Wayne Radinsky

2025-10-30 02:39:56

Every economic system fails in an AI world according to Ruri Ohama. She starts off questioning experts' predictions for the near term, where people think jobs like taxi drivers will be replaced. She thinks lawyers will be replaced sooner than people think. AI lawyers have the potential to be less biased than humans, and they are good at recognizing patterns. Maybe a little too good right now, as current AIs hallucinate, but she thinks future AIs won't have that problem.

She thinks AI will be able to automate not just low-end "junior" software engineers, but the expensive "senior" software engineers.

She thinks AIs are capable of creativity. Creativity is combining existing ideas in an original way.

(As soon as I saw the "Napoleon as a rat" picture, I knew AI could combine existing ideas in an original way and thus were creative. If you asked any human to make a picture of Napoleon as a rat and they came up with the picture Dall-E came up with, you wouldn't hesitate to say they were "creative".)

She thinks the job of managers can be automated by AI. Management is resource allocation. AIs are already better than humans at data analysis. CEOs may still be needed as visionaries, but the management layer can be replaced by AI.

(I think it may take longer than she thinks because she's not taking into account office politics. People will "play politics" to keep their jobs as long as possible.)

She thinks AI will become better than doctors at diagnosis. AIs can be taught millions of drug interactions and symptom combinations.

If AI is cheaper than hiring humans, the likelihood of replacing humans is quite high, but it all depends on what AI excels at.

Even if AI creates new job opportunities, AI learns and adapts faster than humans. So humans learn the new jobs but they disappear quickly, and then human have to learn new jobs again, faster each time. It's as if we humans are playing a game of musical chairs, with the music getting faster and faster and the chairs disappearing.

Gen Z already trust ChatGPT more than human salespeople.

People say the economy will shift to an "experience economy", but she wonders who is going to pay for these human-created "experiences", when the majority of people don't have incomes and don't have purchasing power.

(I know the answer to this: the economy has already shifted significantly towards luxury goods and services for the wealthy instead of consumer goods for the masses, and I expect this trend to accelerate.)

She brings up the subject of universal basic income (UBI). She thinks the people who provide goods and services can simply raise prices to absorb the UBI. She uses the example of landlords and rent.

(She doesn't realize this, but in the San Francisco Bay Area, which is geographically constrained -- bounded by the Pacific Ocean, the San Francisco Bay, and the Diablo mountain range which separates the Bay Area from the central valley -- economists have estimated that essentially all increases in average salary have been absorbed by increased rent, and it's actually the landlords that have made all the money, more or less, from the explosion of economic activity in the area. If you ever move to the Bay Area, make sure you become a landlord. Does this principle apply everywhere? Maybe not as not everywhere is geographically constrained the way the Bay Area is.)

She uses this as a segue into a discussion of socialism vs capitalism and her conclusion that there is no economic system that works in an AI world. With AI automating away labor, it also takes away all the bargaining power workers have, even collectively.

She doesn't think AI will create unlimited "abundance" because there will always be things, such as land and human attention, that are not abundant.

"What if the answer is that there is no answer? Because in a VUCA age which is basically high volatility, uncertainty, ambiguity, error, the future is never in an extension of the past. So no one actually knows what will happen in the future. What if we're heading towards an economic system so different that we can't even conceive of it yet?"

("VUCA" is a military acronym that stands for "volatility, uncertainty, complexity, and ambiguity.")

"No one knows what's going to happen next because of the accelerating pace of the changes that are happening in the world."

"So in my opinion asking the question of which jobs will disappear, which jobs will emerge is actually a pointless question to ask."

"I read bunch of books about the experts' predictions from years ago about AI. Even like two years ago and most of the things that they said were completely wrong. So even experts are failing at trying to predict the future because it's so different than we have ever experienced before."

(Most of the books she read are in Japanese, so you might have trouble reading them yourself.)

#solidstatelife #ai #technologicalunemployment