The biggest issue with 2027 was that it didn't understand the economy.
For AI2027 to be real, the money has to come from somewhere to carry on building the economy. If >10% of the workers suddenly become unemployed, and the rest taking paycuts, then money supply dries up. (unless central banks do something, but then that can be highly inflationary)
Without massive amounts of investment, AI development stops dead.
In this post, they hand wave about the USA being able to acutally 1) build concensus locally for regulation and 2) the rest of the world actually follows suit.
It fails to understand that actually the progress of AI is not actually the gift of the USA. It requires a constant supply of things from china.
Also its assuming that having 74 billion agents doesn't cause economic distortion. Like what value are these agents generating that justifies them being run?
I really wish people would just ignore this for what it is: bad sci-fi with an incomplete world.
Which predicts that explosive growth of robot production will lead to problems such as
> a deflationary debt spiral, where the AI and robot companies can’t pay back loans in dollars because the robots and AIs are worth nominally less than the loans written the year before.
In other words, the companies go bankrupt because they produced an oversupply of cheap goods, the bubble pops, and there's less new investment for a while. Plenty of precedent for such a development.
But instead of adjusting their predicted output growth downwards accordingly, they instead propose that
> One way to solve this could be for the loans to be denominated in AI and robots, so the companies pay back the loans with some percentage of the AI and robots instead of dollars.
Try doing this today with a battery factory for example. You expect that battery prices will fall to the point where the revenue from selling batteries won't ever cover the cost of building the factory. So you propose to a bank that they'll be the ones to build the factory, and you'll borrow it from them (not paying rent?), make your batteries, then give back the factory when you're done. All the profit is yours, all the risk is theirs! Which is of course why a real bank won't agree to this, all you're going to get is a dollar loan with the factory as collateral.
> Without massive amounts of investment, AI development stops dead.
Would the US government not pour enormous resources in AI labs if needed, knowing that China might be doing the same? What happens if an adversary develops an AI capable of finding and implementing exploits in every software run by your country's strategic infrastructure?
> For AI2027 to be real, the money has to come from somewhere to carry on building the economy. If >10% of the workers suddenly become unemployed, and the rest taking paycuts, then money supply dries up. (unless central banks do something, but then that can be highly inflationary)
The economy doesn't need workers as consumers necessarily. It would of course be a huge shock to the economy but the economy could adjust to it eventually. Maybe it compromises the 2040 timeline. Still, billionaires are increasingly holding the assets. The money supply drying up can be countered, as it is controlled by central banks and can be arbitrarily increased.
> as it is controlled by central banks and can be arbitrarily increased.
Its really not controlled by central banks. Its influenced, but not controlled.
When central banks "print" money, they effectively just add money to the accounts of investment banks
But investment banks are also "printing" money. Double accounting effectively uses assets to double the available pool of money. If you then sell off those loans based on those assets, then you crystallise that new money. Investment banks are inflationary.
For AI2027 to be real, the money has to come from somewhere to carry on building the economy. If >10% of the workers suddenly become unemployed, and the rest taking paycuts, then money supply dries up. (unless central banks do something, but then that can be highly inflationary)
Without massive amounts of investment, AI development stops dead.
In this post, they hand wave about the USA being able to acutally 1) build concensus locally for regulation and 2) the rest of the world actually follows suit.
It fails to understand that actually the progress of AI is not actually the gift of the USA. It requires a constant supply of things from china.
Also its assuming that having 74 billion agents doesn't cause economic distortion. Like what value are these agents generating that justifies them being run?
I really wish people would just ignore this for what it is: bad sci-fi with an incomplete world.