I wanted to make one remark about Noah Smith’s recent observation on AI’s “sales pitch”:

“Hi. Do you have a moment? I’m from the Cursed Microwave company. Our product is much better than a traditional microwave. Not only can it automatically and perfectly cook all your food, it also microwaves your whole body, so you and your family are paralyzed and unable to ever work again. Don’t worry, though, because when everyone has a Cursed Microwave, our society will probably implement Universal Basic Income, and you and your children can just go on welfare! Oh, by the way, we estimate that there’s a 2 to 25 percent chance that our microwaves will put out so much radiation that they destroy the entire human race.”

If a door-to-door salesman gave me this pitch, I would gently see him out the door, and then quickly call the FBI.

Quite to the contrary, it’s a great sales pitch if you define “great” as one that converts paying customers. But neither Mr. Smith nor ordinary people are the intended audience.

It’s a compelling pitch to short-horizon, competition-constrained decision-makers even if the long-term equilibrium is unclear or unstable. Phrasing it more kindly the pitch is:

Using AI you can accomplish the same results you are now with fewer employees. We can improve your bottom line.

It’s perfect when you identify its audience correctly.

Last night my wife and I saw Gabriela Lena Frank’s 2022 opera, El último sueño de Frida y Diego at Lyric Opera, our last opera of the 2025-2026 season. We were drenched by the sudden downpour in downtown Chicago during the walk from our parking garage to Lyric’s home at Civic Opera House.

The picture above gives you a good idea of what the opera looked like. The production design drew heavily from the works of both Diego Rivera and Frida Kahlo. I thought the concept and book were great, the production good, and the staging a throwback to the way operas were staged 50 years ago with the performers standing in lines across the stage and singing to the audience. Some of that may have been deliberate.

I found the music unengaging. In a sort of revival of opera seria it was cerebral and static, maintaining a slow, stately andante and similar vocal and orchestral densities throughout. The final scene, which should have been an emotional highpoint, maintained the same, detached tone.

My wife asked a good question: “Is this magical realism?” I think it sat right on the boundary of both, leaning more into fantasy. In magical realism the supernatural is ordinary and expected. In fantasy it is more extraordinary and used as a sort of intellectual device.

All in all it was a decent conclusion to the season but won’t become one of my favorite operas.

Kyle MacMillan, Chicago Sun-Times

Bringing together piano, celeste, gongs and myriad other percussion with traditional orchestral instruments, Frank draws forth a rich palette of sometimes hard-to-identify, discordant sounds. The music can be spooky and foreboding with rumbling marimba runs, shrill piccolo blasts and alien-sounding trumpet lines or moving and reflective with gently cascading harp or searching clarinet.

Intrinsic to this opera is the first-rate Lyric Opera Chorus, which is almost a constant presence, either seen on stage or heard offstage, helping to shape the opera’s reverential, almost sacred feel at times. Sometimes these singers are part of the action as cemeterygoers or underworld denizens, but other times they act as a kind of Greek chorus, prodding or commenting on all that is happening.

Hannah Edgar, Chicago Tribune

This review is more about the revival of interest in Frida Kahlo than it is about the opera itself.

Lawrence A. Johnson, Chicago Classical Music Review

El último sueño de Frida y Diego has much going for it: Frank’s compelling score, a strong cast, and a striking, at times visually stunning production. All elements came together Saturday to create a vivid, colorful and impressive night of theater.

Frank’s flowing, restless music is powerful, often haunting and eminently listenable. Even when the stage action was static or dragged, the crystalline timbres and iridescent scoring coming out of the pit consistently beguiled the ear.

And yet, paradoxically, engaging as Frank’s music is, there is little that sticks in the memory (an issue I’ve also had with Frank’s orchestral music). More crucially—and Cruz’s libretto bears some blame for this—there are no true musical peaks or standout solo moments for the singers. Yes, there are solos and a late quasi-duet for Frida and Diego. But they appear and go quickly in an efficiently scrolling theatrical canvas that doesn’t pause long enough for the singing to make a strong impact. At times Frida y Diego feels less like an opera than a Nilo Cruz play with intriguing incidental music.

Once again, Mr. Johnson clearly saw the same opera I did.

The nonpartisan Committee for a Responsible Federal Budget has proposed capping Social Security retirement benefits at $100,000:

This Trust Fund Solutions Initiative white paper offers a new option to improve Social Security solvency by establishing a maximum benefit level.

The Six Figure Limit (SFL) would set a $100,000 cap on the total benefit a couple retiring at the Normal Retirement Age (NRA) can receive starting this year. The SFL would be adjusted based on marital status and claiming age, with a $50,000 limit for a single retiree collecting at the NRA.

I wonder if they recognize what a “poisoned pill” such a reform would be for Democrats? Clearly, some Democrats do. Connecticut Rep. John Larson:

“Nobody’s getting wealthy on Social Security,” Larson said in a recent interview with CNBC, noting that more than 5 million Americans receive monthly benefit checks that are below the federal poverty level.

“It is the very sustenance that 40% of all Americans need just to get by, and it hasn’t been adjusted in more than 50 years,” Larson said.

Massachusetts Sen. Elizabeth Warren:

Washington, D.C. – U.S. Senator Elizabeth Warren (D-Mass.), Ranking Member of the Senate Committee on Banking, Housing, and Urban Affairs and member of the Senate Finance Committee, published an op-ed in Fox News hitting back at the Trump administration and Elon Musk’s attempts to gut the Social Security Administration (SSA) as a means of slashing benefits for Social Security recipients.

A key problem that goes quietly unmentioned: how can Democrats coherently support a $100,000 cap on annual Social Security benefits without also supporting a $100,000 cap on defined benefit pensions for those who don’t participate in the Social Security system, e.g. teachers?

I don’t oppose the reform in principle but even the CRFB acknowledges it won’t solve much of Social Security’s problems. As I’ve pointed out before the real problem staring us in the face is assumption failure—the 1983 reform didn’t foresee the sharp increase in income inequality that took place in the 1980s to the present day. My proposed solution was to make all wage income subject to Social Security withholding.

What do the midterm elections portend? Should we be prepared for a Democratic sweep? A Republican? Are big changes ahead or more of the same?

In his most recent Wall Street Journal op-ed Karl Rove analyzes the prospects for Democrats in the upcoming midterms. After noting that midterm elections tend to incline towards referenda on the sitting president and pointing out that Trump’s favorability isn’t particularly high, he points out something interesting, even paradoxical:

But Mr. Trump’s popularity isn’t the only consequential number. The RCP average has the Democratic Party at 35.8% favorable to 55.7% unfavorable. The Republican Party does better, 39.3% to 53.9%. How did the Democrats become less popular than Mr. Trump? It took many years, much hard work, wacky policy stances and plenty of missteps.

Even the RCP’s generic House ballot is a problem for the Democratic Party’s midterm prospects. Although Democrats lead Republicans, 46.9% to 42.1%, that may point to only modest Democratic gains in the House. Only a handful of districts are up for grabs this fall. Prognosticators suggest there are 17 districts that are toss-ups. Thirteen are held by Republicans, four by Democrats. On top of these are 13 seats held by Democrats and two held by Republicans that lean blue and one Democratic and three Republican seats that lean red. Finally, there are two GOP seats and one held by an independent (former Republican Kevin Kiley) that are considered solidly Democratic after recent resdistricting shenanigans.

Altogether, that isn’t a lot of targets for Democrats. If they swept all 39 at-risk seats, that would translate into a gain of 21. That’s about half the 41-seat boost they saw in 2018, when their generic-ballot advantage before the election was 7.3 points.

concluding that Democrats are acting as though trashing Trump is all they need to do. The tone of Mr. Rove’s op-ed is hortatory: he wants Republicans to get off their duffs and do something. His warning reflects the reality that even favorable structural conditions can be squandered.

The paradox to which Mr. Rove gestures isn’t particularly surprising. Indeed, it’s close to the modern norm, following the patterns of George W. Bush and the Republicans in 2006 and Barack Obama and the Democrats in 2014. Presidential approval ratings are personal rather than reflecting on their respective parties.

This also aligns with Ruy Teixeira’s recent lament about the Democratic Party’s declining standing with voters. The problem he identifies is not primarily candidate-specific but structural: the party brand itself has weakened. The polling Rove cites with Trump trailing generic Republicans, and Democrats trailing both fits that pattern. Democrats may benefit from dissatisfaction with Trump without having rebuilt broad confidence in their own party.

Larry Sabato projects the likely outcome of the 2026 midterms is that Democrats will gain control of the House by 21 seats while Republicans retain control of the Senate. That is also supported by prediction markets.

What does that mean for the next two years? If precedent is any gauge, it will mean continued gridlock and posturing by both political parties. Divided government without a strong mandate typically produces symbolic conflict rather than durable policy. If House Democrats remain true to form they will impeach Trump (again) and it will go nowhere (again). If Senate Republicans remain true to form they won’t accomplish a great deal.

John Halpin has just announced that he’s shuttering the Liberal Patriot Substack:

We tried our best to make a non-profit media model work with entirely free analysis and commentary published throughout our existence. But it turns out upsetting the partisan applecart on multiple issues is not a particularly good fundraising or business model.

He’s late to the party. I reached that conclusion more than 20 years ago.

Ruy Teixeira’s valedictory post there is a sort of lament for the Democratic Party. In it he reviews the folly of viewing what are called “trans rights” through a civil rights prism as part of the ongoing liberal project of defending and expanding personal rights.

He also proposes ten qualities of what he sees as a “realistic immigration policy”, eventually arriving at the observation that what the Democrats are presently supporting is a de facto open borders policy:

Going forward, Democrats must show voters they understand these realities and are willing to dramatically change the incentive structure for illegal and irregular immigration. That means strict border enforcement, elimination or radical restriction of immigration loopholes, and a credible interior enforcement regime that recognizes illegal immigrants, even if they stay out of trouble, are still illegal and therefore susceptible to deportation. Otherwise illegal immigrants who manage to enter the country will quite reasonably assume that they can stay here forever which of course is a massive incentive for more illegal immigration.

He also wishes that Democrats had some sort of coherent economic policy beyond opposing Trump or vague generalities like “affordability” or “abundance”. He concludes:

It’s time—past time—for Democrats to discard the conceit that they are on the right side of history and that therefore their positions are, and have been, noble and correct. Until they do so, I do not expect them to develop the dominant majority coalition they seek and vanquish right populism. Indeed, it could be the other way around. That’s a sobering thought.

I’ll miss Liberal Patriot. It was a pro-Democratic site with which I was frequently in agreement. I doubt we’ve heard the last of the John Halpin, Ruy Teixeira, or the other contributors there.

In my search for open source intelligence on the war in Iran one of the sources I have found is Data Republican’s Substack. To date she has produced two updates:

Data Analysis of the State of the Iranian Conflict on March 8, 2026

Data Analysis of the State of the Iranian Conflict on March 16, 2026

I’m looking forward to her next update which I expect in the next day or so. I offer these links without further comment.

Other sites offering open source intelligence on the war include:

bellingcat

Institute for the Study of War

I will add additional sites to these as I discover them.

If you have any additional suggestions, please place links in the comments.

I wanted to comment on Ramesh Ponnuru’s remarks on “Medicare for All” in a recent Washington Post op-ed before it disappeared. In the op-ed Mr. Ponnura sketches four reasons a national single-payer healthcare system modeled on Medicare won’t work. His reason are (in sequence):

• It would require a substantial increase in the taxes paid by people earning middle incomes which might or might not be made up by savings in payments of private healthcare insurance.
• Savings aren’t likely to materialize.
• The systems most similar to “Medicare for All” (Canada’s and Britain’s) have long wait times, particularly for specialists.
• It would require the abandonment of employer-subsidized and private healthcare insurance and Medicare Advantage.

I would like to propose several other issues. First, the costs of such a plan are wildly understated. The graph at the top of the page illustrates the real growth in real Medicare spending from inception to date. That is orders of magnitude faster than was estimated when it was originally enacted far exceeding the original projections, which proved not merely inaccurate but systematically and persistently optimistic. Here’s the real growth in Affordable Care Act subsidies since its inception:

Those, too, have grown considerably faster than estimated at the time.

When healthcare spending comprised 3% or 5% of spending, a five-fold increase would have been painful but, possibly tolerable. Now that it’s 16% of spending a ten-fold increase in healthcare spending, as occurred with Medicare, isn’t arithmetically possible. A five-fold increase is hypothetically possible but absurd. The implication is that for such a plan to work costs must be controlled. Growth on the scale Medicare experienced cannot be repeated without overwhelming the rest of the fiscal system and, indeed, the economy.

The second issue is that Congress has demonstrated that it is unwilling to control healthcare costs. That was proven dispositively by the experience with Medicare SGR, demonstrated repeatedly by Congress.

The third issue is that once implemented such a plan would be irreversible. There would be no alternative system to return to. Once private and employer-based systems are displaced, rebuilding them would be economically and politically prohibitive.

The relevant question is not why Medicare spending exceeded projections. The relevant question is why we should believe that future projections will be more accurate than past ones. Medicare and ACA estimates were produced by serious analysts using the best available data, yet both proved systematically optimistic. That shifts the burden of proof. Proponents of a vastly expanded system must identify concrete, politically sustainable mechanisms that will prevent the recurrence of those errors. Absent that, assurances of cost control are not evidence. They are repetition.

I don’t oppose a national single-payer healthcare system in principle. I oppose it in practice. For any such proposed system to work it must be structured to align the incentives of consumers, providers, and administrators. It is not sufficient to assert savings from administrative simplification or bargaining power. Proponents must identify specific mechanisms that (1) constrain provider behavior, (2) are politically sustainable, and (3) have not historically been reversed under pressure. As the graphs above clearly show more than blithe assurances are required. The problem is not that we lack ideas for controlling healthcare costs. The problem is that every serious attempt to implement them has been reversed when it encountered political reality.

I found the “talking heads” programs yesterday remarkably vacant. On Face the Nation Margaret Brennan tried her best to get NATO Secretary-General Mark Rutte to condemn President Trump’s war against Iran, rephrasing the question multiple times in various different ways, but he refused to take the bait. For example, in response to her observations about President Trump’s scolding NATO countries for being reluctant to secure the Straits of Hormuz, he more or less tut-tutted the question:

So it is only logical that European countries needed a couple of weeks to come together.

and praised President Trump’s conduct of the war heretofore:

And that is exactly why I feel in Europe, that most politicians, it resonates with them. What the President is doing here, which is taking out- degrading Iran’s capability to be, again, an exporter of chaos, sheer chaos to the region, to the world.

She was much more successful in getting the responses she wanted from Colorado Rep. Jason Crow. She just needed to turn him loose:

They started this war without congressional authorization. There’s no imminent threat. They actually have even stopped trying to pretend there was an imminent threat, which is what is necessary for the president to take action without congressional approval. So now here we are spending tens of billions of dollars of taxpayer money, losing American lives. Congress isn’t involved. We’ve been stonewalled. The American people are not in the driver’s seat here, right? Because Congress has been sidelined in this they are the ones that need to have the voice and whether or not they’re going to send their sons and daughters to go fight this war, whether or not they’re going to finance it. It’s time for this to end.

At ABC’s This Week Jonathan Carl followed a similar pattern. Administration representatives were broadly supportive of the war while the other guests including North Carolina Sen. Thom Thillis and California Sen. Adam Schiff were opposed to it.

Journalists have multiple strategies available to them for shaping the message conveyed about news stories. The first is which stories they elect to cover. In this case they are choosing an important story—the war in Iran. The second is explicit editorializing which, wisely, they avoided.

The third strategy is how they choose to cover the story and that is clearly what was happening yesterday. That strategy may be working. A majority of Americans think the war isn’t going well for the United States. Public understanding is largely a function of what is made salient, how it is framed, and what shortcuts people use to process it. Decades of research on agenda-setting and framing show that what is emphasized in major media becomes what the public regards as important, and how it is presented shapes how it is understood. Most people do not independently verify or seek out competing accounts; they rely on what is repeatedly surfaced.
Personally, I would rather that journalists just report the news and let us decide what to think about it.

As far as I can tell Fareed Zakaria’s thesis about the effects of relatively inexpensive unmanned drone aircraft on warfare is summed up in the closing paragraph of his Washington Post column:

In 1991, the Gulf War taught the world that advanced technology could make war precise. In 2026, Iran is teaching the world something more consequential: Precision will now be mass-produced. The countries that prevail will not simply be those with the finest platforms. They will be those that can combine small numbers of exquisite, expensive weaponry with vast numbers of cheap drones. Human judgment will over time give way to computer algorithms. That is the future of war. And it is arriving faster than most of us imagined.

His implication seems to be that drones erode or even negate the tactical advantage possessed by the United States, China, and other major powers.

Unfortunately for Mr. Zakaria’s thesis, the statistics he cites (along with some he doesn’t) actually point in a somewhat orthogonal direction. Consider the following:

• Since the beginning of the war Iran has deployed about 3,000 drones.
• That has resulted in about 30 total casualties.
• A Shahed drone costs about $25,000.
• The cost to Iran has been around $100 million.
• Ukraine’s STING interceptor drone costs around $2,000.
• Iran’s drone usage have not yet reduced U. S., Israeli, Saudi, or other Gulf state military capability or will to use it.

The available evidence does not yet support the claim that inexpensive drones negate the advantages of major powers. What it suggests instead is that drones shift warfare toward a cost-optimization problem in which defense currently appears to have the advantage. There are reasons to believe that particularly from Ukraine’s use of its fixed geography and detection in its defense. Iran’s experience demonstrates that mass without integration—without targeting precision, coordination, and doctrinal coherence—produces low returns. The future of warfare will not belong to those who simply deploy large numbers of cheap drones, but to those who can integrate them effectively into layered systems of offense and defense. In that respect, drones may reinforce, rather than diminish, the advantages of technologically sophisticated militaries.