Is Our Children Learning from Education Tech?

And more from this week...

Good morning, everyone. A slight deviation in form this week, as I had more links than usual to share, but no “one thing to read” into which I wanted to dive in depth. So we’ll go with a couple of best reads, some other things you should be reading, and then plenty about which I wish to complain.

BEST READS

The All-New Big Tech American School | Emily Brownlee, After Babel

The whole thing is powerful, but I am especially pleased to see an experienced educator note, in response to the tobacco-like assertions that no one has “proved” technology’s harm to children, that “no one in the field of education can prove why it was introduced in the first place.” This is what I find so baffling about all the debates around technology in the classroom: What evidence is there that the digitization of education over the past 20 years has accomplished anything? We know that test scores are down, attendance is down, costs are up, mental health problems are up. By all means, confuse correlation for causation! I’m looking for anything that might plausibly suggest the massive infusion of technology into the classroom has delivered anything except profits for the companies developing and plying the technology. DMs are open.

The Reality of America’s Multi-Racial Working Class | Robert Ordway, Aaron Renn’s Substack

We are big fans of Robert Ordway and of Aaron Renn’s Substack here at Understanding America, so when Ordway writes a guest post for Renn, it’s self-recommending. The multi-ethnic working class has long existed, he notes, what’s new is the Beltway Bubble’s awareness of it. A proud product of that class, from a fascinatingly diverse Indiana community that fits none of the contemporary political frames, his reflections are a uniquely useful anecdote to the typical op-ed page.

WHAT ELSE SHOULD YOU BE READING?

The Biggest Business Myth Is That PE-Backed Companies Are Well Run | Auren Hoffman, Summation

Hoffman makes a great point: “The number one exit for PE firms are other PE firms.” If private equity’s premise were that ownership by private equity yielded substantial value creation, PE firms would never want to buy from other PE firms—the opportunities would already have been seized. If, on the other hand, the premise were to trade piles of assets in circles, financing each others’ deals and charging management, consulting, and transaction fees along the way, it’s exactly what you’d expect.

• Bonus Link: What to do when your industry has been underperforming dartboard throws for a decade and its predominant activity is handing underperforming and illiquid assets back and forth? The SPAC Is Back!

Why Didn’t We Know Randi Weingarten Was a DNC Member? | Erika Sanzi, RealClearEducation

The title says it all—seriously, how was this not constantly discussed, and how was it allowed?

What if MAGA Has a Point About Science? | Paul Sutter, Undark

I found this essay charming, like the Father’s Day card that comes home from school exhibiting great care and effort, notwithstanding some crumpling in the backpack and “father” spelled wrong on the front. Sutter, a Johns Hopkins cosmologist (different from a cosmetologist), seems entirely genuine in sensing that something is in fact wrong with science, honestly confused about what has happened and why, and clumsily obtuse in working through any of it, managing to inadvertently insult more than clarify along the way. It’s worth considering, as we impose demands upon our broken institutions to reform themselves, that even those eager to make amends may in good faith struggle as mightily as Sutter does here.

Garry Tan, Y Combinator CEO, on X | posting video of Scale AI co-founder Alexandr Wang

“Alexandr Wang: The future isn't humans vs AI - it's humans managing AI chaos. Even self-driving cars need 1 human operator per 5 cars to handle edge cases. The last 10% of accuracy? That's where humans shine. Management isn't going away, it's evolving.”

I’d always assumed that various automation technologies sold as “replacing humans” would in fact continue to require humans in the background for the many edge cases. (In The Once and Future Worker, published in 2018, I wrote “If a truck driver can leave his cab and work from a central control facility, in which he toggles to whichever of numerous trucks requires human guidance at a particular moment, he will likely be happier, healthier, and better paid as well.”) This is the first time I’ve seen it discussed so openly in the context of self-driving cars, and recognized—by AI optimists—as a permanent condition rather than a stop-gap.

• Bonus link: Sam Altman and Brad Lightcap at Hard Fork Live. The first part of this probably belongs in the special AI Corner / Cringe Corner crossover episode below, but I found this remark from Lightcap, OpenAI’s Chief Operating Officer, very revealing: “We will kind of wake up one day with this incredibly powerful thing and will the world be different that day? I think what we've all kind of agreed now is it probably won't. I think these things really have to be kind of integrated into people's lives, they have to be felt, and that change is more gradual. And so we work really closely with companies and as much as we do with users to figure out what that process will look like.”

  Yeah, no kidding. Sorry to quote The Once and Future Worker to you again, but I do want to continue emphasizing that these are the fairly obvious insights available to anyone with experience operating in the real world: “The prophecies ignore the gradual timeline on which transformations inevitably occur. Technology takes time to adapt by fits and starts to real-world conditions. And organizations take time to adapt to new technologies.”

CRINGE CORNER

Falling Birth Rates Don’t Have To Be A Crisis. Just Look At Japan. | Jim Pethokoukis, Washington Post

Looked at purely as a matter of economics, Pethokoukis is “hopeful” about Japan’s slow demographic death. (“If a country that’s considered an archetype of demographic decline can go gently into that good night, so, too, might the United States and other rich countries.” Hooray.) Wage growth and inflation are up, so “better to go with the flow than fight against it.”

On the upside, it’s fascinating to see that whereas the Old Right and Wall Stret generally view labor shortages as disasters necessitating more immigration to fill jobs Americans won’t do at low wages, it turns out they can see the virtuous cycle in which tight labor markets lead to productivity enhancement. “As Goldman sees things,” reports Pethokoukis, “the demographic decline that once drained vitality is now creating a ‘virtuous cycle’ of tightening labor markets, increased worker bargaining power and more investment in productivity-enhancing tech.” What explains the sudden adoption of this pro-worker lens? Perhaps the convenience of using it to argue against pro-family policy: “The Japan scenario seems a more promising path forward than further natalist nudges.”

American Democracy Might Not Survive a War With Iran | Robert Kagan, The Atlantic

I was in the middle of researching Kagan’s and colleague Bill Kristol’s role in laying the groundwork for America’s misbegotten efforts at “benevolent hegemony” in recent decades, when I came across this doozy. Come for his vague use of the term “Saddam Hussein’s weapons programs” (what kind of weapons, Robert?). Stay for the claim that actually the liberal world order would still be perfectly sustainable, the problem is just that America’s purported turn toward illiberalism undermines the system.

Gilead’s HIV Breakthrough | Editorial Board, Wall Street Journal

The Journal’s opposition to drug price controls underscores the extent to which their positions tend to be pro-corporate-profit rather than pro-market. As usual, the editors argue here that, “President Trump’s threat of drug price controls risks slowing such advances by reducing the financial incentive to innovate.” But there is no natural rate of profit in the drug market. The “financial incentive” is almost entirely a function of policy choices, both to provide intellectual property protection and permit collection of rents on the supply side, and to operate public programs that pay the prices set by these government-granted monopolies on the demand side. If more profit leads to more innovation and thus greater social benefits, why stop at current levels? Let’s lengthen patent duration and increase drug prices, to make the incentive even stronger to develop new drugs, and ensure drug companies have even more resources with which to do so.

There’s a term, incidentally, for recognizing that an industry’s investment and innovation generate massive social benefits and thus should be encouraged, even with policies upstream that intervene in free exchange between consenting individuals and downstream that guarantee profitable sales beyond what the market would otherwise afford. The term is “industrial policy.”

U.S. Immigration Crackdown Will Leave Deeper Scars Than Tariffs | Tej Parikh, Financial Times

This is supposed to be an article about the wonderful economic benefits of illegal immigration and the harm of restriction. But here’s a reordered series of sentences therein:

“The impressive growth in US jobs following the pandemic has been driven by foreign workers. Recent growth has depended on foreign-born labour. In some hands-on occupations, such as brick masonry and roofing, which employ a high proportion of undocumented labourers, labour-saving technologies are still limited. Simply put, the loss of foreign workers is akin to removing an economic input. It would leave the US extra reliant on generating significant productivity gains.” Don’t threaten me with a good time!

I hope Goldman Sachs won’t mind if I borrow their analysis of Japan from above: “The demographic decline that once drained vitality is now creating a ‘virtuous cycle’ of tightening labor markets, increased worker bargaining power and more investment in productivity-enhancing tech.”

SPECIAL AI CORNER / CRINGE CORNER CROSSOVER EPISODE

Does AI Make Us Stupid? | Tyler Cowen, The Free Press

You may have seen the recent study from MIT finding that reliance on ChatGPT reduces cognitive activity and engagement. I was kind of lukewarm on it—seemed a bit contrived—until I saw how the effort to undermine it looked. If this is Cowen’s rebuttal, then I’m in!

Cowen argues that the researchers ignored the potential for people to allocate more cognitive effort to other activities after outsourcing what they could to an LLM and, in a truly hilarious twist, attempted to apply… as Justice Ketanji Brown Jackson would say, wait for it… “The important concept here is one of comparative advantage, namely, doing what one does best or enjoys the most.” I suppose it is my burden as an economist to explain that this is not what comparative advantage means. Comparative advantage would leave humans doing that which they do relatively more efficiently, regardless of absolute efficiency or enjoyment. It would also require the LLM and the human to be in a trading relationship. If what he means is, “ChatGPT will give people more time to do what they want,” that’s fine, but it’s not comparative advantage.

“The real question—again not measured by the researchers,” writes Cowen, is whether students relying on ChatGPT to write their essays “did something meaningful and memorable with their freed-up time.” That’s a question, and maybe one you’d focus on if you thought that the immediate gratification of early-adulthood hedonism is the thing our society, technology, and education system should optimize. But it’s reminiscent of the UBI supporters who envision an explosion of self-actualizing poetry if people were freed from the obligation to support themselves. Maybe, here me out, the welfare-maximizing option would be for people to do their homework.

Everyone Is Using A.I. for Everything. Is That Bad? | Kevin Roose and Casey Newton, New York Times

I’d been thinking that a really good litmus test for how useful and important LLMs have become is to, start JFK voice, ask not what your LLM can do for you, ask what you cannot do without your LLM. On cue, Kevin Roose, our inveterate AI enthusiast, offered this: “I can’t think of another technology, besides maybe the smartphone, that has gone from ‘doesn’t exist’ to ‘basically can’t function without it’ in less time. … A.I. has essentially replaced Google for me for basic questions: What setting do I put this toaster oven on to make a turkey melt? How do I stop weeds from growing on my patio? I use it for interior decorating — I’ll upload a photo of a room in my house and say, ‘Give this room a glow-up, tell me what furniture to buy and how to arrange it and generate the after picture.’ A friend of mine just told me that they now talk to ChatGPT voice mode on their commute in their car — instead of listening to a podcast, they’ll just open it up and say, ‘Teach me something about modern art,’ or whatever.”

Those were the best examples. Roose’s co-host, Casey Newton, then noted that he uses it for research and fact-checking, and his boyfriend (a software engineer at Anthropic) relies on it for coding. So in short, it’s an improved Google that’s really important for coding. Sounds about right, and not like something you, or I, or these hosts “basically can’t function without it.”

AND AT COMMONPLACE

The Internet Comes of Age by Brad Littlejohn. In Paxton, SCOTUS protects childhood from the porn industry.

Democrats Just Can't Quit Their Gender Ideology by Brandon Showalter. But they may be starting to hope it will fade away.

This week on the American Compass Podcast, I moderated a friendly debate between Sohrab Ahmari and Josh Hammer about applying “America First” in practice, the wisdom of U.S. involvement in the war between Israel and Iran, and the limits of a Jacksonian foreign policy in the 21st century.

Enjoy the weekend!