“A welfare state and free immigration will obviously bankrupt any country,” the Tesla and xAI chief wrote on X on Sept. 14, adding that “AI + robotics is the only path to universal high income” for every person on Earth.
Will someone please tell him that Universal High Income is, by definition, a welfare state.
He warned in April that once AI and robots flood the economy with output, governments must “issue dollars to people” or trigger massive disinflation.
Who is going to pay for that UHI without MASSIVE taxes - especially if fewer are working due to robots.
That fact that at least 30% of the average middle class family’s budget is the result of for-profit “skim, scam and fraud” disturbs me, but at least I’m able to avoid most of it in my own daily life.
Have you watched “The Expanse”? There’s a lot of space adventure. But there are two details that really caught my attention. First, the sea walls around Liberty Island and Manhattan to deal with sea level rise. They don’t tell you that, but you can see it in every aerial shot.
Second, 80% of the population is on “basic assistance” because there isn’t anything for most of the population to do. This future society has so much automated, that only 20% of the population has a job (including, apparently, military and government).
I am quite convinced this is our future (assuming we survive to see it). Both sea level rise and the basic assistance. They didn’t comment on the tax structure of that society. However, just to maintain a relative peace among the populace, you will absolutely need basic assistance (hungry people get really angry and violent). Which means drastically higher taxes on the people that are working and the people with all the money (like Musk). And, probably, the companies that “employ” robots instead of humans to pay for the humans that aren’t working.
People have been predicting that for quite a while. I was immediately put in mind of John Maynard Keynes, the famous economist, who predicted back in 1930 that within a century that so much of our economy would be automated that we would see working hours similarly reduced:
We are being afflicted with a new disease of which some readers may not yet have heard the name, but of which they will hear a great deal in the years to come–namely, technological unemployment. This means unemployment due to our discovery of means of economising the use of labour outrunning the pace at which we can find new uses for labour
* * * * *
For many ages to come … everybody will need to do some work if he is to be contented. We shall do more things for ourselves than is usual with the rich today, only too glad to have small duties and tasks and routines. But beyond this, we shall endeavour to spread the bread thin on the butter—to make what work there is still to be done to be as widely shared as possible. Three-hour shifts or a fifteen-hour week may put off the problem for a great while.
A famously wrong prediction. He thought we’d have “solved” the need for work, and resort to those kind of make-work proposals to satisfy psychological needs, by 2030. Which obviously isn’t happening.
New technology creates more opportunities for work. Jobs we can’t even imagine existing today. Just like the internet didn’t kill jobs. It didn’t even actually kill retail jobs. People were convinced that the internet would hollow out employment in the retail sector, but it turns out that e-commerce generated massive numbers of new jobs. Just those folks were doing very different things than the old retail employees.
`But he was almost right. ChatGPT tells me that U.S. real GDP per capita increased about 6.38-fold from 1929 to 2019.
So a worker earning an inflation-adjusted $40K/year in 1929 would be earnings $255,000 in 2019 ($40K * 6.8 = $255K).
Or a worker working 40 hours a week in 1929 would be as productive as a worker working 6.2 hours a week in 2019. If she worked 15 hours a week, she’d have roughly double the income.
Keynes’ first mistake was assuming capitalists would share the productivity gains with the workers. The second was Javon’s paradox. If workers are more productive, you want to hire more of them.
Well, there is still 4 years to go and we are getting closer. And maybe he was off by 10 or 20 years. It is hard predicting the future. My concern is the types of work that will be available. Remember, half the population has an IQ of less than a 100! Where are they going to be working?
I think he was just early. A lot of stories from the past haven’t materialized yet, and a lot have. We’ve already passed “Blade Runner” (2019, as I recall). “The Terminator” was 2029, I think. Clearly, none of this is happening. Arguably, Jules Verne nailed nuclear submarines. That Keynes’ predictions haven’t YET come true, doesn’t mean they won’t. At least some of them.
Technology is improving to the point that machines are going to be able to do whatever humans can do. Even if new jobs are created, at some point the machines will be able to fill those positions, too.
Yes, Keynes predicted we’d only have to work 15 hours/week in his 1930 essay.
People may be only working that little today, but they still have to show up in the office for 40’hours/week, plus, whether they have anything productive to do, or not.
You’re talking Skynet? As long as we keep the really big weapons air-gapped under human control, the only worry would be AI falsifying orders to those humans.
One thing that would stop it is quantum computing. Secure the comms and big weapons with quantum computing, and they are (as I once heard it said) “guaranteed by the laws of physics”. Just don’t run an AI on the quantum computers.
If you’re referring to the “useless” population, that is probably in the ballpark once automation and AI are doing most things. Which will require some sort of government assistance, which will have to be paid by taxes on the wealthy, and corporations that use robotics and AI.
Probably not in my lifetime, but we really should start planning for it now. So we’re ready when it happens.
How do you determine what the “wage” to tax is for each robot? Does the robot that does home health care get taxed on $35,000 income, and does the robot that does surgical procedures get taxed at $450,000 income? What if the very fact of robots doing surgery reduces the equivalent human cost of that labor from $450,000 to $35,000?
What if robots doing “everything” reduces the equivalent “wage” to $5000? Or if everything is self-replicating, reduces it to near zero?
Perhaps his second mistake was not realizing how much governing ourselves will consume in the future. In 1929, government (all levels) consumed 11.29% of GDP, by 2019 government consumed 35.7% of GDP.
This assumes that the “consumption” is unproductive. In 1929 we had 3 million miles of roads, most of it within urban areas and very little intra-city or interstate highway. We now have 42 million miles of roads, endless ribbons of asphalt connecting every city, village and hamlet with the ability to deliver goods and services, and to visit Grandma at a moment’s notice.
Along with the asphalt there are bridges, there is electricity to run the stop lights, there is the maintenance and repainting and improving of intersections. - And that’s just “roads.”
Like to do “airports”? (It’s from 800 then to 3,000 government owned and run airports now.) You could likely do the same for “Ports” or any other kind of vital, major infrastructure.
In 1929 we didn’t even have “antibiotics”. Now we fund the CDC to monitor and (hopefully) ameliorate disease outbreaks (including animal populations), and the NIH and other health organizations or improve “the general welfare” of the country. The average lifespan in the US has grown more than 20 years since 1929.
Should we talk about satellites, which did not get off the ground (bad pun, sorry) anywhere in the world without government funding? Want to know when the FBI was founded? Good idea? Bad idea? How about support for higher education?
And say, Social Security didn’t start for half a decade after 1929, Medicare for decades longer.
Yes, we spend more money on “government” now. Somehow people of this meme (not you, just “people”) somehow don’t think it means “we’re spending money on common goods to make YOUR life better” and somehow translate it into “those lazy government slobs sitting around all day doing nothing and I’m paying for them.”
Where are they working now? AI is likely to modify knowledge work - working with information. If having lower IQ correlates to lower facility with language and information skills, then they are less likely to be disintermediated with AI.
As has always been the case. Most of the jobs that humans did 200 years ago are done with technology today. Back then, 80% of the U.S. population worked in farming - now, barely a few percent are.
IOW, technology will be able to do whatever jobs humans are doing now, not whatever humans can do.
