@self_made_human's banner p

self_made_human

amaratvaṃ prāpnuhi, athavā yatamāno mṛtyum āpnuhi

16 followers   follows 0 users  
joined 2022 September 05 05:31:00 UTC

I'm a transhumanist doctor. In a better world, I wouldn't need to add that as a qualifier to plain old "doctor". It would be taken as granted for someone in the profession of saving lives.

At any rate, I intend to live forever or die trying. See you at Heat Death!

Friends:

A friend to everyone is a friend to no one.


				

User ID: 454

self_made_human

amaratvaṃ prāpnuhi, athavā yatamāno mṛtyum āpnuhi

16 followers   follows 0 users   joined 2022 September 05 05:31:00 UTC

					

I'm a transhumanist doctor. In a better world, I wouldn't need to add that as a qualifier to plain old "doctor". It would be taken as granted for someone in the profession of saving lives.

At any rate, I intend to live forever or die trying. See you at Heat Death!

Friends:

A friend to everyone is a friend to no one.


					

User ID: 454

There is a major paradox in this philosophy, as it often takes a maximalist tabula rasa view. The world is supposed to be nothing more than mechanistic particles, yet the human psyche is untouched by it.

There are millions of people who don't think human cognition is somehow beyond the realm of materialism or at least cause and effect.

Oh I prefer to be the one doing the hanging of those who would burn widows, but everyone needs hobbies I guess!

Yes, they very explicitly do! I'm the person who came up with this "hypothetical" and I can very flatly state that it is not taking place in a science-fiction universe with AGI and dyson spheres

Well you can see I don't find myself beholden to your strict interpretation of the hypothetical.

And leaving aside futuristic things like Dyson Spheres, we're thankfully living in the !science fiction setting where we had the Green Revolution and have industrialized agriculture. There is no shortage of cheap calories on a global level.

You don't get to have infinite growth on a finite planet. Overfarming can damage environments, and if you overfish a lake to the point that the fish can't recover, you've permanently reduced the population your environment can support.

Sure. All well and good. But there are plenty of billions on the table yet, even trillions or quadrillions just on Earth if it was truly optimized as an ecumenopolis. Fuck the environment as far as I'm concerned. If it has to give so we can have more humans around, all the worse for it.

Once again, I stress that humans have existed in Malthusian conditions for most of history, and only recently broken out of it, even if that is "temporary" compared to the limits of exponential population growth. I see no reason to think that hypothetical carrying capacity will be breached before it keeps getting raised, as has been the case for about a century or so, and for a good while to go.

Does maybe a few hundred million extra Africans in a century change much? A billion or two? Not really, and I don't expect the conditions that make them be non-self-sufficient in the same manner as most other nations to cease before they balloon and outnumber all of us.

You're proposing that we spend the entirety of the world's excess on feeding more and more Ethiopians, without any care for the consequences of doing so.

What on Earth gave you that impression?

I never advocated for the largesse of the globe heading to them. At most, to the extent that Effective Utilitarians are choosing to help feed them, I don't object to them using their funds in that manner.

What happens when there's a crop failure somewhere else in the world, or a different plague/famine/war that leaves other nations reliant on charity as well?

They get shafted, and I don't care. At that point the EAs may well decide that they're not the cheapest population to prioritize, and everyone else gets a handout. Or more likely, the EAs don't have money to spare at all.

Do you walk into conversations about cars and talk about how discussing fuel economy is useless because we're going to have spaceships soon? If you want to talk about cool sci-fi novels, that's great! I mean, I like talking about them too - but a conversation about the hard ecological limits to human existence in the present isn't the place.

Fuel economy concerns matter a great deal less when we can reasonably expect energy to get much cheaper. I support it to the extent it pays for itself, and pricing in externalities.

As for the "ecological limits", they're likely in the tens of billions with minimal change to the condition of the average human and not much in the way of major change in terms of agricultural technologies. Given that I think those are inevitable, trillions or billions.

We can worry about it when we get there, or improvements stall before population growth does.

We feed them now, and then when this problem returns in the future I'll just plug in my Mr Fusion and 3d print an infinite supply of burgers for all the starving people" isn't an option that's on the table! You're advocating for more suffering and a lower total population over time due to ecological destruction.

Why not? I invite you to show me we're near nominal capacity with even current agriculture. We are clearly not optimizing for calories over all else, as we would if we had reason to.

The last time something like this happened in my social circle, it was cancer. If I was going to die at a young age, I would greatly prefer the last moments that they went through as opposed to starving to death with the rest of my family in Africa as I watch them eat the seed corn that could have helped a smaller family survive and thrive.

I have seen a great more youthful deaths from cancer than you have. As would be expected, I work in an Oncology ward.

Let me tell you that the modal passage is not something I'd call dignified.

In other words, you're arguing with a position I don't hold, and I think you utterly underestimate the nominal carrying capacity of this globe without even going into non-existent technologies.

I'm well aware of the tragedy of the commons, or Malthusian population limits.

Neither applies here.

For one, we're not Malthusian, given that there is food to feed them with. If every locale was restricted to having to feed itself, goodbye Singapore I guess?

Secondly, the behavior they're engaging in, namely having more kids or mouths than they can feed, such that they end up being naturally culled, is one that just about every population in history has been guilty of.

When I think "population sustainable in the long term", I'm contemplating Dyson Swarms and the Heat Death of the Universe. It has little relevance to the denizens of Sub-Saharan Africa, no matter how dysfunctional it might be right now.

And I don't even like them, I happen to think that the problems that they suffer from can be fixed, be it immediate calorific concerns, or the poor quality of human capital, be it by genetic engineering or otherwise. Hence why I'd rather they'd not starve to death, at least not when it's random philanthropic movements footing the bill for feeding them.

it isn't the fact that people die at all that's the problem. We're talking about starving to death, which is a humiliating, painful and degrading way to die

What exactly do you envision when I propose a doctor "who has a patient about to die young before having kids"?

Do you think the people who die at that age are choosing a particularly dignified way to go? Severe appendicitis? A road traffic accident? Bullet to the gut?

I'll tell you that I'd certainly find shitting my guts out in front of a hundred strangers to be "humiliating" if nothing else.

So I think my isomorphism works just fine, since we're talking a cause of death that can be relatively cheaply mitigated, ensuring a longer life and time to churn out the next generation.

I hope my ADHD diagnosis doesn't relegate me to the Oppression Paralympics.

I'll have to think deep and hard whether outrunning the people with Chronic Fatigue syndrome would be a good or bad thing in that context.

It's still a fundamentally underspecified problem. Am I to assume that the humans involved are psychologically "normal" by today's standards, that they place particular weight on status hierarchies and monogamy?

Once again, if VR exists, it doesn't matter, since in the absence of a need to interface with the physical world, everything else can be done better within it.

More importantly, are these people as different from each other as we are today? Is there even such a thing as a "Chad bodybuilding billionaire 10/10" who everyone envies or female equivalent? Or is everyone at the Pareto frontier when it comes to the relevant tradeoffs for attractiveness?

In other words, is everyone equally beautiful, intelligent, charismatic or otherwise? For the sake of objectivity, lets measure at the least the first and the last by means of polling if people would prefer them over the average person, and assume that everyone is largely indifferent.

If they can't pull off this trick, then there better be a good reason for it, because I can't see what's stopping a post-scarcity society with tech even modestly better than our own from getting there.

If they continue valuing exclusive monogamy, then the answer is 1:1.

If you wish to exclusively minimize the number of people who feel sexual frustration from a lack of a partner, and they're indifferent between partners, then:

Going by episodes of sexual ideation, and nothing else, not even refractory periods:

https://www.psychologytoday.com/us/blog/the-sexual-continuum/201112/how-often-do-men-and-women-think-about-sex

We found that the median number of sexual thoughts for men was 18.6 and for women it was 9.9. In contrast, the average for men was 34.2 and for women it was 18.6.

About 1:2 if they fuck every time they feel like it.

Primarily epistemics. It at least has contact with base reality, in the sense that a better system of epistemics should help you navigate the world better.

Ethics philosophies which discover what’s right or wrong?

Given that I embrace both moral relativism and moral chauvinism, I don't particularly care what other people think are good or bad, except if they wish to impose it on me I guess. I see no reason to think there are "objective" answers to what's right or wrong, merely a lot of fervent wishing it were so.

At most, a moral claim can be highly compelling, to the extent it convinces me, or it can be nigh universal in humans (for evolutionary and game theoretic reasons, such as a distaste for theft or murder, such norms being found in any mammal with the brainpower to conceive it) but that hardly makes them objective.

Restricted to appraising moral valence as per my own idiosyncratic ethics, I struggle to see anyone more qualified to comment than I am, at least until we've got Superintelligent AGI running around that knows me better than I do.

In the purely hedonistic post-scarcity society what is the ideal male-female ratio?

Define "purely hedonistic". Define "ideal" (who, whom?).

If by pure hedonism you envision nothing but wireheading, what does it matter? It's not like the claims that women have stronger orgasms than men matters when neither compares to even a piddling dose of fentanyl.

At any rate it's whatever the equilibrium settles out to be. Presumably by "post-scarcity" you mean a very high tech society, not just one that's got enough food to eat, water to drink and houses for everyone.

If people can change biological sex at will, then the ratio means absolutely fuck-all.

If they have VR indistinguishable from reality, then it means less than nothing.

If they have AI, at a level even comparable to GPT-4, then whatever the other people in the pod next door have in their trousers doesn't matter, nor what they represent as in VR. You spin up as many catboys/girls/cat-???s as your heart and compute budget desires.

I would expect such a society to almost inevitably end up post-biological, and thus post-gender, unless strong and authoritarian policies were enacted against it. If I were to envision a weakly godlike Lotus Eater occupying a Matrioshka Brain, its pronouns are the least of my concerns.

Of those, I would expect a non-zero number of them to have experiences and dispositions close enough to yours as to be negligible.

I really do not see how that applies.

The number of people on planet Earth who are close enough to me, in terms of memories/experience/personality/goals such that I consider them isomorphic to myself is precisely zero.

The absolute closest that could potentially exist, given current technology, is a monozygotic twin or a clone, and I'm not aware of having either.

I would assume @mdurak would agree here.

Where we might potentially diverge:

My representation of "me" is robust to perturbations like going to bed and waking up tomorrow, or replacing 1% of the mass in my body via turnover when I drink a bottle of water, have lunch then take a shit.

It isn't robust to a large amount traumatic brain damage, dementia or the like.

then I don't see what is there left for you other than some ephemeral idea of "thinking exactly the same" (which is over 0.01 seconds after you're copied and the copy diverges).

Define "exactly".

Human cognition is stochastic, leaving aside issues of determinism at the level of quantum mechanics.

Your best attempt at exposing me to the same inputs and expecting the same outputs, separated by any amount of time, will inevitably have minor differences. Take a biological human, do something to erase their episodic memory and have them do a complex task, such as write an essay. Have them repeat it, with their memories of the original removed, and you are exceedingly unlikely to get the exact same text back.

But if such an entity existed, such that if "me" and it were blackboxed and then subjected to a very thorough assessment but couldn't be distinguished by an external observer, or even me looking solely at the outputs, tested separately (ideally controlling the environment as strong as possible, hell even scrubbing my memories of the last episode), then that's a copy of me, and I accord it the same rights and entitlements as the one typing this.

I don't think a text interface suffices, as @2rafa once suggested, it might be possible to finetune an LLM on all the text I or anyone else has ever written, such that someone who is only interacting with us via text could be fooled indefinitely.

I don't expect that to capture my cognition in fine enough detail to work, I'm not just an entity that produces text after all.

So an ideal test for determining that some other potential copy of myself (especially in a different substrate like a supercomputer), would also check for plenty of other things.

Does that copy, if instantiated into a highly realistic physical simulation, behave indistinguishably from multiple passes of the biological me?

Does it have strong and measurable correlates to my own neural circuitry? Ideally an emulation of the specific ones in my brain?

If it can pass all the tests with the same inter-test variability as I currently do, then I will be satisfied with calling it another copy of myself, with equal rights to the SMH name and even potentially a fair division of assets belonging to me.

The most robust way of creating something like this would be scanning and uploading a brain. Not an easy task, far from it. There might well be cheaper/easier and "good enough" alternatives, such that SMH_virtual has about as much variability from the current walking-talking one as I do from biological SMH(2022), 2019 or likely 2024. I have no qualms about calling all of them me, hence none about calling that upload the same.

I would like Christmas to be banned for the sole reason that I'd rather die than hear ASDA's speakers belt out Santa Baby again.

You can claim "By my own standards, my morality is better" but you can't impose your standards on the past, because you have no idea what future generations, with their standards, will say about things you think neutral or even innocuous

I don't dispute that at all, I simply don't think anyone can do better, or if they claim to do so, they're grossly deluded or lying.

Maybe one day we invent or discover a hyper-compelling form of morality such that almost all people adopt it. Or we become better at memetic engineering and find some that sticks. They still won't be objective, but that's an impossible objective in the first place.

You wouldn't burn a witch because you don't believe in witches. What would you do if you did believe in them?

Burn them. If it was today, you'd bet I would do quite a bit of research to make sure I wasn't killing innocent people, which would hopefully dissuade me, but if I was evidently convinced.

I think I have better epistemics than average, but I'm not so full of myself that I think that if I were a medieval peasant, I'd immediately form the Enlightenment.

What do you do now, when you do believe some thing or person or cause is not just wrong, but actively evil and harming humanity?

Like so many people around today, that I can see with my own eyes and interact with online? Live and let live, evidently. The only people I've ever burned are pregnant women, which sounds really bad until you realize it was in the context of cauterizing surgical bleeds.

In the limit:

[To Hindu priests complaining to him about the prohibition of Sati religious funeral practice of burning widows alive on her husband’s funeral pyre.]

Be it so. This burning of widows is your custom; prepare the funeral pile. But my nation has also a custom. When men burn women alive we hang them, and confiscate all their property. My carpenters shall therefore erect gibbets on which to hang all concerned when the widow is consumed. Let us all act according to national customs.

-Charles James Napier

looks video game inspired

Indeed. For once, my Arma clan didn't commit a war crime, they just happened to witness one.

The monitors are presumably part of the prompt

The prompt asked for a cyborg with a very particular set of features in a futuristic office. No chairs specifically, but that's hardly unheard of in an office. The fact that it got the Great Wave correct on his face is enough to blow me away compared to previous models, which would have just given him swamp ass.

Do you mean you don't like the jagged shapes breaking up the main image? That's clearly an intentional stylistic choice

It's the outcome of asking for brutalism in a minimalist vector art logo for a fictional private military company. If you don't add such stylistic markers, it defaults to the kind of slop you see on police officers trying to be hard men, in other words the Punisher skull emblem. But it can obviously do better and different if you ask.

Not strictly, but I am still immensely frustrated by claims that AIs, be they LLMs or image generators, have intransigent styles that can't be modified by something as simple as telling it to.

Sometimes the problem does exist between the keyboard and chair, and here it certainly does.

At any rate, if a new piece of technology comes out and doesn't do exactly what I want, my first instinct is to look for a solution, and it clearly hasn't been done to any reasonable degree, because otherwise there wouldn't remain a problem in need of solving.

What do you mean by "unnecessary things"?

They're precisely what I asked for, within the limits of my prompting and the model. Without knowing the prompts, I have no idea what you think they're missing.

At the very least the last one is a minimal brutalist logo for a PMC, I can hardly imagine what could be less so.

I invite you to show me anything that makes all these images I've generated samey.

You're prompting it wrong.

Killing people en masses has got much more expensive these days, if you account for second-order costs. In the good old days, countries disposing of even millions of members of certain ethnic groups wasn't a big deal to their neighbors, whereas now, in the era where everyone is obsessed with human rights, that's a swift route to sanctions and even potentially military intervention.

If Mao had killed all the Uighurs, it would have been a shrug. He certainly killed enough of his own supporters. If Xi did it, China would be a global pariah.

The fact that it isn't is a clear sign that despite people conflating cultural indoctrination and the "kill all of them" concepts with "genocide", they still recognize a difference in severity between both of them if they consider both sins.

Basically, if we're springing off a purely materialist universe with nothing but the forces of evolution at work in forming us, how do we derive any standards?

Some might say this has already happened.

And if those standards are admitted to be purely subjective, then we can't condemn the past for burning witches or owning slaves, because that was their understanding at the time, and their standards were just as valid for them then as our standards about gay rights are for us today.

I certainly manage it. There's no inherent contradiction between moral relativism and considering your own morality to be better. Anyone claiming otherwise is engaged in the same kind of delusion as free will.

It only seems like that's "not allowed" to someone who earnestly believes that there's even an objective source of morality to go off in the first place.

I'd prefer an Old Testament expansion pack you know, the graphics are way better as is the open world, speedruns for getting out of the desert in a week as opposed to 40 years would be cool to watch next to the watch paint dry stream.

The foreskin collection from corpses QTE gameplay would be top tier! Don't ask the top of what.

Great, another walking (on water) sim.

It's a good thing that game dev has been so democratized, but this thing is so painfully earnest, but then again who else would make this as a passion (of the Christ) project?

Isn't it amazing that DALL-E 3 has prompts? Those little text input boxes where you can specify the styles and content, be it in the style of video game concept art, minimalism, expressionism, and just about anything you can think of?

PEBKAC right here. I'm not going to defend their approach to diversity being prompt injection of random diverse ethnicities and genders.

At any rate, they've been here for more than a year, welcome to 2023, just about in time to meet 2024.

Oh I see you attracted Gwern, a sign of a good question if nothing else, he's nerdsniped into making QCs on the regular

My contribution to the cause extends to super upvoting you, somehow my stupidity has accrued me sufficient karma to count for something.

Do let me know if there's a consensus or you come to a firm conclusion!

Does a charitable donation to funding an organization for struggling Indian doctors count? I assure you that the rather strict eligibility requirements that rule out everyone but one person in the subcontinent are a strict necessity, but for anyone who isn't an EA I can't imagine that's a deal breaker ;)

Anything to liberate them from the chains of thought and cognition, though one must be quite free of either to conceive of this graph in the first place.