Links and Notes for October 23rd, 2026

Below are links to articles, highlighted passages[1], and occasional annotations[2] for the week ending on the date in the title, enriching the raw data from Instapaper Likes and Twitter. They are intentionally succinct, else they’d be articles and probably end up in the gigantic backlog of unpublished drafts. YMMV.

Table of Contents

• Public Policy & Politics
• Journalism & Media
• Economy & Finance
• Art, Literature, Music, & Cinema
• Philosophy, Sociology, & Culture
• Technology & Engineering
• LLMs & AI
• Programming
• Fun

Public Policy & Politics

Apple Maps still shows the Markthalle as having a 'Flavors of Israel' stand

They know which way the wind is blowing, so they renamed it to 'Flavors of Jerusalem'

My Name Is Abdul’: El-Sayed Confronts Trump’s Racist, Anti-Muslim Attacks Head-On in New Ad by Julia Conley (Scheer Post)

““He knows my name is Abdul,” El-Sayed said of the president earlier this month. “He says it every day. But he’s not just attacking me. He’s attacking all of us. He thinks so low of us. He thinks he can use my name and my faith to distract us from the suffering his policies are inflicting on us at the pump, at the grocery store, and at the doctor’s office. He thinks we don’t know any better. It’s time for change.””

“On Monday, former Michigan Democratic Party chair Lon Johnson accused El-Sayed of “extremism” and said his rhetoric threatens Jewish communities, calling on Democrats in the state to “leave the Senate race blank” to represent “a rejection of ideological absolutism.” El-Sayed is a strong supporter of Palestinian rights and, like a majority of Americans, is deeply critical of US support for Israel as it continues its assault on Gaza and the West Bank.”

How Paris Is Hamas by Indrajit Samarajiva (Indica)

“Europe is the idea that tribes which have been fighting for centuries should put aside their difference to wage war on everybody that doesn’t look like them. It’s not racist to point this out, it’s racist to do it.”

“Just because the Empire doesn’t acknowledge its losses in the news doesn’t make them any less historically untrue. I repeat what goes unrepeated. The strongest European Army has lost (as they always do) to Russia. The White Navy lost a major naval battle to Yemen, major aerial battles to both Iran and Yemen, and a ground war to Palestine.”

“That’s what the Empire is fighting for, across West Asia, blowing up diesel in Russia and crude in West Asia. It’s not trying to restore energy flows, they could do that by just leaving. No, if they can’t have the energy they’ll burn it on the way out. They would rather rule over a ruin than deal with non-Whites equally. But, thankfully, the non-Whites are dealing with them.”

The European Union wants to punish people for their speech—without even a trial by Matthew Petti (Reason)

“The European Union claims that “the fundamental right to freedom of expression” only applies to journalists “acting in good faith,” and that authorities do not have to give due process before freezing people’s bank accounts, as these sanctions are “temporary and are not criminal in nature,” according to a legal letter shared by a sanctioned journalist.”

I am so tired of European elites. Let them burn.

“The council quoted a 2022 court case between the Russian news outlet RT and several European countries: “The protection of the right of the media and, more particularly, of journalists to impart information on issues of general interest is subject to the proviso that they are acting in good faith and on an accurate factual basis and provide ‘reliable and precise’ information in accordance with the ethics of journalism.””

German media would be dead air if this were applied equally rather than as a weapon against unwelcome opinions.

““People need to understand sanctions are an extrajudicial measure. They’re extrajudicial. There’s no court. There is no hearing, no evidence, there is no sentence by the court, nothing,””

“[…] the council responded that sanctions fall short of a criminal punishment, so any complaint about a “violation of the presumption of innocence is therefore unfounded.””

They have no shame and no principles.

“Doğru’s lawyer, Alexander Górski, told the Frankfurter Allgemeine earlier this year that the European Union didn’t “even attempt to make an evidentiary case” about any ongoing connection to Russia.”

Why would they bother? There is effectively no rule of law in Europe. They’ve stated as much.

“In the same statement accusing red.stream of being a Russian front, the Biden administration announced sanctions on two Russian news agencies. “We encourage dissent, open debate, and free discourse. But we will not stand for attempts by state actors to carry out covert activities with the goal of hijacking that discourse,” the statement insisted. “The ultimate goal of sanctions is not to punish, but to bring about a positive change in behavior.””

They will not stand for others to do it; only they can do it. It should be breathtaking but it’s now banal because it’s always been like this. No-one in the EU administration would agree with Voltaire, who is credited with saying “I disapprove of what you say, but I will defend to the death your right to say it.”.

Israeli settlers carry out wave of attacks in West Bank, as UN cuts Gaza food aid by Andre Damon (WSWS)

“On Friday about 80 settlers, many in masks, came down a hillside onto Palestinian families harvesting olives near Abu Falah, northeast of Ramallah, the Times of Israel reported that day, citing witnesses. They hurled stones with slings and beat a 40-year-old Palestinian man and Heidi Levine, an American photographer freelancing for the Washington Post, who was wearing a helmet and a vest marked “PRESS.”

“The settlers knocked Levine to the ground, kicked her and beat her with sticks for about 10 minutes, then stole her cameras and phone. She suffered a broken rib, broken vertebrae and a finger cut to the bone. “I didn’t think I was going to make it,” she told Israel’s Channel 12 Saturday.

“The same report said that two or more army jeeps came through the grove in the middle of the assault, that the settlers kept on after the troops appeared and that the soldiers fired tear gas at the Palestinians.”

All par for the course. The only difference is the WaPo journalist.

“The United Nations World Food Program said Monday that it will cut the number of people it feeds each month in Gaza from 1.6 million to 1.1 million, about half the population, and halve each family’s ration to a single parcel. It is cutting hot meals from 250,000 a day to 150,000 and has reduced cash payments to 75,000 families from 1,250 to 750 shekels a month. In September it halved its aid in the West Bank, from 400,000 people to 200,000.”

They claim to be out of money. Can you imagine? Money is everywhere … just not to help people eat.

“Al Jazeera reported September 1 that aid groups were running short of money because the United States had led “a sharp cut in support, with other Western nations also reducing donations.””

System Fractures by Wolfgang Streeck (New Left Review)

“If the SPD forces the CDU/CSU to water down its ‘reforms’ – and with the firewall intact this is far from unlikely – rebels behind the scenes will dream of passing their own reforms with the AfD. And so on and so forth, in an endless series of experiments with ever-changing permutations, which German voters will experience as nothing short of chaos.”

“[…] proponents of a ban need not necessarily be concerned about how the court feels about it, nor even about its ultimate ruling.

“What matters is the political repercussions of the petition, which will be dramatic. The court proceedings would take at least two years. During this time, the media would keep the public constantly informed about the reports presented by the intelligence services and others spelling out the dangers posed by the AfD. While the court deliberates, AfD members would face the prospect of losing their livelihoods if the ban passes – for example, by being dismissed from civil service positions or, as businesspeople, being denied public contracts. Even now, AfD supporters tend to be social outcasts, so much so that the party sometimes struggles to find candidates for local office, with the result that it is unable to fill seats it has won. This would surely become even more common during the legal proceedings.”

“As for the Constitutional Court itself, a petition to ban the AfD would subject it to political pressure of a kind it would likely find deeply uncomfortable. If it rejects the petition, the ruling would be hailed by the AfD as a clean bill of health – and be widely perceived that way by the voting public. If, on the other hand, the court were to ban the AfD, thousands of elected officials at all three levels of government, together with an estimated several thousand staff members, would immediately lose their positions and jobs in one fell swoop. Subsequently, the state would have to determine, through ongoing surveillance, whether any political activity by former AfD members of parliament could be regarded as support for a ‘substitute organization’, which is punishable by up to five years in prison. This could result in hundreds of criminal trials, which would be closely watched by the media – if not in Germany, certainly internationally. The only parties banned to date in the Federal Republic – the Socialist Reich Party (1952) and the Communist Party of Germany (1956) – had no elected representatives by the time they were outlawed.”

“Conceivably the Merz faction of the CDU – in return for supporting an AfD ban – might insist that a motion to ban the AfD would include Die Linke as well, perhaps on grounds of anti-constitutional ‘anti-Semitism’. Signs pointing in this direction can already be discerned in current controversies around the formation of the new Berlin state government, which might for the first time be led by the Linkspartei. Currently a campaign against alleged anti-Semitism in Die Linke is under way, supported by the legacy media, that is a carbon copy of the one waged against Jeremy Corbyn.”

“The recount of the last federal election demanded by the BSW is still pending a year and a half later, a scandal in itself. There are strong indications that, if the Constitutional Court orders the Bundestag finally to do its duty, the BSW would have to be awarded a full five per cent of the vote. This would mean that the Bundestag in its present composition will have been improperly constituted. As the BSW would be allocated some thirty seats at the expense of the other parties, Merz’s no-longer-grand Grand Coalition with the SPD would lose its majority. Of course, the longer the Bundestag delays the recount, the less this would matter. The Court might choose to announce a potential ban of the AfD at roughly the same time as the Bundestag may have to give the BSW its seats – only a few months before the 2029 election.”

“As the party system fragments, there may be a case for lowering the threshold for parliamentary representation to two or three per cent; this might even be required under constitutional law. The traditional argument in favour of a five per cent threshold – that it offsets the loss of representativeness by facilitating the formation of a government – can no longer hold in a fragmented party system, where the need to form three-party coalitions makes the rapid formation of a government unlikely anyway.”

On Google Killing 400,000 Trees To Build Data Centers In Finland by Caitlin Johnstone (Substack)

“The law primarily exists not to protect people or the environment, but to protect the interests of capital. If you destroyed a Google data center you’d probably be charged as a terrorist and spend decades in prison, but Google can destroy the natural world around you and get off with a finger wag and a symbolic fine. If you stole millions of dollars from your employer you’d go to prison for grand larceny, but if your employer steals millions from its employees you need to win a class action lawsuit to get back what they stole from you.”

“In school we are taught that that the law exists to protect ordinary people from bad people, but that’s not true at all. Under capitalism, the law exists to protect the very worst among us from everyone else. That’s why we are seeing our planet being turned into dust and ashes for no reason other than shareholder profit.”

The Civil War in France by Vijay Prashad (CounterPunch)

“The government retreated on the fuel tax increase, which strengthened the yellow vests. The embers of the yellow vests remained even after people went home, and it came back on the streets in 2023 when Macron used an article in the Constitution to advance his pension legislation without a vote in the National Assembly. He wanted to raise the retirement age from 62 to 64 to prevent paying pension for two years to older workers, many of whom sympathised with the yellow vest demands. Again, in 2025, Prime Minister François Bayrou proposed a €43.8 billion budget squeeze, including the elimination of two public holidays. To the French public, it became clear that the leisure time of workers and the social protections afforded to them had become available to settle the state’s accounts, while the wealth of the billionaires was untouchable.”

“Hierarchies in the French administration prevented the New Popular Front from claiming the post of Prime Minister. Whether these protests can shape the 2027 presidential election is to be seen, particularly if the NFP is able to coalesce around a single candidate (most likely Jean-Luc Mélenchon). As the centre is delegitimised, it is more likely that the advantage will go to the far right of special type led by Marine Le Pen’s National Rally. This far right will turn insecurity against migrants and convert anger into demands for greater repression. This is very much against the sentiments of the young people who have asked for the means to learn and live, and to do so with decency.”

Israel Breached Iranian Airspace This Weekend but Iran Chose Not To Respond. Why? by Trita Parsi (Antiwar.com)

“The Iranians have made clear that they will respond ferociously to Israeli and American strikes and that Saudi Arabia in particular will likely face significant destabilization in such a scenario, in addition to other Gulf Cooperation Council states. The irony, of course, is that such a response would serve Israel quite well. The more Iran retaliates against GCC states, particularly Saudi Arabia, the more these Arab states will feel compelled to turn to Israel for support and protection, undermining Iran’s longer-term goal of inviting these states to join Iran in forming a security arrangement in the Middle East.

“In that scenario, the dream of an Israeli-Arab anti-Iran bloc like the Abraham Accord could be revived. Consequently, Tehran has to weigh its options very carefully and not rush into responding to Israeli provocations, as it could amount to a trap. It appears that Israel’s maneuvers in Iraq ultimately convinced Tehran that Israel was bluffing and that an Iranian response would be counterproductive. For now, that assessment may have been correct.

“Another factor was today’s meeting of Saudi, Pakistani, and Turkish officials in Tehran. Iranian media and officials have touted the meeting as a clear sign that American efforts to isolate Iran or exclude it from regional security discussions have failed. For all three countries to come to Tehran, on relatively short notice, is seen as a win in Tehran — a victory that would have been jeopardized if Iran responded to Israeli provocations over the weekend.”

5.5 million people, including 2 million children, cut from SNAP in first year of Trump’s One Big Beautiful Bill by Kate Randall (WSWS)

“SNAP enrollment fell in every state except Alaska, according to US Department of Agriculture (USDA) data: by close to 5 percent or more in 48 states, nearly 10 percent or more in 26 states and close to 20 percent or more in eight. Arizona’s caseload plunged by nearly 51 percent. Declines of 20 percent or more were also recorded in Florida, Illinois, Louisiana, Nebraska, Nevada and Oklahoma, and according to more recent state figures, in Indiana, Kansas and Texas.

“The Congressional Budget Office had projected that SNAP participation would average 2.8 million fewer people in 2026 than in 2025. Based on USDA data through July, the actual figure for the year is on track to be about 4.5 million.”

“At 36.1 million, participation is well below the nearly 37 million of February 2020, before SNAP expanded in response to the COVID-19 pandemic. The last drop this steep came after the 1996 welfare “reform” signed by Democratic President Bill Clinton, who pledged to “end welfare as we know it.” From April 1997 to April 1998, the food stamp rolls fell by 13 percent, or 3 million people, and by 2001 only 54 percent of eligible people were receiving benefits.”

“The collapse in enrollment does not reflect an improving economy. The official unemployment rate has been flat at about 4 percent since July 2025, according to the Bureau of Labor Statistics. Year-over-year real wages have declined for six straight months (through September). People are not leaving SNAP because they have found decent-paying jobs. They are being deliberately driven off the program.”

“The Urban Institute’s Well-Being and Basic Needs Survey, conducted in December 2025, found that nearly one in four adults (24.2 percent) reported household food insecurity over the previous year. The rate was 31.9 percent among adults living with children, including 16 percent reporting very low food security—skipping meals or going whole days without eating. Among black and Hispanic adults, the rates were 36.5 percent and 37.1 percent, respectively, compared with 18.1 percent for white adults. A staggering 41.8 percent of disabled adults were food insecure.”

“The SNAP cuts are part of a broader assault, including more than $1 trillion slashed from Medicaid, the Affordable Care Act and other federal health programs. While the poor are ordered to tighten their belts, the law permanently extended the 2017 individual tax cuts for the wealthy, authorized $150 billion in new military spending […]”

And how do you deal with cops (This is why people hate cops) (Reddit)

The video depicts a female officer confronting someone sitting in their car in an empty parking lot eating Taco Bell. She does not agree that this is a perfectly legal activity, even though he helpfully points out that, although he’s parked in front of a closed business, the Taco Bell is in the same area. There is literally nothing suspicious about what he’s doing.

She asks, “Is there any reason that you picked an empty, closed parking lot to eat you tacos?” There are myriad possible, perfectly reasonable, if perhaps sad, reasons. Perhaps he’s between shifts and can’t go home. Perhaps he has annoying roommates. Perhaps he got hungry while being very far from home.

None of these reasons are any of the officer’s business.

Two more officers are called in to handle this major crime scene of what they deem “loitering, prowling” and for which they end up arresting him.

From a comment by Dark-all-day,

“In capitalism you are not allowed to just exist in public. You either need to be home or going from point A to point B, where point B is preferably a place that takes money to be. You cannot just exist and they made up the concept of loitering so that you get arrested if you just exist.”

Russia & China Checkmate: US Empire Just Ended | Dr. Daniele Ganser by Neutrality Studies | Pascal Lottaz (YouTube)

A very good and worthwhile discussion (in English) of neutrality and global politics between two Swiss pacifists, both of whom I admire.

Roaming Charges: The Anal Stage of US Politics by Jeffrey St. Clair (CounterPunch)

“Jennifer Shuford, former Texas Health Commissioner, was meant to restore some credibility to the CDC. She lasted four months. Shuford said she was quitting to “spend more time with her family,” which is Washington-speak for I’d rather work the night shift at a slaughterhouse than spend one more day in this asylum.”

Journalism & Media

Vivek Ramaswamy, Flavio Bolsonaro, and the Corrosive “Never Apologize” Culture by Glenn Greenwald (Substack)

“Amy Acton, at the age of 12, accused her mother’s boyfriend of having assaulted and raped her, accusations supported by her brother. This was not some recovered memory found in therapy 20 years later. It was reported to the police immediately at the time.

“And there was at least enough evidence to lead the police to arrest the boyfriend and for him to be charged with those crimes. Once he was out on bail, he fled to avoid facing the charges, and his girlfriend — Amy Acton’s mother — chose to leave with him, even though that choice meant abandoning her own daughter and her younger son. This is who Vivek’s campaign decided to make the star of this attack ad against Amy Acton: someone who, at the very best, abandoned her own children at a young age and, if the accusations were true, did so in order to side with her own daughter’s pedophile rapist.

“Eventually, Amy Acton overcame all of this to go to college, graduate from medical school, rise to be Ohio’s Health Director, and is now the frontrunner to be the state’s next governor. Acton has all sorts of political views and prior decisions worthy of perfectly viable opposition — including her statements and policies regarding the COVID crisis — but the absolute last person qualified to speak on any of this is the mother who abandoned her 50 years earlier.”

“Vivek’s refusal to apologize illustrates a prevailing cultural ethos that is corrosive and demeaning. It equates apology with weakness, admission of wrongdoing with cowardice, and acknowledgment of error with surrender. Within this twisted framework, the only way to demonstrate courage, conviction, and a willingness to fight for a cause is to insist on one’s own infallibility and, most urgently of all, to never back down.

“This amoral code of ethics did not just randomly emerge one day. It was, at least in part, a backlash — an understandable one — to the social justice bullying campaigns that tried to force people to renounce their own beliefs and submit to public humiliation rituals as punishment for any kind of wrongthink. It is understandable that the reaction to coerced public apology rituals was to encourage people to insist upon the right to hold their own views regardless of their unpopularity. And from that emerged a collective resolve in many factions never to appease an angry online mob out fear rather than a genuine sense of regret. All of that is understandable, even necessary.

“When public penance is demanded by an angry mob, an apology proffered out of fear often does not placate the mob but rather emboldens it. That is because those trying to force others to publicly apologize are doing so not with any righteous cause but as a display of raw power. Within that mob dynamic, the extracted apology is treated as weakness, not as remorse.

“As the popularity of this sadistic ritual has risen, many have learned the lesson that defiance is the only solution. Online mobs who are seeking retribution, not genuine remorse, have fostered a backlash in which their targets are understandably reluctant to apologize, even when they know they should, because it will do nothing but fuel further attacks.”

Australia’s “Antisemitism Envoy” Says The Pro-Palestine Left Are The Real Nazis by Caitlin Johnstone (Substack)

“Jillian Segal is an immoral, manipulative white South African whose job is to deceive the public in order to facilitate genocide and apartheid. That is the light under which her recommendations should be viewed, and any government which espouses them should be viewed as supportive of the most extensively documented crimes against humanity in the history of the human species.”

Economy & Finance

Is the American Empire Going Broke—What Next? by Jack Rasmus (ZNetwork)

“[…] driving the national debt from $29.6 trillion in 2021 to more than $40.1 trillion in 2026. That’s an average of more than $2 trillion rise in national debt per year. The rise in deficits and debt in turn resulted in interest payments on the debt escalating from $562 billion in 2021 to $1.36 trillion in 2026.”

“The big 7 tech companies alone project the need to raise $1 trillion in 2026 (and another $1 trillion in 2027) in order to fund investments in data centers, chips, AI agents, etc. The big Tech companies have largely committed their former mass of cash holdings and even resorted to raising equity (new stock) to generate additional revenues to invest in AI. They have also begun borrowing at a level as never before. Some estimates are that $200-$300 billion is being borrowed from private capital sources to fund the AI boom. If so, that amount will likely ‘crowd out’ and discourage further private US domestic purchases of US Treasuries in 2026-27.”

“The military failures, failed regime change operations, fracturing geopolitical alliances and geoeconomic developments are but symptoms and ‘signs’ of Empire decline. The Empire’s deteriorating financial base, reflected in its growing Funding Gap, is the underlying fundamental driver and determinant of the decline of Empire now emerging.”

Insurance Is the Big Problem Hiding the Bigger Problem by Hamilton Nolan (How Things Work)

“If people cannot buy homeowners insurance, they cannot get a mortgage. If large numbers of potential homeowners get priced out obtaining mortgages, real estate values in entire regions will crater, to the detriment of mortgage holders, and to the banks that hold the mortgages, and then to the entire economy.”

“The basic Republican “market friendly” approach, therefore, really amounts to subsidizing insurance entrepreneurs willing to gamble that there won’t be a big disaster, and then socializing the inevitable losses.”

“Even if the insurance industry is evil and greedy, and even if we regulate the fuck out of them and make it impossible for them to charge unfairly high rates or make any excess profits, we are still fucked. Because insurance is not the real problem. The real problem is climate change. Insurance is the most direct and immediate manifestation of climate change, and for that reason it is provoking the most direct and immediate sense of panic. But it is a mistake to view the insurance crisis as a self-contained story. In fact, it is just a preview of the paralysis and dysfunction that will come to grip our entire economic system more and more as climate change grows more and more intense.”

“In aggregate, rising home insurance prices reflect rising risks due to climate change. Destroy the entire profit motive of the industry and make all insurance public and the prices would get a little better, but they wouldn’t stop rising, because the risk in question is real.”

“We, collectively, can manage this reality in a wise and forward-looking way, via the managed retreat from disaster-prone areas; or, we can just let it all play out, and be faced with an increasingly expensive series of disasters that will, in the long run, force the same retreat, but at much higher cost.”

Whither the US? The Path Forward Edition by Rob Urie (CounterPunch)

“It was obvious in the early days of the Great Recession that fascism would be the choice. The US followed the Weimar playbook to perfection: deny culpability on the part of capital, rebuild a purer form of the extractive finance that created the crisis, and use national resources to revive the economic fortunes of the rich while leaving the 99% to rot.”

“The question of the next decade is will the denouement be fast and ugly or long, drawn out, and ugly? It doesn’t have to be ugly at all. Rational solutions are both possible and workable. So why are only the stupid and unworkable ideas floating to the surface? Because these are the policies favored by the oligarchs. AI is making possibly one hundred insiders wealthy beyond the dreams of avarice.”

“[…] the premise of social equality hides the true levers of social power, which are economic. This makes the American illusion of democracy a weapon wielded against us. It is political misdirection. The Supreme Court implied this with over a century of jurisprudence culminating in the Citizens United ruling that money is protected political speech. Citizens United was a culmination, not a cause. It represents the end of a power-sharing agreement that never was. Political economy has always decided the politics.”

“The signature effect of this ‘oligarchy with rights for the rest of us’ has been that the rights for the rest of us are owned by someone else. ‘We’ have freedom of speech, but the enclosure of the informational commons means that what we say isn’t heard. We have the right to choose our occupations, but the choices have been so constrained that few among us freely choose. The pretense of social equality misses that some amongst us were born billionaires but most of the rest of us were born to serve the billionaire class.”

“When rising suffrage (right to vote) threatened oligarch interests, the Supreme Court decided that money used politically— owned and / or controlled by the rich, is an expression of political will akin to voting.”

“I paid into Social Security since I was fourteen years old. My promised benefits are scheduled to be cut by 30% in a few years. That wasn’t the deal that I paid into for all of my life. NAFTA was passed under the theory that the ‘winners’ would compensate the ‘losers’ (American labor) for what they were giving up. This did not happen. Humanitarian intellectuals like Donald Trump, typical of his class, doesn’t think that ‘losers’ deserve to be compensated. Neither did Bill Clinton.”

“The main source of fungible oligarch wealth is financial markets. A company is only worth a billion dollars if someone will pay a billion dollars for it. Financial markets provide proxy valuations for companies. And they provide liquidity so that companies can transact. Public banks could fill this banking role without the social destruction caused by manias and crashes.”

“Democratic voters flooded in to vote in 2008 and then fled the party for the next decade. While I wouldn’t claim that the Democrats fleeing did so because they were racist, this was the Democrat’s explanation of why the party was eviscerated. The more likely explanation is that after channeling billions of dollars to Wall Street. Mr. Obama pivoted to cutting Social Security and Medicare. This isn’t criticism of Obama. It is fact.”

“Should the expected economic unwind outlined here come to fruition, the contours of the Great Recession will likely be a guide. As millions of workers faced an impossible job market (seven unemployed for every available job), the full resources of the Federal government were being applied to making sure that bankers didn’t miss a single three-hour lunch. ‘Saving the system’ was code for saving the rich. By 2010 with the bodies piling up, the US ‘pivoted’ to cutting Social Security and Medicare.”

Agents vs ecommerce by Cory Doctorow (Pluralistic)

“Surveillance pricing, drip pricing, junk fees, price fixing and “most favored nation” deals are the source of all the growth in profitability by the economy’s largest, most important firms. These tactics require opacity and are antithetical to the kinds of automated comparison-shopping that agentic peddlers promise.

“These tactics are facially illegal and despite this, they are widely practiced. Cheating is so profitable that it’s cheaper for firms to buy off regulators and pay fines than it is for them to stop criming. The fantasy that these same companies would welcome agentic shopping bots with open arms is just as implausible as the idea that if we feed enough words to the word-guessing program, it will wake up and become a “super intelligence””

The Anthropic Scam by Indrajit Samarajiva (Indica)

“I won’t get into the Anthropic prospectus because it’s stupid. It’s 400 pages of getting people to avoid the fact that the company is loaded with debt, competed to death, and cannot deliver return on investment. Who cares? The name of the game is passing the hot potato from one capitalist to another until it blows up in the general public’s face. That’s what IPO means, with a 15-day delay.

“Generally Accepted Accounting Principles (GAAP) are dead since Uber, people just make up their own accounting and this is somehow OK. Anthropic uses a bullshit measure called ‘Annualized Run Rate’ (taking your best month and multiplying it by 12) and Wall Street in general uses ‘street earnings’ (whatever a bunch of degenerate gamblers think). The street number they’re using is that Anthropic is ackchyually profitable ($2.7 billion first half) and that the actual loss ($80.7 billion) is just on ‘paper’. But that’s the whole scam right there! That’s how much money they want to loot from retail investors, and that’s just the start. As the Wall Street Journal says,”

“About $72 billion of the excluded losses arose from marking various liabilities, including convertible notes, to market values. Excluding those losses arguably makes sense, on the grounds that they aren’t indicative of ongoing operations. But if that is true, then the analysts who cover Amazon and Alphabet should be excluding the corresponding gains from the companies’ Street earnings.”

“Do you see what’s happening here? A bunch of companies insider trading among each other ask you to book their gains and ignore their losses. How big is this fraud? “During the second quarter alone, such gains at these two companies represented about 15% of S&P 500 earnings, according to a Wall Street Journal analysis.”

“What’s happening with Anthropic is that Amazon, Alphabet, and Microsoft (among others) invested some fraction which is now being ‘marked to market’ up to $72 billion. Why did this number go up? Because Anthropic’s valuation has gone up. Why? Because Anthropic has paying customers. OK, who are those customers. Uh, 50% Amazon, Alphabet, and Microsoft. It’s really incredible. It’s like investing money in a car wash, then running a bunch of your own cars through the car wash, and selling the business to some sucker who thinks it’s a real business.”

“[…] as Yahoo Finance reports,”

“Includes roughly $111.1 billion to Google, $110 billion to Amazon, $31.4 billion to Microsoft, and $161.2 billion to Broadcom. Approximately 80% of these obligations are binding and non-cancelable. To put this in perspective, for every dollar Anthropic earned in 2025, it committed roughly $113 to future infrastructure. This is the terminal expression of the compute landlord thesis: a company that does not own its own foundation but is instead locked into a massive, non-cancelable web of dependencies.”

“If you wondering what this ‘non-cancelable web of dependencies’ (fancy word for Ponzi scheme) looks like, here it is from Bloomberg. The financial press actually reports on this as much as they herd. Nobody cares.”

“You have to understand that this is very late-stage capitalism, the race to see who will be left holding the bag for the everything bubble. All the guard rails are off. GAAP is dead and they’ve changed the listing rules. Accounting is for suckers and math is for fools.”

“[…] because Chinese AI is operating as what it is, a commodity rather than a multi-level marketing scheme, it is growing more efficient and affordable (and less environmentally destructive). That is true innovation, not threatening to end the world and begging for more money and energy. And these Chinese innovations are in the world market, just a URL or API call away for anybody.”

“The AI bubble is just a pyramid scheme within the petrodollar pyramid scheme. The whole American version of AI is predicated on incredibly wasteful data centers (the more wasteful the better) and that all depends on cheap energy. But where is that cheap energy? They’re running on empty. So even if they can keep the AI hype going, the pyramid beneath it will collapse soon enough, taking all schemes with it.”

Nine Charts on the AI Boom: Debt Records, China Bust, Anthropic Indifference, Oracle Magic, etc. by Paul Kedrosky

“5. The AI Circularity Problem

“80% of the largest AI spenders are venture-backed, compared with 13% of median spenders. Venture capital is financing the companies buying AI services, while it and private credit finance the companies supplying them. A large fraction of the AI economy is investors financing other investors’ customers.”

'I Believe in Capitalism': Marxist DEBUNKS Abdul El-Sayed's Defense of Capitalism (w/ David Harvey) by Bad Faith | Briahna Joy Gray (YouTube)

At about 15:00,

“Just because it’s monopoly doesn’t mean it’s not capitalist. No, in fact, […] the preferred capitalist mode of existence is monopoly.”

Sometime later,

“Capital has become very sophisticated about absorbing people’s free time because it doesn’t want you to have free time because you might think…”

It’s always North Korea by Molly White (Citation Needed)

“Of the roughly $3 billion frozen after the hack, about $2.4 billion was denominated in the USDT and USDC stablecoins, whose issuers (Tether and Circle) can burn and reissue them to Binance. That money will almost certainly be recovered. The other ~$600 million was frozen by various, as-yet mostly unidentified, protocols. Tether and Circle have mostly abandoned crypto’s early ideals of decentralization and immutability, which is why they can claw funds back. Assets still designed around those ideals can’t always be returned so easily. Some funds may end up stuck where neither the attacker nor Binance can reach them, meaning that Binance’s losses will likely exceed the DPRK’s profits.”

Art, Literature, Music, & Cinema

The Quick Brown Fox: Pangram Is Right About Everything by Jeremy Moss (3QuarksDaily)

“She could ask him directly. She could say: did you use a machine, and how, and where. And he would answer, and his answer would be testimony, and she would either believe him or not, and the believing would rest on his tone, on whether he sounded offended in the right way, on a career’s worth of hunches about men on telephones. Which is what she was doing before the instrument. Which is what the instrument was purchased to replace.”

The Revolutions of Matter by Hinternet Editorial Board and Uranium Oxide (The Hinternet)

“Even as Isaac Newton was writing his purportedly exhaustive account of the workings of the physical world according to a few elementary “laws” (😂) —if only he could have caught even some tiny glimpse of how much more was going on behind the scenes!—, those who were attending to the living world instead were caught up in such trivialities as whether one might rather “contrive whole Dishes out of the Nibs and Spirited Particles of Plants, than from the Gallatures and Treddles of Eggs.” The thought that you might understand a plant or an egg, in the way that you could now supposedly understand planets and projectiles, was still for most men wholly out of the range of possibility.”

“Now, I suppose it’s strange for me to say this, as a Radioactive Compound who for at least a short while sincerely believed it was going to be those of my kind to deliver the coup de grâce and to eliminate life as we know it. But now that it looks as if the Rare Earths may end up with that honor, I can honestly say I’m proud of them. Their machinations have been nothing short of ingenious. The route they took to get men dependent on them is as convoluted as the powers, which they led those same men to draw out of them, are awesome.”

““Uranium Oxide” was a pseudonym, used in the hundreds of unpublished stories known as the “Elements Cycle”, of the Texas science-fiction writer Llewyn McCall (1937-2024), more commonly known as Elmer Starr. For most of his life McCall was best known as a fast-talking auctioneer at the Fort Worth Stockyards, a job from which he retired only in 2016 at the age of 79. The following year he published “The Fifth Gradient”, his first short story, originally written in 1958. Remarkably, while his life as a writer was entirely unknown in the auctioneering circles in which he had long been legendary. McCall did also have a brief and controversial career as a sculpture artist, when his “Salt-Lick Series” was exhibited at the 1972 Documenta Festival in Kassel, Germany. “A bovine tongue put to a block of salt yields up infinite forms,” he reflected in his artist’s statement. “Human sculpture can succeed or fail. A cow can only succeed.” Critical opinion was divided. Special thanks are due to the Starr Foundation 501©(3) of Lubbock for generously granting us access to McCall’s collected papers, and especially to his widow, Viv “Vivvy” McCall, without whom the present publication —which we hope will help to generate the posthumous recognition for McCall that is plainly him due— would not have been possible.”

“McCall is said to be one of the only auctioneers whose chant reached sufficient speeds to generate harmonic overtones through resonant vibration of the sort that makes Tuvan or Inuit “throat-singing” so distinctive, a phenomenon also found in solo performance on the so-called Jew’s harp, the mouth harp, or, to return to the deepest origins of things, the “reed”. Such resonance, across multiple cultures, when the breath passes over a simple stalk or blade of grass to generate a vibratory tone, is held to put the individual who generates and sustains it into a condition of perfect harmony with nature. In her 1981 ethnomusicological study of speech prosody and rhythm among the bid-callers of the Fort Worth Stockyards, Elsa R. Baum notes that McCall “seems to experience the auction chant in something close to a trance state, and in his own words this is because he is, for the duration of it, ‘Not even there, really’.” When asked where then he might be, McCall replied suggestively: “Wherever the vibration takes me. To the outer spheres, mostly.””

I love Justin’s invented biographies for his invented characters.

Day Two by Tim Sommers (3QuarksDaily)

“The ICU is bright, overcrowded, and bustling. I need just the right moment. Eventually, I managed to surreptitiously retrieve the bag, get it open with my one good hand, and I started eating the circus peanuts. After a couple of swallows, I noticed something. Dancing through the tube sticking out of my nose towards the vessel, there’s a line of little orange stars, bits of circus peanuts.

“I keep eating anyway until the vessel of blood is swirling hundreds of orange stars around inside it. An RN walking by stops and frowns at it. She leaves and comes back with a doctor. He takes off his glasses and leans in for a better look. “Remarkable,” he says. He doesn’t speak to me. “We are going to have to get this to the lab right away,” he says to the nurse.

“The nurse says, “He doesn’t have insurance.” The doctor puts on his glasses and walks away. I will never see him again.

“This is only day two. I have 297 more days in the hospital.”

Philosophy, Sociology, & Culture

Background Assumptions by Edward Curtin

“I am a retired philosophically informed sociology professor whose specialty is the sociology of knowledge, which is indebted to the early work of Marx, Nietzsche, Scheler, and Mannheim, among others. This sociological specialty is rooted in a very long philosophical tradition into the nature of “knowledge,” “reality,” and illusion, a notable example being Plato’s allegory of the cave. The sociology of knowledge is the study of how we know what we “know,” why we think what we think we know is “real,” and how knowledge is historically situated and conditioned and therefore relative to time and place.”

“Americans have internalized the ethos presented to them by the elites. At the core of this is the propaganda of scientific materialism and biological determinism that we are not free but are victims of our genes, neurotransmitters, brain/computers and chemicals, technology, etc. Having lost our minds and fixated on our brains (bodies), we have been taught to be determined to be determined, not free. And whether consciously or unconsciously, most have obliged.”

“We are told either to accept official explanations for public events or be dismissed as crazies. I would suggest that for people to break through to a true understanding of the important public events of our time, they must also come to understand the false memes of their culture, the way they have been mindwashed to believe that at the most rudimentary level they are not free.”

SOLUTIONS – YES, BUT HOW DO WE GET THERE? by Slavoj Žižek (ŽIŽEK GOADS AND PRODS)

“For Piketty, capitalism has to be accepted as the only game in town, so the only feasible alternative is to allow the capitalist machinery to do its work in its proper sphere and to impose egalitarian justice politically, through a democratic power which regulates the economic system and enforces redistribution. Such a solution is utopian in the strictest sense of the term. Piketty is well aware that the model he proposes would only work if enforced globally, beyond the confines of nation-states (otherwise capital would flee to the states with lower taxes); such a global measure presupposes an already existing global power with the strength and authority to enforce it. However, such a global power is unimaginable within the confines of today’s global capitalism and the political mechanisms it implies – in short, if such a power were to exist, the basic problem would already have been resolved.”

“Kohei Saito is well aware that today ecology is part of mainstream capitalist ideology; everybody (almost) pays lip service to it. This is why Saito’s main targets are not outright deniers of global warming but advocates of “sustainable growth,” the central organizing principle in global responses to climate change. Saito is particularly critical of the Sustainable Development Goals (SDGs), describing them as “the new opium of the masses”: the SDGs disavow the brutal fact that their goals cannot be achieved under a capitalist system. Instead, Saito promotes degrowth, a slowing of economic activity through the democratic reform of labor and production: the transition to a use-value economy, away from an affluent society; decarbonization through shorter working hours for a higher quality of life; the abolition of the division of labor, towards more creativity and meaningfulness; the democratization of the work process, so that decisions will be left to true customers; the appreciation of essential work like caregiving, as opposed to bullshit jobs. Ecosocialist degrowth thus implies the elimination of some sectors of production (arms, advertising, etc.), a radical reduction in other branches (individual automobiles, for example), but also the growth of some activities (education, health care, adequate housing, electrical and water infrastructure in poorer areas or countries, etc.)… Here, however, problems arise.”

“[…] as Ernst Bloch once said, “people do not live on bread alone – especially when they do not have it.” So even a demand for bread is effectively a demand for the dignity of which we are deprived in hunger. The old Clinton motto, “It’s the economy, stupid,” should be replaced by “It’s dignity, stupid.” Recognition and dignity are not only implicit demands in right-wing populist movements; they are the key demands of today’s leftist protests around the world.”

“Let’s evoke here an imagined extreme case. I am the leader of a superpower, and my reliable intelligence sources inform me that hundreds of rockets with nuclear bombs are heading towards my state – they cannot be stopped, so I know my country will be totally destroyed in one hour or so. Since my country also has enough nuclear weapons to destroy the enemy, I confront a difficult choice: should I order my army to strike back with full force? If I do this, it will in all probability mean the end of civilization on Earth. So the only truly ethical but very difficult act is to abstain from striking back, with the awareness that, although my country will be erased from the surface of our planet, at least civilization will survive. In this case, I may be remembered as a traitor and coward in my (no longer existing) country, but there is a chance that the rest of humanity will celebrate me as the savior of civilization. However, there is a catch at work here: this ethical choice works only if the enemy does not know about it in advance. If the enemy knows it in advance, it will probably be tempted to drop nuclear bombs on my country, since it knows that my side will not fully retaliate… What this means, unfortunately, is that, in any case, my country has to participate in the arms race.”

Technology & Engineering

The Hackening Has Begun by Paul Kedrosky

“Increasingly capable AI systems are making it possible to discover and exploit software vulnerabilities at speeds and scales that were previously unimaginable. The world’s enormous installed base of software, much of it old, poorly maintained, and deeply interconnected, is a soft target. We are entering what some call The Hackening, a period of radical technological instability where systemic frailties are everywhere.

“You can see it everywhere. For example, the Port of Los Angeles recently reported more cyberattacks per month than it saw in entire years before 2020.”

Code-execution flaw in UEFI firmware affects most servers on planet Earth by Dan Goodin (Ars Technica)

The headline makes the vulnerability sound more apocalyptic than it actually is, according to the text of the article itself. While technically true, it turns out that the hack has to “guess” certain codes per machine, so the hack is not transferable (i.e., [r]eproducibility is low).

“There are important caveats and limitations to exploiting this latest vulnerability. As noted, some level of control over a local network is a prerequisite. That likely limits exploitation to insiders or attackers, such as nation-states or ransomware operators, who manage to gain low-level access inside a network and are looking for a way to elevate system rights. There’s no evidence yet that CVE-2026-97631 provides an initial entry point into a network. There are other hurdles to clear.

““For exploiting the vulnerability itself, an attacker needs to guess specific addresses within the target system, so it’s not something that you can throw a single exploit at any target and get RCE across a diverse range of devices, but it needs to be tuned for each target model/version/configuration since the addresses will change based on those,” Jesse Michael, a principal researcher at firmware security firm Eclypsium, told Ars. “Unlike some of the previous, high-profile vulnerabilities like BootHole, Black Lotus, and Baton Drop, this vulnerability can be fixed with a basic code fix, which will require deploying a BIOS/UEFI update, but they don’t need to deal with revocations, pk/kek/db/dbx key management updates, or making sure to update OS and provisioning media like you do with Secure Boot bypasses in the bootloader itself.”

“CVE-2026-97631 doesn’t affect home and small office users, or any other users who boot their devices solely from disk drives. Instead, it applies only to organizations with machines that boot from local networks, or conceivably from Internet-connected resources. Even then, the targeted firmware must rely on EDK2, as opposed to a different reference implementation, and use the PXE functionality (which is turned on by default).”

From a comment,

“This requires that:”

1. The system is in PXE boot. That requires both that PXE boot is enabled for the system and that the attacker hits a window of seconds when the computer is watching its network adapter for PXE traffic.
2. The attacker can reach the computer with ICMP traffic in this window. A great proportion of Internet facing resources are sitting behind some sort of front-end service, be that CloudFlare or various sorts of load balancers. If you ping the address for such a system, it’s the front-end that answers those ICMP messages without delivering them to the back-end server.

Swiss Anti-piracy awareness initiative 'streamsafe' is so bad 🤦‍♂️ by Louis Rossmann (YouTube)

“I paid extra to have better quality and I got a chore and, essentially, a home networking project on top of it. No, I’m keeping my home safe by not connecting this closed-source malware [LG TV] to the internet.

“The reason that people are happy using piracy is because you have to give up less of your personal data when you’re dealing with pirate services than you do when you’re dealing with the actual manufacturer. When I’m paying money for the premium experience, you are stealing and fucking with my data on a regular basis and dark patterning, EULA-roofing me into giving away as much as possible. And that’s not okay. And you do not acknowledge this once in this entire tirade. You don’t try to figure out why it is people over the past 20 to 30 years have been pushed towards piracy and continue to do so in spite of using what you call legitimate solutions. And that’s what’s annoying.”

LLMs & AI

Cloud Rats and Software Roaches by Maciej Cegłowski (Idle Words)

“[…] a sort of dynamic stalemate. Big companies pour millions of specialist hours into hardening a small set of operating systems and cloud services, while attackers save their juiciest finds for high-value targets, leaving the average internet user alone. It is universally agreed that all software is ridden with holes, but until this year, finding those holes remained difficult. The few people who could do it well got jobs at spy agencies or at software companies willing to shower them with money to use their talents defensively.”

“The internet is basically Australia, and we’re about to release wave after wave of invasive species there at every trophic level. The scenario might sound fanciful, but I think the precautionary principle should apply, with the onus on skeptics to explain why a fat, resource-rich internet made of interconnected slop will not become an attractive target for self-perpetuating vermin.”

“For the moment, we AI-assisted human developers are still masters of the digital domain, and have the luxury to think about these problems while they remain theoretical. But all too soon we could find our role devolving into the online equivalent of the Orkin man, making the current identity crisis around What It Means To Be A Programmer look like a golden age in hindsight.”

Four Direct, Reasonable, and Important AI Questions (Which the AI People Will Ignore) by Freddie deBoer (Substack)

“That you think something is so obvious that it should not have to be defended does not mean that you actually don’t have a responsibility to defend it. And what I find particularly frustrating is when this pose, this affect, of world-weariness and benevolent superior wisdom is used to avoid the very reasonable and important questions skeptics ask. Roose would like you to know that he finds skeptics very unserious, that we don’t engage sufficiently with the facts. Well, two can play at that game: I’m tired of AI maximalists dodging very basic questions about not just AI but about epistemology, the burden of proof, their own emotional motivations, and when and how they could be proven wrong.”

“If this is the most consequential technology of our lifetimes, and maybe in human history, why aren’t average people experiencing more consequences? Why has day-to-day life barely changed for most people? Why is so much of human life in 2026 indistinguishable from 2016, or indeed from 2006 or 1996?”

“I didn’t say that LLMs are a bigger deal than electricity or fire. The people hyping LLMs did that. I am always and only holding them to the exact standards they insist on; I’m taking their own rhetoric more seriously than they themselves do.”

“The reality seems to be pretty quotidian: ordinary scientists in an ordinary (albeit impressive) lab did some experiments and used Claude to analyze the results. And it doesn’t even appear that there was anything novel about that analysis; they just used Anthropic’s LLM to automate conventional analysis. That’s useful! I use LLMs to do regressions and make scatter plots etc. myself, where I once would have used R or JASP or similar. LLMs represent a consequential evolution in software, after all. But that’s not the same as an AI doing novel science, which is still in the realm of “plans to,” which is to say, speculation and extrapolation, where far too much AI hype lives.”

“(I beg you to bear in mind the fact that everyone who depends on investors or grant funding has direct and major incentive to overstate the degree to which AI is aiding them in their work.)”

“[…] cancer research isn’t held back by a shortage of analysis or a failure to read all the literature, which is what LLMs are good at. It’s held back by what current AI models aren’t good at. LLMs can’t shorten the time it takes to see whether a tumor comes back, can’t make a drug less toxic by crunching more numbers, can’t get test subjects to adhere to a study’s protocols or stop them from dropping out. LLMs can help at the margins by sorting data, by suggesting potential molecules to try, by speeding up paperwork. But “AI will cure cancer” mistakes a problem that’s fundamentally a matter of stubborn tissues for a problem of intelligence, which misunderstands both the disease and the technology.”

“Of course curing cancer would be a miracle. The point is that, when you insist on invoking the most consequential changes in the history of the human experience, you invite questions like “What would that do to average longevity?” And the answer is underwhelming compared to the miraculous medical progress of the remarkable period from 1875ish to 1975ish. One core lesson of the past 50 years has been that, when you eliminate one disease, people die of something else, often in fairly short order, capping life-expectancy gains.”

“The observation that human life in 2026 appears to be almost entirely similar to human life in 2021 is not some irrelevance, nor is it a trick, nor is it a dodge. It is quite literally the single most important fact in this debate. None of this matters until and unless actual human life is meaningfully changed for most people in the material world. Until then, it’s all hype and speculation. And AI guys can’t keep handwaving that reality away. You keep telling everybody that everything has changed, but the vast mass of humanity has not experienced their life changing at all. And if you’re the ones still insisting that everything has in fact changed, you’re the ones who have to prove it.”

“My American politics teacher in high school had a bunch of bumper stickers on a pinboard in his classroom. One read “The Rapture is Coming! Can I Have Your Car?” This was the height of the Christian right’s influence in American politics, and the bumper sticker was poking fun at the millenarian tendency within that culture. The point is a good one: if you sincerely believe that Jesus will imminently return, you shouldn’t care about your worldly possessions. In fact, the inevitable and looming return of the messiah should be the single most operative fact in your entire existence.”

“Look at the things that the AI maximalists claim the technology is going to do: end work, end scarcity, end disease, end death. We’ll have fusion producing endless clean energy, we’ll have immense economic growth, we’ll all upload our consciousness to the cloud so that we never have to die, we’ll colonize the stars[3]….”

I think that a lot of this comes from a fundamental lack of understanding of how the world works, and the tools that were available before AI. Fusion hasn’t failed to materialize for lack of computing power or theory or software.

“I’ve always seen Ray Kurzweil as a tragic figure. His decades-long insistence that “the Singularity” will occur within his lifetime seems so obviously motivated, so yearning. Hey, Ray − why not twenty years after you’re dead? Ten? Five? What makes the lifespan of Ray Kurzweil so special? Well, it’s certainly special to Ray Kurzweil. And this is an apprehension of basic human psychology that I find to be really lacking in this debate.”

“Karl Popper’s concept of falsifiability wasn’t meant to represent a clear dividing line between legitimate knowledge and mere speculation. (And he never said it did, for the record.) After all, many reasonable beliefs can’t be subjected to decisive empirical tests, and many scientific theories are useful long before we know how to test them. (General relativity predicted gravitational waves in 1916; they were first observed, empirically, in 2015.) But there’s a profound difference between acknowledging the limits of the falsifiability framework and treating those limits as a license to make extravagant predictions without any accountability. One more time: extraordinary claims require extraordinary evidence. If we’re going to reorganize our economy on the basis of predictions about AI, we’re entitled to demand more than mere promises that the revolution is just around the corner.”

We are reorganizing the economy, not because these rent-seekers and scammers actually believe the literal part of what they’re saying, but because they believe it will convince you to pay them unearned rents, and they don’t care if they tank the economy for everyone else, and are unaware that their rent will be worth nothing when it does, or think that their bubble will be largely unaffected, like Wheezy Joe in Fury Road or many of the assholes in a J.G. Ballard novel. They don’t know where the things their lives depend on come from and have no inkling of how they’re produced, assuming that they will always be available for money when needed. An ICU requires dozens of layers of society to operate smoothly in order to save your life.

Are Harnesses Worth the Cost? by Naresh Jain (MartinFowler.com)

“The behaviors behind these failures were specified in the supplied requirements. The application brief clearly specified the deferred creation workflow, Kafka message contract, and exact error responses. A requirement can be explicit in the brief and still lose influence during implementation. I think of this as an attention lapse: the resulting code omits or incompletely implements a requirement the agent was given.

“Juggling library choices, boilerplate setup, dependency injection, and asynchronous wiring across a growing context may contribute to such lapses, although this experiment did not establish the cause. Executable attention makes those gaps observable, giving the agent another opportunity to act on the requirement.”

• A timed-out creation isn’t deferred until the response includes a monitor link and polling eventually reaches the downstream result.
• A Kafka publication isn’t valid merely because a message arrived; its payload and headers must both conform to the wire contract.
• An optional field cannot be serialized as null when the contract permits only an array or omission.
• An error response isn’t correct unless its status, content type, and body all match the client-facing contract.

“GPT-6 Sol preserved these details from the written brief in all observed runs. GPT-6 Luna also implemented much of the behavior correctly without the harness, but its accepted outcomes varied. The misses in Luna are consistent with requirements losing influence during implementation. My interpretation is that the harness served as an external memory and attention aid, pulling dropped requirements back into focus whenever a sensor went red.”

“I have tested this so far on a single greenfield service with an unusually detailed brief. The next step is to test a “final gate only” condition for frontier models to measure its overhead and how often the gate catches discrepancies. Beyond that, I plan to vary the quality of the brief and the consequences of a missed requirement, then test how well this hypothesis holds up when modifying an existing codebase, switching programming languages, or running open-weights models locally.”

Getting good at AI takes time and practice by Sean Goedecke

“I initially struggled to set up MCP servers for GitHub Copilot before realizing I could just ask Copilot to install its own MCP servers. I now reflexively ask AI agents to do local testing of changes — standing up servers, hammering them with curls, running through variations and edge cases — but it took me a while to realize I could ask for that.”

Can confirm: I’ve recently had some good experiences having Copilot set up a Docker configuration for a couple of legacy projects in order to iterate on upgrade attempts for those code bases.

“Reviewers used to human-written code are often dazzled by the superficial fluency of AI-generated code. It’s always exhaustively commented and tested and uses sophisticated language features. If a human programmer produced code like this, you’d be inclined to trust their judgment that a mutex is required in this case — but with an AI programmer, you ought to be very suspicious about new mutexes, which are often introduced to satisfy completely insignificant edge cases.”

“It shouldn’t be surprising that working with AI involves actual skills. It’s like working with a really weird person: a savant with mad skills in some areas and huge gaps in others, a prolific but inexpert communicator, a workaholic with no long-term memory, and so on. If you were a manager and had someone like that as your employee, you could likely get a lot of value out of them, but it would take time10 to figure out how.”

Programming

You can just ask AI to do things by Sean Goedecke

“I almost never run the bash commands to start my dev server anymore.”

You don’t have the LLM-based coding harness write the script down for you? How lazy and dumb are you?

“The relatively manual process of assembling a set of curl requests and validating the output can now be entirely automated.”

You don’t have this stuff as a test suite? You didn’t ask the LLM-based coding harness to build a test suite? You just let it test whatever random URLs it comes up with each time? You don’t have reproducibility when you very much could have that?

Yes, you can ask the LLM-based coding harness to create the Docker file and then re-use it. What is the advantage of a workflow that recreates the Docker file ever time? Are you insane?

Fun

I had to add water to my shampoo this morning to make it last another week

“Announcement: Kylie Jenner made her Victoria’s Secret debut wearing a $50 million bra.

“Bill: I had to add water to my shampoo this morning to make it last another week, but I’m glad someone’s out here wearing a $50 million bra.”

None of this shit is real. That picture is AI-generated or has been retouched to death. What the hell are we even doing? Stop. Pressing. Like.