The risks of AI are real but manageable (2023)

It took over a century to go from Oliver Twist to the broad Post-WWII Middle Class. How long this time?

The first thing this wave of automation seems likely to replace is automation itself.

For this reason it's hard for me to give much attention to comparisons to previous industrial revolutions. It would have looked way different if the first thing they built was a factory factory.

I will give this post some slack though, it was written in 2023.

Would love to know what he thinks nowadays

This page fails to load for me (Firefox on MacOS with UBlock Origin AND DNS-level pi-hole adblock)

https://www.gatesnotes.com/search-reader?readerfocus=the-ris...

This link works, though

The whole 'AI can't automate a lot of jobs, because that's exactly what everyone in the past said about technological advances then!' is a classic fallacious argument. It simply fails as a logical thought pattern. Past technological gains that happened decades or a century plus ago are discrete events separated by time and space from today. 'It didn't happen in the past so it can't happen today' is the same logic as 'I've flipped this coin and gotten heads 5 times, so it will land on heads on every future throw too'. It's completely possible that several times in the past, humanity was wrong about widespread job automation- but they can be right this time.

The whole argument just irritates me on a logical basis. You're basically saying 'the future will unfold exactly the way the past did'. No that doesn't follow at all. We may have dodged widespread automation in past history, but that doesn't mean it will automatically happen every time in the future. You haven't discovered like an Iron Law Of History that repeats every single time. You're just reasoning from a very small sample size and overfitting. Sorry for the rant, the whole argument just annoys me

Almost daily I feel like im arguing with Ai by proxy, loose out on jobs because the recruiter just consults Chat get me tea and cant be bothered to read official docs

Just like with STDs

His analogy misses a lot of subtleties.

Consider cars.

He says something like "we fixed the problems with cars". I say "what problems and what we"? Technologies change the social and political order, sometimes preventing us from solving their problems.

In the U.S., cars changed the landscape in a very different way from the rest of the world. Because the car industry got a lot of political clout they were able to restrict public transportation (particularly in cities) and built an environment where you're barely functional without a car if you live outside New York or San Francisco (both cities had public transportation before the car industry came up). The rest of the world uses a lot more buses, trains and bicycles than the U.S. does.

Cars created a world with methyl-lead gasoline, global warming, corrupt petro-cleptocracies and several other distortions that we didn't completely fix yet. Mr. Gates only talks about accidents and drunken-drivers. That was easy because the responsible for those problems don't have political and economic clout. Now good luck trying to fix the problems caused by people with lots of money (e.g polluters, big tobacco, health insurance industry) or lots of votes (e.g. cars and guns owners).

This is so incredibly naive to me; the problem isn't AI, it's (as always) profit-above-all driven sociopaths that are running the show.

Bill Gates should stick to pretending to be a philanthropist. I.e., don’t quit your day job, Bill.

“Everything was fine before, so it’ll be fine now” is the level of argument I expect to find in thousand-year old religious texts.