Tuesday’s Forum
Steven L. Taylor
·
Tuesday, August 26, 2025
·
22 comments
OTB relies on its readers to support it. Please consider helping by becoming a monthly contributor through Patreon or making a one-time contribution via PayPal. Thanks for your consideration.
About Steven L. Taylor
Steven L. Taylor is a Professor Emeritus of Political Science and former College of Arts and Sciences Dean. His main areas of expertise include parties, elections, and the institutional design of democracies. His most recent book is the co-authored
A Different Democracy: American Government in a 31-Country Perspective. He earned his Ph.D. from the University of Texas and his BA from the University of California, Irvine. He has been blogging since 2003 (originally at the now defunct Poliblog).
Follow Steven on
Twitter and/or
BlueSky.
A summary of the actions state governments are taking to resist the current regime:
https://open.substack.com/pub/cmarmitage/p/its-time-for-americans-to-start-talking
Dave Barry discovers that Google AI wants him to be dead.
https://open.substack.com/pub/davebarry/p/death-by-ai
Back at work and with all my vacation time used up. I should declare an international day of mourning…
It seems El Taco is dead set in ending the US dollar rule as global reserve currency.
We have become an authoritarian state, and our top newsrooms are in denial (Dan Froomkin)
In recent days there have been more than a few musings on the ai front about bubbles. Financial ones.
Apparently, our economy is being kept afloat by ai investment. Consumer spending, normally a big piece of the economy, is persistently down.
The benefits overall of ai are also in question. Duh. Vast investment in a very expensive product that requires vast resources with pretty pie in the sky possibilities. The job losses are mounting. Young people with degrees in computer sciences and student loans to pay off have no entry level jobs because ai has the gig. And it’s not that good at the job either.
The whole thing reeks of FOMO and not sound business or economic sense. And if it crashes…boom.
America Tips Into Fascism (Garrett Graff)
@Kingdaddy: The first task I ever attempted on Chat-GPT was to sum a list of about two dozen three- and four-digit numbers (representing extra cash income for the month). I naturally assumed its answer was correct because…simple computation. Any basic computer program can handle that, no?
But the next day, I got a sudden urge to double check. I added the numbers myself with a calculator, got a different result. Hmm, that can’t be right. Repeated the task, same outcome.
I told Chat-GPT its math was wrong and gave it the corrected answer. This scary, world-changing, world class, job-and-person-replacing phenomenon responded: “Oh! You’re right!”
So. That’s AI. Useful, but also a joke.
It occurred to me while watching the powerful documentary on the Ohio State University athletic departments sexual abuse scandal that the victim-survivors story is a snapshot of the US today. They were asked why did they let the doctor do what he did. Big strong wrestlers would just knock the crap out of him. But they didn’t. They took it because it appeared everyone knew the doc was a predator of young men and just keep your mouth shut and go along to get along or else no scholarship, no degree and a vengeful power on your back.
The good news is that those OSU wrestlers did finally do something. They took and are still taking the heat, but they did speak out.
I think there might be a lesson here.
@DK:
Ironically you would have been better off asking it something to the effect of … “What buttons should I push on a scientific calculator to get the sum of x+x+x+x+x+x=”
Hell, it would probably be able to tell you which buttons to push to get the correct calculations to put a satellite into low earth orbit. But unless someone feeds it a times table or some other subroutine, AI can’t do math or calculate. 🙂
@DK: generative AI is incredibly useful when you don’t care about the quality or accuracy of the result.
There are a number of applications where Ipsum Lorem would be inappropriate but where you’re mostly filling space. Health insurance letters denying coverage of needed medical care, for instance.
@Rick DeMent: I recently saw an explanation of why AI may sometimes say 1 is more than 2. All it’s doing is stringing words together in a way it’s “learned” the words are often strung together, and it’s often said that 1st is better than 2nd. It “knows” add as a word, not as an action.
In case anyone missed it… It was reported on Friday that the SecDef fired the second of two generals who testified at the House Intelligence Committee hearing in March, when much of the questioning was about sensitive military details on the Yemen strike having been discussed on the Signal app. According to the AP this past Friday,
“Defense Secretary Pete Hegseth has fired a general whose agency’s initial intelligence assessment of U.S. damage to Iranian nuclear sites angered President Donald Trump, according to two people familiar with the decision and a White House official.
Lt. Gen. Jeffrey Kruse will no longer serve as head of the U.S. Defense Intelligence Agency, according to the people…”
Gen. Tim Haugh had already been dismissed as head of the National Security Agency in April. Recall how each general had addressed the classified nature of the operational information pertaining to the Yemen strike:
LTG Kruse: “I think what I would offer are two things and hopefully they’re helpful to you. One of them would be, in addition to that full package that we would talk about, when the Secretary extracted individual uh details from that and provided those on Signal, whether individually or in aggregate, that is his decision of what is classified is not from an operational aspect…”
GEN Haugh: “…the full packages were transmitted within classified, within traditional classified means.”
@Gustopher:
I’ve wondered if I can ask an LLM to write a prompt that would get it to do what I want.
The Supreme Court addressed the removal of Federal Reserve Board of Governors members when it sided with trump on May 22 in granting a stay of District Court orders preventing the removal of an NLRB member and an MSPB member. The unsigned two-page Supreme Court decision included this:
“Finally, respondents Gwynne Wilcox and Cathy Harris contend that arguments in this case necessarily implicate the constitutionality of for-cause removal protections for members of the Federal Reserve’s Board of Governors or other members of the Federal Open Market Committee. See Response of Wilcox in Opposition to App. for Stay 2−3, 27−28; Response of Harris in Opposition to App. for Stay 3, 5−6, 16−17, 36, 40. We disagree. The Federal Reserve is a uniquely structured, quasi-private entity that follows in the distinct historical tradition of the First and Second Banks of the United States. See Seila Law, 591 U. S., at 222, n. 8.”
And Justice Kagan’s much lengthier dissent included this:
“Except apparently for the Federal Reserve. The majority closes today’s order by stating, out of the blue, that it has no bearing on “the constitutionality of for-cause removal protections” for members of the Federal Reserve Board or Open Market Committee. Ante, at 2. I am glad to hear it, and do not doubt the majority’s intention to avoid imperiling the Fed. But then, today’s order poses a puzzle. For the Federal Reserve’s independence rests on the same constitutional and analytic foundations as that of the NLRB, MSPB, FTC, FCC, and so on—which is to say it rests largely on Humphrey’s. So the majority has to offer a different story: The Federal Reserve, it submits, is a “uniquely structured” entity with a “distinct historical tradition”—and it cites for that proposition footnote 8 of this Court’s opinion in Seila Law. Ante, at 2 (citing 591 U. S., at 222, n. 8). But—sorry—footnote 8 provides no support. Its only relevant sentence rejects an argument made in the dissenting opinion “even assuming [that] financial institutions like the Second Bank and Federal Reserve can claim a special historical status.” And so an assumption made to humor a dissent gets turned into some kind of holding. Because one way of making new law on the emergency docket (the deprecation of Humphrey’s) turns out to require yet another (the creation of a bespoke Federal Reserve exception). If the idea is to reassure the markets, a simpler—and more judicial—approach would have been to deny the President’s application for a stay on the continued authority of Humphrey’s.”
Wow. But I’m not confident that the court will stand by their recently-minted “Federal Reserve exception”.
@DK:
It’s also been noted that LLM-“AI” is hopeless at playing games such as chess.
The current models appear totally incapable of generating internally consistent rule-sets based on available information, and acting on their basis.
As opposed to simply locating what appears to be a valid “answer” to a query from a dictionary of scraped data with no internal ability to sift valid from invalid data beyond “data majority vote”, as it were.
Zero capacity to “test against conditions”.
Given the contents of the internet its a minor miracle that all LLM answers to queries aren’t: “Cats!”
(Or p@rn.)
@Eusebio: I’ve tended to see Roberts as caught in a difficult position and trying to put off confrontation. A recent article in The Guardian has pushed me toward seeing him as a standard issue RWNJ, just smoother than most. https://www.theguardian.com/us-news/ng-interactive/2025/aug/21/justice-john-roberts-supreme-court
He also seems inordinately fond of the Unitary Executive crap. Roberts famously said judges just call balls and strikes. He never promised to call them fairly.
@gVOR10:
…..I’ve tended to see Roberts as caught in a difficult position and trying to put off confrontation. A recent article in The Guardian has pushed me toward seeing him as a standard issue RWNJ, just smoother than most.
….. He also seems inordinately fond of the Unitary Executive crap. Roberts famously said judges just call balls and strikes. He never promised to call them fairly.
___________________________________________
I generally agree with you, however I think he’s surprised that the radical rightwing Republican takeover that so many spent decades building and hoping for actually happened seemingly so quickly. But he helped to make it happen with his Immunity and Official Acts opinion last Summer.
Prior to Trump he could appear to be a reasonable, perhaps moderating, institutionalist as in, ‘sure he’s a conservative, but not one of those burn it down types.’ But if one looks closely – at his majority decisions in Citizens United and Shelby County – he actually is radical, just not in an openly angry vengeful way like Alito or Thomas.
I made peanut butter noodles. I got the recipe online, and then made some changes:
1/2 onion thinly sliced
Several cloves of garlic minced
2 Tbsp roasted sesame oil (divided)
Black pepper to taste
2 Tbsp soy sauce
1/2 Tbsp vinegar (preferably rice vinegar)
2 1/2 Tbsp peanut butter
200 grams noodles (pre cooked)
So, cook the noodles to taste. Meantime, in a sauce pan add 1 Tbsp sesame oil and sauté the onion until it’s a bit soft, then add the garlic and cook for a minute or two. In a cup or bowl mix the soy sauce, vinegar, remaining sesame oil, and peanut butter. Mix well. If it feels thick, add some pasta water to loosen it up. Add to the pan, mix some more. add the noodles and toss well. You can add more pasta water here if it still feels too thick.
Today there will be yet another attempt to blow up the Xtarship. Best of luck on that. Can’t leave well enough alone…
recently I came across the claim that space travel from Earth is jutsssst barely possible. That if gravity were higher and/or the atmosphere denser, we’d find it impossible to launch anything to orbit.
I don’t think so. Maybe its jussst barely economical. But if the escape velocity were higher, then you’d need bigger rockets. Say a Saturn V, all three stages of it, would be needed to put a comms satellite on geosynchronous orbit. Given the high cost, it might not be worth doing, but it could be done.
Now, Jupiter has an escape velocity of 55.9 km/s. That’s 5.3 times Earth’s 11.2 km/s, and more than even a Saturn V can manage. Therefore any Jovians are trapped in their planet forever.
Maybe. and maybe not.
How about a nuclear fusion engine? Especially an air-breathing one. That is, like a jet engine. Except it needs reaction mass rather than oxygen to “breathe”. How fast could that go in Jupiter’s thick, hydrogen and helium atmosphere?
How much would it cost (and what’s the exchange rate for the Jovian Baryon to the Euro anyway)?
If anything, our hypothetical Jovians would find it harder to contain their very thick atmosphere inside a space capsule. It would be built like a bathyscaphe, only much, much, much, stronger. Like meters thick walls. the pressure is higher than at the bottom of the Pacific Ocean. Which also means more mass for the launch vehicle to carry, alas.
Popping up out of my bunker to share surprisingly happy news:
Jeanine Pirro’s Office Fails Three Times to Convince Grand Jurors to Return Indictment.
The bar for securing indictments in criminal cases is relatively low — so low that there’s a saying that prosecutors could indict a ham sandwich. But prosecutors in the nation’s capital have failed a shocking three times to get grand jurors’ approval for a felony case to move forward in an alleged assault against a federal agent.
It appears her office may attempt to beat this by making the indictment a misdemeanor instead of a felony, thus skipping the GJ and going directly to court, but I’m giggling and cackling in my G&T about the look on the judge’s face when/if this shows up on their docket.
https://www.msn.com/en-us/news/crime/jeanine-pirro-s-office-fails-three-times-to-convince-grand-jurors-to-return-indictment/ar-AA1Lg0Mf?ocid=msedgntp&pc=LCTS&cvid=68ae354d80b4401c982523d111ba30e2&ei=55
ETA for some reason, my ability to indent/block, or even to use a keyboard in a quasi-Luddite fashion, seems to be failing me today.
ETA2 But have no fear, Cracker and Luddite continue to do our imitations of Waldorf and Astoria heckling Kermit (at least in our heads). Carry on, everyone!
The V-2 heavy did not blow up. It wasn’t recovered, either.
I don’t know. I’ve an idea how hard this is and all, but scoring a suborbital flight as a success because the previous two tries blew up strikes me as empty bragging.
They can call it fast iteration all they want, but to me it looks like half-baked designs they’re trying to get right. Consider the worst thing the Saturn V did was shake loose plaster on the ceiling of the Kennedy control room during an early test.
@Flat Earth Luddite: Tell Cracker we miss him!