[andrewmutz]

If you want to understand the likely capabilities of AI technology in the future, listen to software engineers like this guy.

If you want to understand the impact of AI technology on the economy, don't listen to software engineers, listen to economists.

[smallmancontrov]

Just remember that the US purged left-leaning economists during the cold war and the field re-grew under intense think-tank incentives towards the economic right, so if you think labor/capital dynamics might be important to the AI revolution you really ought to balance your "random" sampling of US economists with some Piketty (Atkinson, Stiglitz, Zucman -- but in an era where reading even one book is considered a herculean feat of focus, "Capital in the Twenty First Century" by Piketty is the canonical pick).

[pirate787]

Piketty is just a marxist flailing around, backfilling data to fit his belief that communism is the solution for every problem. He's been spectacularly wrong in his predictions so far, for example he said Milei would be "devastating" for Argentina and the opposite is the case.

[carlob]

I'm pretty sure no Marxist considers Piketty a fellow Marxist, he's much closer to social democracy and Keynes. Piketty advocates for market economy and a global capital tax, Marxists on the other hand advocate for the violent overthrowing of capitalism and collective ownership of the means of production. Can you spot the difference?

[nahuel0x]

Piketty is not a marxist, he even declared he never read Das Capital.

[viccis]

>for example he said Milei would be "devastating" for Argentina and the opposite is the case.

Love to see you mix unsubstantiated vitriol with overt lies like this.

[smallmancontrov]

Let's see.

https://en.wikipedia.org/wiki/Thomas_Piketty

> A visit to the Soviet Union in 1991 was enough to make him a firm "believe[r] in capitalism, private property and the market"

Ok, that's what he says, but what does he want? Does he want to eliminate social classes (communism)? Eliminate private ownership of the means of production (socialism)?

> His 2013 book Capital in the Twenty-First Century, relies on economic data going back 250 years to show that an ever-rising concentration of wealth is not self-correcting. To address this problem, he proposes redistribution through a progressive global tax on wealth.

No, looks like he just wants taxes. Case closed: this is instance #54367 of an economic conservative pretending that it's marxism to tax a penny from a billionaire. And you call yourself "pirate"? Sigh.

[chadgpt3]

A visit to the USSR makes one a capitalist, a visit to the USA makes one a communist. Maybe they both suck?

[wyre]

Maybe critique his ideas instead of his predictions. Piketty is an economist, not a future-teller.

[Maxatar]

What's the significance of his "ideas" with respect to economics if they don't lead to actual predictions?

Why did he decide to make these predictions if his ideas supposedly have no bearing on how predictions are created?

Does economics no longer even have the pretense of being a science, which by its very nature exists to make testable predictions?

[lll-o-lll]

If we are going to start discounting economists based on failure to correctly predict things like - interest rates - then you’re going to be chopping the entire mainstream field. If any economist was ever beholden to making accurate predictions, the field would currently be vacant.

[MSFT_Edging]

Economics isn't a science. It's a vibe check. That's why unprofitable companies make up a majority of the stock market.

[jdkoeck]

If he is confident enough to make economic predictions, it’s only fair that we take him to task on these.

[AngryData]

Anyone who flat out dismisses others work because they are "marxist" is only proclaiming their own ignorance on economic topics.

[TalkingCodeMonk]

Also remember that the Koch Bros (plus numerous other billionaires) have spent billions [0] poisoning economics academia by funding their narcissistic version of pseudo-libertarianism across think tanks, lobbyists, and colleges... They're so "Libertarian" they believe corporate dictatorship is freedom and poisoning their communities is the communities problem.

I remember first going to a party at an Economics students house 20 years ago, and thinking they all seemed like they were in a cult. Wasn't until fairly recently I figured out it was from propaganda.

[0] https://wikipedia.org/wiki/Koch_network

[hurtigioll]

Taleb joke:

junior trader for a bank looses $10 mil. boss asks him what happened. trader says he sold oil because bank economist said oil price will go down. boss fires him. junior asks how could he become a good trader if he's fired on the first losing trade. boss says "no, you idiot, I didn't fire you because you made a losing trade, I fired you because you listened to our economist"

[koe123]

I come from a research background, and transitioned to software later. There is an interesting tendency of software engineers to believe they have skills outside of their skillset.

Relevant here: the would we trust a Software engineer, which in general don’t always obtain the mathematical foundation to understand deep learning in the first place, on the trajectory of AI?

[PaulHoule]

Part of my software engineering skillset is "going native" with subject matter experts to be able to get more out out of them and even work around the lack of sufficient SME on a project.

I see software development as part of a broader science, technology and even ideology of simulation. But I came from a research background too.

[koe123]

Sounds like a similar track, and I agree that its a useful skill and talent.

What I mainly noticed was, after really understanding my domain, the confidence of the SWEs I was working with despite being incorrect. Now I am a SWE and I try to stay humble.

[kakacik]

This is correct behavior. There must be a name for this effect of having some more or less shallow understanding and feeling like an actual expert with decades of experience from various sides of the topic.

SWEs I think are more susceptible to this since as you say we often dip in many areas and industries. No, we are not actual SMEs and proper experts (barring exceptions of course, but in any case we usually have a specific view on domain, while proper experts understand many/all views).

[drtz]

> would we trust a Software engineer, which in general don’t always obtain the mathematical foundation to understand deep learning in the first place, on the trajectory of AI?

Valid point, but it suggests a mathematician who understands the math behind AI is more capable of grasping its trajectory, which is probably not the case.

People who are deep in the inner workings of this stuff day in and day out are the only ones who have a chance at having any real insight.

[koe123]

I think more broadly that grabbing attention with predictions and hot takes has become lucrative, and we definitely don’t celebrate prediction accuracy.

[827a]

If you want to understand the impact of AI technology on the economy, don't listen to software engineers, in fact, don't listen to anyone, no one was able to predict what the economy was going to do pre-AI, no one has any clue what's going on.

[amelius]

> listen to economists

Did anyone ever keep track of how often economists turn out to be right about anything? I didn't, but have the feeling that it isn't much better than flipping a coin.

[corimaith]

Reality is messy. But they are the ones at least thinking the most about this, and I'd say the coin is still overall weighted in their favour than listening to an uninformed opinion.

[amelius]

That could be true, but the self-reflection could be a lot better.

[athrowaway3z]

No serious economist would claim such a thing. In fact, it was all the rage a decade ago for economists to remind everyone that they - in fact - should not be looked at to do this.

> An economist is an expert who will know tomorrow why the things he predicted yesterday didn't happen today.

[david_shi]

"The growth of the Internet will slow drastically, as the flaw in 'Metcalfe's law'–which states that the number of potential connections in a network is proportional to the square of the number of participants–becomes apparent: most people have nothing to say to each other! By 2005 or so, it will become clear that the Internet's impact on the economy has been no greater than the fax machine's."

[Animats]

I expected saturation to happen, having been working on the Internet since the early 1980s. I did not see the dot-com boom coming, with it becoming a necessity for all companies, down to the dry cleaner level, to have a web presence. That was pushed over the top by hype and overfunding before it was cost effective. Like Uber and Space-X.

[Jtarii]

Krugman is a bitch.

https://www.youtube.com/watch?v=GiJXALBX3KM

[CamperBob2]

I wouldn't call Krugman nasty names, but I do have to wonder why anyone pays any attention to someone who has been so consistently and uniformly wrong about virtually everything he's commented upon. Even a Nobel should only earn you so much slack in life.

If you look at his track record, it's hard to explain without resorting to accusations of a humiliation fetish.

[david_shi]

Unfortunately a lot of (economics) research is co-opted to reinforce and obfuscate in service of power.

[setsewerd]

> I wouldn't call Krugman nasty names

> If you look at his track record, it's hard to explain without resorting to accusations of a humiliation fetish.

It may not have been a nasty name but damn that was brutal.

[throw0101c]

Are there specific things that Krugman has been wrong about that you have in mind, because he has made a number of notable good calls: he predicted that the 2009 stimulus would be too small:

* https://archive.nytimes.com/krugman.blogs.nytimes.com/2010/0...

A bunch of folks predicted that QE would cause all sorts of bad things:

> We believe the Federal Reserve’s large-scale asset purchase plan (so-called “quantitative easing”) should be reconsidered and discontinued. We do not believe such a plan is necessary or advisable under current circumstances. The planned asset purchases risk currency debasement and inflation, and we do not think they will achieve the Fed’s objective of promoting employment.

* https://www.hoover.org/research/open-letter-ben-bernanke

He took the other side of that and was right. When QE2 came around, Pimco—the famous bond trading folks—predicted certain things would happen and put large amounts of money on those predictions, and again he said it was all wrong, and Pimco turned out wrong:

* https://archive.nytimes.com/krugman.blogs.nytimes.com/2014/0...

This isn't to say he's always right. He has been wrong, but more importantly he admits when he's wrong and tries to figure out why:

> The first thing I want to say, I think most of our audience here, at least the economists, remember the big debates over QE early in the last decade, and all of the economists who predicted dangerous inflation and then refuse to admit that they've been wrong. And I don't want to be one of those guys, so I need to start out by saying that when we had our last discussion I was relaxed about the inflationary outlook and I was wrong, it turns out that inflation is coming way higher than I expected and I think the important thing on stuff like that is to try and figure out why you were wrong and learn from it. It's when you get to the why I was wrong, what's odd is it's not the simple script that I might have expected. For those who remember our earlier debate, that was centered mostly on fiscal policy, which I think is not going to be the case now and it was centered around the American rescue plan, which was a very big slug of money. And Larry was arguing, like a lot of people, that it was just way too big, that it was too much stimulus and that the economy would massively overheat. I was, I agreed that it was a lot of money, but argued that it wasn't designed as stimulus and, in fact, you know that had other purposes, and that it was likely to have low multipliers, so that there wouldn't be that much overheating. […]

* https://bcf.princeton.edu/wp-content/uploads/2022/01/Webinar...

More generally, if you look at the people Krugman has been arguing against, they've been more wrong, more often, more longer than he probably has, pushing garbage for decades:

* https://wwnorton.com/books/9781324005018

Is there someone(s) else that should be listened to more? (Certainly not Austrians or (Friedman) Monetarists, or anyone putting forward trickle down non-sense (that's basically most folks on the political right).)

[throw0101c]

To add to this, a 2011 paper looked at the predictions of the various talking heads, and Krugman was found that he got the most right (15/17=88%):

* https://www.hamilton.edu/documents/an-analysis-of-the-accura...

Second place was 82% correct, third was 75%.