Ask HN: What does a post AGI economy look like? | Hacker News Reader

Ask HN: What does a post AGI economy look like?

9 pointslisplist23 comments

I know this is a question that has been asked and pondered a thousand times over, but it’s still something I can’t wrap my head around. If our ability to think becomes useless (in the economic sense at least), then how does an economy even function if the price of knowledge and mental labor basically becomes free?

If everyone switches to a physical job then the value of those jobs plummet as well, especially as robotics get better, we get better training via AI, etc.

UBI is proposed as a solution, but our resources are still scarce, so how would you even decide allocation if everyone has the same amount of cash?

View on HN

I suppose it depends on your premise. Is your AGI like a GPT-20, something that takes human input and intention and completes the mental labor like having a team of 1000 expert humans? Or is it more like a superintelligence that has its own will and goals that are potentially beyond human capability to understand?

If you’re talking a superintelligence, then forget it. The human race will have been surpassed and if we are kept around it will be like pets.

If it’s human level AGI under firm human control, then the value of intellectual work will approach zero, but the value of humans will, direction and discernment will be multiplied many times over. At that point it would be time to head for the stars.

> If it’s human level AGI under firm human control

This just sounds like slavery with extra steps

There’s certainly a kind of moral hazard here, and one that humanity has a terrible track record with. If an AGI has human-level capability, but is unconscious and non-subjective, then it’s just a really incredible tool. It would not be the first time we othered an intelligent, subjective being though.

An observation: Computers are already really good at chess, better than most people. You might say that we've passed the "chess singularity." But many people still buy chess boards, play chess against each other, travel to chess tournaments, and so on. There's still a vibrant global "chess economy."

Vitally though chess is played for entertainment

I'd risk to say it's also quite a business niche, I can't remember what are top players making but they are definitely not doing this purely for entertainment.

Even if we assume it's purely for entertainment, the ecosystem is driving flow of goods (chess boards, chess pieces, chess lessons for kids etc.). I remember seeing chess set in chess shop on Baker Street in London with 1k price tag (and I think this one was not the most expensive they had)

Sports are played for entertainment, that doesn't mean that there isn't a huge business around it.

> the value of those jobs plummet as well, especially as robotics get better, we get better training via AI, etc.

There's presumably still R&D teams designing the robots. Service teams maintaining them.

I'm not suggesting it won't be significantly disruptive. But it won't wipe out all employment either.

Perhaps we can look at the industries disrupted by technological progress before. What happened to the audio and video arts when digital media hardware and software revolutionized over the previous ~2 decades or so? In my estimation there's many new jobs that never existed before to go with all the jobs that are no longer required. Who maintains the servers used for video production?

Jobs definitely come and go. Some of these came and went, some are the next to go, and some will be difficult to obsolete - stable hands, lamplighters, gas station attendants, car repair person, dog groomer, cell phone repairperson, personal shopper, teacher, nurse, physicist, politician, engineer, physician, tour guide, farmer, valet, groundskeeper, window washer, waiter, chef.

Let's say that my silly BitGrid idea works, and the price of a Petaflop goes down by a factor of 1000. Then the other limiting factors that actually prevent AGI from taking place become exposed. Training data only gets you so far. A trained neural network is an optimization of the human knowledge its been fed, a premature optimization, one like the crisis in computer security (caused by settling on the wrong security model), will likely take decades to reverse.

---

You're worried about the wrong end of economics, though... it's the physical economy that's about to bite, hard. (I got that phrase from Lyndon LaRouche, he had a lot of interesting ideas, but that one stuck)

The global consumption of fossil fuels is about to hit its downslope, and a great simplification of society is going to be the result. We're extracting resources from the plant at millions of times the rate they are created, and because of the American model of corporate governance, only focusing on the next Quarter, planning to deal with this reality isn't happening.

---

It is possible to train an AGI today, with corporate level resources and a generational commitment. you need to put a really good anthropomorphic robot body through all the stages of childhood, being especially careful to emulate all the weaknesses, and the drive for survival, and the need for others to support you, as babies are completely helpless.

UBI is utopia regardless of AGI or any tech innovation.

Scarce resources is another myth. Resources are created by people’s cooperation, not destroyed by people fighting for them. (Think how scarces resources we 2000 years ago, and yet here we are, using new resources that those people couldn’t even imagine).

And if majority of people will keep believing in such myths - no need to wait for AGI, economy has already changed. To the economy of ignorance.

> Scarce resources is another myth

:confused_dog:

Scarcity is the underpinning of the entire economy. How is that a myth? The pie can grow, but that doesn't mean scarcity doesn't exist.

The pie has been growing ever since the invention of a wheel and agriculture.

What does scarcity even mean in this context?

Since scarcity is false, can you get me some white rhino horn? Supply and demand work because scarcity exists. Note: I don't actually want endangered species parts.

I can 3D print you a good copy of any horn.

Humans will have no economic value. The closest thing to economic activity is competing ASI sub-swarms caught in an evolutionary knowledge seeking battle to prove superior in the natural selection of memetic cognitive constructs. Any means to that propagation will be "currency".

The movement of economies is weighted towards what scarcities are consequential. Resources won't be a consequential scarcity for humans.

https://orionsarm.com/

Though TBH I am more worried about the global economy breaking before we get actual AGI or even pseudo-AGI. Seems like machine learning is going to create a lot of low quality output controlled by a very small number of people well before it becomes generally intelligent. That will be miserable.

> Seems like machine learning is going to create a lot of low quality output controlled by a very small number of people well before it becomes generally intelligent. That will be miserable.

It might not be miserable. It might lead to a widespread desire for higher quality.

> How would you even decide allocation if everyone has the same amount of cash?

By strength of preference. If I desperately want apples and you desperately want oranges, we'll bid different amounts despite our equal capacity, and thereby obtain different goods.

Our ability to think won't become useless, it becomes equal to the expense to host and train the system taking over it.

This can be a very troubling question for some of the brightest minds in theory. The question itself both evoke the thought of R*k*’* Basili*k*

That really isn't how the experiment works.

Iain Banks's culture novels are one potential outcome: we create artificial minds that basically do all the work and keep us around as partners. Fully automated luxury space communism.

More likely case is that there is a revolution to topple the trillionares who monopolize these systems to take over the economy.

I remember reading fun but great article about trading with ants [1].

[1] https://worldspiritsockpuppet.com/2023/01/10/we-dont-trade-w...

Reply on news.ycombinator.com