit inside a species of mostly hairless apes here on earth. Lots of reasons to believe fundamental understanding of the universe is more accessible to int…

>Obviously intelligence plays a role in how fast science develops

Sort of. Most science is based on tinkering and experiment, with serendipity thrown in, including most major inventions, and in many cases the actual invention history is changed to make it more like a story of rational thinking process and regular steps than it was.

>No reason to believe the absolute pinnacle of intelligence happens to sit inside a species of mostly hairless apes here on earth.

Note that "hairless apes" is a weasel term, meant to psychologically-pressure the other to feel as if we're talking about some inherently inferior thing ("sounds right, mostly hairless apes can't be the pinnacle of intelligence").

Whereas if you had put it as "the best achievable result after hundreds of millions of years of evolution" it doesn't sound so bad, does it?

> 2) Nope. Could be different hardware. The brain consumes far less power than any human-designed computer that could hope to match its computational power.

Potatoh, potato. Different hardware is still hardware - and will also have limitations of its own.

>4) Meh, this seems to me just the next concession in the line of “we can’t do that, we wouldn’t do that, we do it but it’s not that bad”

Which is an appropriate scalable multi-case response to the similarly multi-level: "AI will control the world through physical access!!!!! SINGULARITY IS COMING! WE'RE ALL DOOMED" which then in practice turns to mean "My smart fridge is connected to Amazon".

>5) Okay so it’s unrelated to this conversation except I suppose some sort of ad hominem?

It's related to the conversation about AGI, since its what one of the most visible proponent of it's dangers wrote as his feared very possible scenario.

Also, if one believes people warning about some subject are delluded, pointing to delluded things they've said about it looks pretty clearly pertinent.

>6) So we agree charisma is at best not necessary to manipulate the world

Sure, in the degree that we agree that intelligence alone is not sufficient.

Re: The good delusion: has effective altruism broken bad?

#370

Post by ethanbond »

coldtea wrote: post #367
ethanbond wrote: post #350

Earlier quoted context omitted.

And yet here we all are consuming and pumping oil because… incentives.

That's assuming (1) "consuming and pumping oil" is a bad thing, and not a complex win-some/lose-some thing, which gave us huge progress and benefits in the 200 years we've used fossil fuels to pump the industrial revolution, (2) and that we can just absurtly stop consuming and pumping it right now, and there won't be huge disastrous 2nd order effects (in economy, society, ability for people to heat themselves etc) un…

It doesn’t assume either of those things. In fact it accounts for exactly those points under the heading “incentives.”

We are engaging in clearly self-destructive behavior because, incentive structure accounted for (your points), it very well appears to be the best option. It wouldn’t be a hard problem if it didn’t.

It should be obvious that it’s entirely possible to follow your (real) incentives all the way into a very bad outcome.

Post reply on HN