The good delusion: has effective altruism broken bad? • Hacker News Board

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

#511

Post by naasking » Wed, Nov 16, 2022, 6:19 PM UTC

bootsmann wrote: post #261

concordDance wrote: post #225

Earlier quoted context omitted.

> 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. As someone with multiple EA friends: this is precisely what it is. > they have their own weird language, there's a hierarchy, and people on the inside operate und…

> If AGI has even a 1% chance of happening this century and those donations reduce the risk of it going wrong and killing everyone by 1% that's still money very well spent. I have yet to see anyone who argues this point show sufficient knowledge of the current state of AI research to credibly make it. If you live in the fantasy world where AI is just an arbitrary construct that can do whatever task by itself it is an…

> If you live in the fantasy world where AI is just an arbitrary construct that can do whatever task by itself it is an easy point to make but anyone who is down and getting their hands dirty in the pile of linear algebra that makes up current AI knows that there is not even a credible pathway to creating an AGI in current research (i.e. you cannot optimize for a task you don't know at train time and if you are optimizing for a specific task you are not creating a general AI).

Humans are optimized for the specific task of reproducing, and yet we evolved general intelligence. "General intelligence" doesn't have to be a goal for it to emerge.

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

#512

Post by rcoveson » Wed, Nov 16, 2022, 6:19 PM UTC

tptacek wrote: post #435

rcoveson wrote: post #241

Earlier quoted context omitted.

Are you describing "giving virtually all of [your earnings] away to effective charities" as "literally buying indulgences"? That's literally the most cynical thing I've ever heard. I can understand having a negative opinion of the EA movement as it exists today. But your reaction to the charitable investor thing makes me think that you actually disagree with... altruism. Or at least the idea that being more altruisti…

No, I pasted a quote from an extremely respected thinker who, in the context of Effective Altruism, said that a financier who gave away virtually all his earnings would be morally superior to most of us. Giving away money is a good thing, everybody does it, by all means do it more. But the idea that you can purchase moral superiority, and, moreover, that the superiority you obtain is proportionate to the amount you d…

It’s not the amount you donate relative to others, it’s the amount you donate relative to how much you are capable of donating. Not many people would work a high-stress, high-barrier-to-entry job and give away almost everything they earned, while their peers live richly.

Even if we accept that the average person usually acts morally when confronted with an obvious moral choice, they still fall morally behind someone who actively exerts themselves to the limit of their ability for the benefit of needy people who they’ve never even met.

Maybe we just disagree about where the “average person” is morally? Or maybe you’re ascribing selfish intent to selfless action because the hypothetical person works in finance and/or makes lots of money?

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

#513

Post by bioemerl » Wed, Nov 16, 2022, 6:23 PM UTC

afiori wrote: post #404

bioemerl wrote: post #397

Earlier quoted context omitted.

We are already made of nanobots, we are the nanobots, we are in intelligence, we are a general intelligence, and we are growing exponentially. AGI is here. AGI is you. You are an incredibly efficient piece of carbon thinkware able to sell for replicate into more carbon thinkware and that that extra carbon think where can go out and create its own life and make its own observations and spread information through the s…

Machines being simpler is the problem. The problems with AGI are 1) runaway growth and the singularity and 2) the scale and reach of its actions. Humans are deeply limited in their senses, their actions, and their ability to communicate. An AGI botnet installed on every computer/control system in the world would have none of those limitations.

Complexity breeds ability. Our computers will have to continue to get more complex in order for AI to do the same.

What humans have is scale. There are 10 billion devices connected to the internet. Cool, right? More than humans? No. Being human has really big perks.

We are mobile.

We have hands and feet.

We have eyes and touch and smell and we use tools to expand what we can see.

We each possess the ability to process data and filter data at scales many magnitudes greater than a computer. Note that data in this case isn't "many miles of spreadsheet data" - data in this case is worldly observation. You notice weird things, you report it to your friends. You try weird things, you report it to your friends.

Communication between humans is limited, as is our memory, but it's limited for a reason. 99 percent of information is worthless junk human communication gets most of the important parts through.

Then we need to parse mass amounts of data? Computers do that.

Comparing your information to all other known information? It's an O n squared problem. You can throw hardware at it but as informations scope grows it will outpace the hardware eventually.

Humans are crazy advanced compared to machines in basically every sense. I think the viewpoint here that machines will have superior senses is extrapolating how computers work today and is thoroughly impractical. We will find these logical and reproducible systems look more and more like we do as we push the line.

I expect eventual marginal improvements that can't outpace human "plentifulness", and biological modifications and creations to be our real future.

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

#514

Post by naasking » Wed, Nov 16, 2022, 6:24 PM UTC

tptacek wrote: post #433

concordDance wrote: post #225

Earlier quoted context omitted.

> 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. As someone with multiple EA friends: this is precisely what it is. > they have their own weird language, there's a hierarchy, and people on the inside operate und…

It's easy to justify almost any sort of behavior if you take some risk and assign it a severity of "infinity". A 1% chance of happening? At severity infinity, you're morally obligated to act at even a 0.00001% chance: 0.00001% times infinity is still infinity . That's the AGI logic I'm seeing from these people.

No, that's a strawman. The severity of human extinction is not infinite. For instance, it would be preferable for humanity to go extinct than for every human to be tortured for every minute of their life, and since there is no number larger than infinity, the severity of human extinction must be finite.

This tells me that you haven't seriously engaged with the arguments over AGI. I'm sure there's plenty of nonsense armchair philosophising happening, but you don't take all the "quantum woo" that's out there as some indictment of real quantum mechanics, do you?

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

#515

Post by tptacek » Wed, Nov 16, 2022, 6:24 PM UTC

rcoveson wrote: post #512

tptacek wrote: post #435

Earlier quoted context omitted.

No, I pasted a quote from an extremely respected thinker who, in the context of Effective Altruism, said that a financier who gave away virtually all his earnings would be morally superior to most of us. Giving away money is a good thing, everybody does it, by all means do it more. But the idea that you can purchase moral superiority, and, moreover, that the superiority you obtain is proportionate to the amount you d…

It’s not the amount you donate relative to others, it’s the amount you donate relative to how much you are capable of donating. Not many people would work a high-stress, high-barrier-to-entry job and give away almost everything they earned, while their peers live richly. Even if we accept that the average person usually acts morally when confronted with an obvious moral choice, they still fall morally behind someone…

That is not what Scott Aaronson said. It's what you might believe; fair enough! Sacrificial giving is a big thing in the religion I belong to, too! Aaronson is making a point that minmaxing absolute donations (even, in this specific case, if it means defrauding thousands of customers) is a pathway towards moral superiority.

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

#516

Post by tptacek » Wed, Nov 16, 2022, 6:25 PM UTC

naasking wrote: post #514

tptacek wrote: post #433

Earlier quoted context omitted.

It's easy to justify almost any sort of behavior if you take some risk and assign it a severity of "infinity". A 1% chance of happening? At severity infinity, you're morally obligated to act at even a 0.00001% chance: 0.00001% times infinity is still infinity . That's the AGI logic I'm seeing from these people.

No, that's a strawman. The severity of human extinction is not infinite. For instance, it would be preferable for humanity to go extinct than for every human to be tortured for every minute of their life, and since there is no number larger than infinity, the severity of human extinction must be finite. This tells me that you haven't seriously engaged with the arguments over AGI. I'm sure there's plenty of nonsense a…

You can replace "infinity" with any suitably huge number; I don't need to quibble over tenths, hundredths, and thousandths of a percentile risk that something is going to happen.

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

#517

Post by c1ccccc1 » Wed, Nov 16, 2022, 6:25 PM UTC

MichaelCollins wrote: post #49

emmanueloga_ wrote: post #18

Earlier quoted context omitted.

Effective altruism (EA) is a philosophical and social movement that advocates "using evidence and reason to figure out how to benefit others as much as possible, and taking action on that basis". [1] Seems like a good idea, if only just to have some sort of open-system for rating charities. I suspect the issue is everything that comes around the idea, specially... people. The problem is always people :-). Perhaps a "…

The problem with EA and other utilitarian systems of thought is two-fold. Firstly, you can't predict the future so really you're working off your biases. Secondly, unfettered by principles other than the end justifying the means, you can easily turn those aforementioned biases into justifications for committing horrible acts. Mass murder, or evidently, theft on a massive scale. For instance: "The technological revolu…

I agree with your second point that having deontological boundaries is important. Interestingly, people like Eliezer Yudkowsky (generally respected in the EA movement, I'm pretty sure?) also agree with you about that. He's been reiterating the point recently on twitter, though this post [1] says pretty much the same thing and it's from 2008. Generally, I'd attribute your examples of "utilitarianism gone wrong" to mental illness or sociopathy rather than to holding a utilitarian philosophy. For the average person, moral philosophy is much more of a way to explain the choices we make after the fact than a thing that actually determines the choices.

[1] https://www.lesswrong.com/posts/K9ZaZXDnL3SEmYZqB/ends-don-t...

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

#518

Post by naasking » Wed, Nov 16, 2022, 6:28 PM UTC

tptacek wrote: post #478

edanm wrote: post #474

Earlier quoted context omitted.

> Your problem is that you're using the concept of incomplete information, rather than that of ceteris paribus. You're right, that's what I meant and I misspoke. > The logic is shot completely to shit as soon as we're trying to compare the marginal sacrifice of a billionaire's giving pledge to the moral decisions people make every day of their life. And yet that's exactly what Scott Aaronson does. Well I think that h…

Consequentialism says that the ends justify the means. They do not.

Last I checked, no one has proven this.

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

#519

Post by zajio1am » Wed, Nov 16, 2022, 6:29 PM UTC

skinnymuch wrote: post #413

edanm wrote: post #400

Earlier quoted context omitted.

> This is the part of EA that triggers a lot of red flags for me. Making money, or rather making significant money, necessitates participation in a system of exploitation that is largely responsible for the ills the charitable donations are meant to alleviate in the first place. I mean, given that you think that I understand where you're coming from. But just note that many, many people would highly disagree with tha…

You missed the major point of being the exact capitalists that are the reason for charity to be so necessary. That’s white washing greed and self interest. Get to have a better life selfishly then donate the money to be morally better.

Well, without capitalism everyone would live in subsistence-level poverty, but it would be a social norm, so no charity would be necessary.

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

#520

Post by naasking » Wed, Nov 16, 2022, 6:30 PM UTC

froggychairs wrote: post #218

devilsbabe wrote: post #193

Earlier quoted context omitted.

re your last sentence: why do you disapprove of funding research into AI safety

Not OP, but I do work in reducing harms of deployed AI systems at a big tech company When most EA/Rationalist folks discuss AI Safety. They’re talking about AGI, a hypothetical construct that we have no way of proving is even possible Now there’s nothing inherently wrong with funding research of that sort. The problem comes is that many EA/Rationalist folks begin downplaying legitimate tangible risks on the horizon (…

> They’re talking about AGI, a hypothetical construct that we have no way of proving is even possible

I really don't understand AGI skeptics. Unless you think reality has some non-physical character that can't be reproduced outside of biology, it obviously follows that a physical machine can reproduce what another physical machine (human body) is doing.