SpaceX’s initial public offering (IPO) attracted intense interest for 2 trillion reasons. But one spectacular feature of this historic launch has barely blipped the public’s radar screen. 

Radio spectrum, the “invisible resource,” is the superhighway of the modern economy. Yet a federal agency has traditionally controlled it, defining exactly what wireless technologies go where and dictating how they roll out. These intricate rules limit some rudimentary forms of interference, but typically block much more—throttling competition and innovation. 

The Federal Communications Commission (FCC) took over three decades to figure out how to create cellular licenses, freezing a World War II technology until the administration of President Ronald Reagan. But since then, market mechanisms have, on occasion, advanced. License auctions commenced in 1994, and in some allocations “flexible use” spectrum rights were authorized. Competitors could define their own services: Why not texting, data, or video, not just voice? Why not digital instead of analog? The FCC now, in the most active bands, lets the market decide. It is hardly a coincidence that they are the most active. 

Rivalry has flourished, and a whole new wireless world has arrived. The iPhone and the competitive Google/Android platform have created a lush ecosystem for thousands of software developers. Applications from Pokémon to YouTube to Waze to WhatsApp to Lyft to Kindle have spontaneously taken seed. 

These apps overlap in frequency space and potentially conflict—just the nightmare scenario regulators asserted authority to protect us from. Yet mobile carriers endowed with exclusive control over frequency spaces, including the right to experiment, seamlessly manage. That coordination enables the mobile technology you now take for granted. 

But this dramatic shift in policy is still controversial. Look to the skies, and follow the trajectory of Elon Musk.

U.S. space communications launched with the Communications Satellite Corporation, or COMSAT—a monopoly consortium dominated by AT&T and the U.S. government—in 1962. The structure assumed the sector would tend to a “natural monopoly.” But the thesis proved false. When the open skies policy was unleashed circa 1972, eager rivals such as RCA, Hughes, Western Union, and GTE rushed in. Prices to transport data were slashed and traffic volumes exploded. These efficiencies enabled the birth of cable television networks, e.g., upending the NBC-CBS-ABC broadcast TV cartel. 

By 2000, there were about 650 satellites in the skies. About half were GEOs, using geosynchronous orbits 22,236 miles high, beaming footprints to Earth that do not move. The other half were LEOs, low-earth orbit satellites zipping across the sky at far lower elevations. Starlink, SpaceX’s broadband service, positions its orbs about 200–350 miles up. These constellations benefit from lower latency, eliminating the stutter of interactive GEO links.

Aided by stunning increases in computational speeds and the miniaturization of electronics, LEOs exhibit exponential growth: from around 500 in 2010 to over 10,000 by 2025—with about 100,000 additional applications now sitting at the FCC. What was a quiet, uncrowded sky is now buzzing. SpaceX’s Starlink serves nearly 3 million U.S. households with high-speed internet; Amazon, Eutelsat, AST, and others are poised to join the fray. 

Simple spectrum sharing rules were thought by regulators to be sufficient to manage space radio traffic. Well, that was then. For about the last decade, SpaceX negotiated spectrum sharing agreements, under federal supervision, to coexist with other satellite operators. But 6- to 13-year delays (conservatively tabulated) reflect the natural bio-rhythms of government spectrum allocations.

SpaceX found the long wait costing too much, even for spectrum access priced at free. In late 2025 it bailed, paying $19.6 billion for wireless licenses held by EchoStar. These permits, along with regulatory waivers the FCC had previously crafted, award exclusive spectrum rights with flexible use. Sharing deals come on the owner’s schedule, not the government’s. 

SpaceX—after investing tens of billions of dollars in satellites—was now assured of unencumbered channels to link to thousands of satellites and millions of paying customers. Today Starlink supplies service to T-Mobile phone customers who wander out of range. But the satellite backup links are confined to be “ancillary” to the “primary” service, subject to a regime called “Supplemental Coverage from Space.” This forces SpaceX to operate on rights leased from T-Mobile with limited contractual flexibility and needless restrictions. The market might configure something superior, particularly in launching the real prize: “Direct-to-Cell” service. Major Tom to ground control. Outer space to your mobile.

A black hole swallowed a decade of FCC-supervised bandwidth negotiations. But when SpaceX struck its deal for secure ownership rights, its IPO package was complete, shares were soon sold, and creative destruction hit liftoff.

Still, the satellite world remains spectrum-famished, even as idle spectrum is scattered all about. The 65 megahertz (MHz) of nationwide bandwidth purchased by SpaceX is proof of concept. Allow more exclusive, flexible-use rights to flow to their highest valued use. 

The Musk-led innovation has already penetrated the FCC’s reluctance to change its ways. In an initiative announced October 7, the Commission proposed a new license auction—just 25 MHz, but it’s a start—to sell flexible bandwidth rights to be used for terrestrial or space links; to carry earth traffic or D2D connections. 

And just yesterday SpaceX announced another $8 billion deal, giving the satellite operator another 14 MHz of nationwide spectrum, controlled exclusively and with flexible use rights. Now folks are starting to notice. The stocks of all three U.S. mobile carriers (AT&T, T-Mobile, VZ) crashed. This is the market validating Musk’s spectrum-owning strategy and rejecting the bureaucratic sharing approach.

The FCC is now, belatedly, seeking to accommodate this bandwidth-hungry world. Some will no doubt say that the Trump-era FCC is playing to its pals, assisting allies. They have a point: Spectrum use should be offered generously, not just when friends knock at the door. An array of competitors are lining up to share the spoils, dreaming of sinking their own trillions into the most advanced communications networks known to mankind. In heaven as they are on Earth.

The post SpaceX Found a Way Around the FCC appeared first on Reason Magazine.

New Space Capitalism: The Entrepreneurial Path to the Stars, by Rainer Zitelmann, Skyhorse Publishing, 282 pages, $32.99

Much of the public is unaware of the gradual transformation of space activity from a purely governmental arena to a commercial industry—or mistakenly assumes that this change arrived only with SpaceX. Rainer Zitelmann’s New Space Capitalism: The Entrepreneurial Path to the Stars offers a solid, brief summary of that history. While the book highlights other companies and the need for competition, it is somewhat too credulous about claims by SpaceX chief Elon Musk. But that is a minor fault in a useful introduction to the topic.

The book’s most interesting chapter is its final one, which argues that property rights should play a central role in our space future. That subject gets too little attention today. As the U.S. races back to the moon, Washington appears to be repeating the mistakes of the Apollo program: It is giving us a government-led, government-directed effort to fly government astronauts to government facilities, even if the launch vehicles this time are privately owned and operated.

The government’s interest in controlling the mission is understandable. The strategic value of space is visible from Ukraine to Iran to the South China Sea, and Washington has good reason not to want another power to dominate space or to deny American access to its resources. The best way to keep lunar resources from becoming an arena of conflict is to ensure that no power can lock market democracies out. We wouldn’t want to live in a world where Americans have no presence on the moon as an adversary uses lunar resources for military ends.

But we cannot rely on wise central planners, kept continuous across many elections, to achieve American goals in space. Nor can market democracies command the needed resources for such a mission. It is commercial interests that provide the sustained motivation to keep such policies moving forward. And since private actors need a profit motive, that is one reason property rights are essential.

Early outposts in isolated territories—think of the Polynesian expansion across the Pacific, or the Norse settlement of Iceland—may thrive locally without generating much export trade. In strict net-present-value terms, the development is hard to justify. But real estate value bridged the gap. Land on a barren volcanic rock is worth little; land next to a living settlement is worth more, and ownership offers investors a return. That land can be sold on expected future value, used as collateral, or treated as a real asset. A lunar mining outpost of 100 people may not have many products, but title takes no mass to ship back to Earth.

Hernando de Soto’s The Mystery of Capital explains not only why formal title matters, but how the U.S. succeeded in part by breaking arrangements that locked land in inaccessible grants and by turning squatter practice—mining claims, homesteading—into law. The Homestead Act of 1862 regularized what was already happening on the ground. We should approach the settlement of space in the same spirit.

Preventing Space Serfdom

The need for property rights doesn’t stop with the incentive for investment. What about the people on this frontier? 

People are what space development most lacks. They are labor—if only to fix and supervise robots—and they are the only source of local demand. People need things; robots, as yet, buy nothing.

The investments required for human occupancy are not mysterious. They have been discussed since the 1960s: closed-loop life support, local resource extraction and manufacturing, propellantless cislunar transportation, and countermeasures such as spin gravity for health hazards. NASA has never given priority to even the basic research behind those technologies, but such tech will be vital for commercial development. These innovations will come from people working in space, who experience needs firsthand, or from entrepreneurs on Earth competing to sell them solutions.

Yet the government is proceeding as if none of this matters. Will government astronauts on a lunar base be allowed to improve their life-support systems? Of course not. They do not own the equipment and cannot turn a screw without approval from a distant Mission Control landlord. That absentee landlord has nothing to gain from cheaper operations or reduced resupply; those costs are budget justifications for a government agency. Will the staff own their inventions? No: As government employees or contractors, their creations will be the intellectual property of the state.

Washington is apparently prepared to abandon the tools that made the American economy work. Instead, space workers will be serfs: people who will not own their habitats, their equipment, or the product of their labor, who must do what they are told by central planners millions of miles away.

A Market Framework for the Frontier

None of this is necessary. NASA can meet its objectives by buying services from private actors. The government can create structures by which customary rights of use and occupancy—already recognized in the Artemis Accords—can be transferred, mortgaged, and sold. NASA could require contractors offering lunar surface services to give the people who live there an equity stake in the physical plant and hence an interest in improving it. We could establish from the start that people living in space or on the moon are themselves customers who can make buying decisions.

As a matter of policy, the U.S. could loosen its grip on U.S.-flagged operations in space. Applying domestic tax and regulatory structures to space environments is unfeasible. Escape from burdensome law at home has always been a reason to move to a frontier. Giving U.S. operators regulatory flexibility could prevent a scenario where commercial space fleets register under foreign flags of convenience. Space does not need to become U.S. territory for property rights to exist. Customary arrangements need only be formalized.

The sudden interest in putting data centers into orbit shows how powerful regulatory relief can be. One major driver for the interest is escaping terrestrial municipal restrictions and environmental permitting. Bypassing that constraint alone may call forth investment that dwarfs NASA’s budget. The same logic applies to other heavily regulated industries on Earth, such as nuclear power, if they acquire the ability to conduct research off the planet.

Without formal property mechanisms, nation-states will enforce claims at gunpoint and billionaires with vast resources will carve out de facto monopolies. If SpaceX settled Mars, Musk could claim territorial control and make it stick, treaty or no treaty. Our goal should not be to stop private pioneers, but to ensure that dozens of competitors can join so that no single entity dictates terms for the rest. The small entrepreneur and the individual settler need the protection of property rights most.

The post Who Will Own the Moon? appeared first on Reason Magazine.

Did you see the news? Everybody is freaking out about artificial intelligence, which poses a significant likelihood of killing everyone on the planet once it achieves superintelligence…at least according to a bunch of tech folks who might have their own reasons for stoking mass panic.

This media frenzy has led to blowback from a bunch of people, including Elon Musk, who say AI doomers are engaged in a “psyop”—in other words, they are using inauthentic means to manufacture a nonexistent controversy over AI. Proponents of the psyop theory point to a high degree of coordination among the doomers, though they are glossing over some less conspiratorial explanations for why politicians and reporters are boosting their claims.

Ex-Anthropic employee Jacob Coxon triggered this meltdown earlier this week when he announced he was leaving the company due to his own concerns that Anthropic and OpenAI—where he previously worked—were racing toward superintelligence without sufficient concern for how an advanced AI could threaten humanity. His thoughts on X went super viral and were widely disseminated by other tech people, politicians, and commentators.

I resigned from Anthropic today. I spent the last three years doing pretraining research at both OpenAI and Anthropic. Neither company is acting responsibly. They are racing straight to self-improving superintelligence and gambling with our lives. More thoughts below.

— Jacob Coxon (@hilbertspaess) September 9, 2026

You might have expected Anthropic to push back on his claim, but one employee, Evan Hubinger, actually chimed in to echo Coxon’s concerns.

“We really do earnestly believe AI could kill all humans!” he wrote. “I personally think it is >10% within the next decade.”

Jacob is correct here—we really do earnestly believe AI could kill all humans! I personally think it is >10% within the next decade. I believe Anthropic is trying its best, but we do not yet have a plan to solve alignment for superintelligence and are not clearly on track to. https://t.co/QAIHiFP3QZ

— Evan Hubinger (@EvanHub) September 9, 2026

A greater than 10 percent chance of something happening is significant. I think most people would not want the AI companies to proceed with their plans if the risk of global extinction was really that high. Note that Hubinger also explicitly rejects the notion that this is a “marketing stunt,” which is an accusation frequently leveled at AI doomers: namely, that they are trying to drum up interest in the product by making it sound dangerously cool, which will help to increase its appeal to and ultimately its value.

That said, it still could be a marketing stunt! There’s like a 10 percent chance our totally awesome product goes nuts—isn’t that amazing and scary? does in fact sound like a marketing gimmick rather than a sober and scientifically reasoned analysis of the likelihood that superintelligence will produce catastrophe. How was the 10 percent figure arrived at? Did they weigh the odds of the superintelligence saving millions of lives? Or counteracting a foreign adversary’s malicious AI? If the people working on these systems really think they’re dangerous, they have a responsibility to be more specific about the kinds of threats we may face.

Vague predictions of tech-related doom are more than sufficient to make the public extremely afraid, however. Coxon’s post was shared everywhere, and he was quickly summoned onto cable news shows. Op-eds appeared in newspapers and magazines. Regulators signaled their readiness to spring into action. Anthropic itself seemed ready for the news blitz, and issued statements begging the federal government to do something to constrain not just its AI—but also its competitors’.

This sort of behavior is nothing new in the tech world. Meta, for instance, has routinely issued public pronouncements in support of greater regulation of social media companies; in years past, critics have argued that the legislation they support could wind up helping Meta’s products by solidifying Facebook and Instagram as market leaders. More recently, following Meta’s decision to settle various lawsuits for billions of dollars, the company said it was in full of support of government-mandated childproofing—so long as the government reined in YouTube and TikTok, too.

But if Anthropic’s decision to align with AI doomers is cynical and self-serving, it’s still not a “psyop,” or an elaborate conspiracy to trick the American people into supporting more regulation—something critics of the doomer position are also alleging. The truth is that neither politicians nor the mainstream media need much of an excuse to succumb to knee-jerk anti-tech thinking: They default toward wanting to prevent alleged harms.

Who is held accountable when AI goes rogue?

I’m launching an investigation into OpenAI – Americans deserve answers pic.twitter.com/bL9zHuHhH6

— Josh Hawley (@HawleyMO) September 10, 2026

Both the media and the government are biased in favor of action: Someone should do something. And AI is a big something. We’ve all been raised on science fiction stories that predicted killer robots would try to take over the world: Terminator, The Matrix, Dune, Battlestar Galactica, and so on. It’s easy to see why flashy headlines warning that tech insiders not-so-secretly believe a rogue AI is only a few years away from destroying humanity is going to sell a lot of newspapers. Similarly, no politician wants to be seen as being insufficiently attentive to this problem. Even Sen. Ted Cruz (R–Texas), a comparatively pro-tech political figure, prefaced his recent remarks on AI by first stressing his bipartisan do-something approach to tech legislation.

So it’s not really a surprise that the world of political commentary is buying what that ex-Anthropic whistleblower is selling. The last thing this debate needs is for the pro-AI side to succumb to the paranoia that broadly characterizes the doomer side. It’s not a psyop; it’s a mistake. (And no, that sentence was not written by AI. I can use this construction too, ChatGPT!)

This Week on Freed Up

We discuss President Donald Trump’s proposal to give every single American a $5,000 “dividend” check if the Republicans somehow win the midterms.

Worth Watching

So far, I am extremely happy with the footage I’m seeing from the Legend of Zelda: Ocarina of Time remake. How about you?

The post AI Doom Is Misguided, Not a Psyop appeared first on Reason Magazine.

The CIA has been using artificial intelligence to write “open source intelligence” analyses, which are being distributed to local police via the Department of Homeland Security’s “fusion centers,” according to documents obtained by Reason.

CIA officials have been alluding to this new AI tool. Open Source Enterprise (OSE) Director Randy Nixon told Bloomberg in a 2023 interview that the agency was building a new chatbot for intelligence analysts to use, and CIA Deputy Director Michael Ellis said at an April 2026 conference that it had produced an entirely AI-generated report for the first time.

The use of AI has clearly grown quickly since then. OSE was creating daily reports on the security situation in Canada and Mexico during this year’s soccer World Cup. Reason obtained the reports during an unrelated request under the Washington Public Records Act for World Cup–related police bulletins.

The reports don’t contain anything secret or particularly shocking. They are an example of “open source intelligence,” a longstanding practice of combing through public media for potential indicators. But they are a rare glimpse into how the CIA uses new technologies—and how U.S. intelligence agencies are integrated into local policing.

“OSE used a keyword search and a keyword relevancy algorithm to select these articles from major outlets in host countries. OSE used [generative AI] to refine the selection, translate the titles, summarize the selected articles. OSE officers reviewed and validated the GAI outputs at each stage to ensure they accurately reflected source material upon initial setup of this serial report,” the methodology note on each of the reports states.

The note names GPT-4.1—which is also commercially available to the public—as the particular generative AI model used for these reports. In other words, CIA analysts are plugging international news stories into ChatGPT and passing on the results.

The reports were passed on to the Washington State Fusion Center, an office for sharing intelligence between federal, state, and local law enforcement. After the attacks of September 11, 2001, the new Department of Homeland Security created the national fusion center network, which has often been criticized for wasting resources and spreading downright incorrect information.

It’s not clear what, if anything, police in Washington state did with the CIA reports. The fusion center redacted several of the documents it provided to Reason for security reasons.

The U.S. government has been increasingly integrating AI into the national security apparatus. Immigration and Customs Enforcement (ICE) has been using AI tools for facial recognition and real-time tracking of individuals, based on massive amounts of data vacuumed up from other government agencies and private industry. The FBI recently put out a request for contractors to build an AI software for maintaining watch lists, as Reason first reported.

Most consequentially, the military uses AI targeting systems, such as Project Maven, to mark enemy targets faster than a human being could. After the Minab school bombing, in which the U.S. military killed dozens of Iranian schoolchildren, there was early speculation that the school was targeted by a mistaken AI program. However, the attack turned out to be a likely case of human error instead.

As the fusion center documents show, artificial intelligence is making its way into national security work in much more mundane ways. And it’s letting the government do more of what it already wanted to do in the first place—in this case, filling fusion centers with paperwork.

The post The CIA Is Making AI-Generated Reports for Local Police appeared first on Reason.com.

Anthropic was never a legitimate national security risk. 

That’s the takeaway from Thursday’s decision in the U.S. District Court for the Northern District of California. Judge Rita Lin ruled that the Trump administration violated the First Amendment and unlawfully retaliated against Anthropic for engaging in “constitutionally protected expressive activities” when it designated the company as a supply chain risk. Lin also found the government’s actions violate the Fifth Amendment’s Due Process Clause “because they deprive Anthropic of constitutionally protected liberty interests without sufficient pre-deprivation notice or opportunity to be heard.” 

The legal battle between the AI firm and the government started in February, when Defense Secretary Pete Hegseth and Anthropic CEO Dario Amodei traded public statements about the Pentagon’s use of Anthropic’s models. Amodei refused to acquiesce to Hegseth’s demand that Anthropic’s model be “free from usage policy constraints that may limit lawful military applications.” Anthropic insisted its contract with the Pentagon include limits on using its technology for fully autonomous weapons and mass surveillance. Hegseth then threatened to nationalize the company and label it a supply chain risk. 

When Anthropic still didn’t comply, President Donald Trump ordered every federal agency to stop using the company’s technology. Hegseth told Defense Department contractors, suppliers, and partners to cease commercial activity with Anthropic, designating it a supply chain risk.

This directive, Lin ruled, was “arbitrary, capricious, and in excess of statutory jurisdiction, authority, or limitations” under the Administrative Procedure Act. Hegseth, she wrote, acted “without observance of procedure required by law.”

The initial directive from Trump extending Anthropic’s designation as a supply chain risk to other federal agencies—such as the Treasury Department and the Office of Management and Budget—was likewise vacated. Lin ruled the president had imposed these conditions “without jurisdiction delegated to those agencies and without authorization by law.”

Lin permanently enjoined the Defense Department from “implementing, enforcing, or giving effect to” Hegseth’s directive labeling Anthropic a supply chain risk. She ordered the department to rescind its previous “guidance, directives, communications, or instructions” on the supply chain designation and to take “all necessary actions” to prevent the designation from being implemented. 

Ironically, the Pentagon was undone by its own words. In her 59-page opinion, Lin states that “it is undisputed Anthropic was given high levels of security clearance and was consistently praised,” adding there was “no discussion of Anthropic’s untrustworthiness or of a potential supply chain risk anywhere in the record” until its dispute with the Defense Department became public.

It didn’t help the government’s case when the Defense Department couldn’t answer whether “terminating a contract for the undisputed sole purpose of ‘stop[ping] public criticism of the President’ would violate the First Amendment.” But federal courts have already answered that question. “Criticism of government is at the very center of the constitutionally protected area of free discussion,” Lin notes.

It’s not all roses for Anthropic. There’s still a path for the Trump administration to decouple the company’s technology from the federal government. Lin states her order doesn’t prevent the government from “taking any lawful action that would have been available” to it before issuing the designation.

Anthropic’s refusal to let its technology be used for mass surveillance was insufficient grounds to consider the company a national security threat. Government skeptics recognized that immediately, and now, thankfully, the courts have too.

The post Judge Says Trump's Clampdown on Anthropic Violates the First Amendment appeared first on Reason.com.

President Donald Trump loves a data center: Last year, the president signed an executive order that “expedites federal permitting, facilitates construction on federal land and makes qualifying projects eligible for financial incentives,” notes The Washington Post. Trump is a huge fan, but much of the rest of his party reflects voters’ fears that data centers will burden residential areas.

Data centers—the huge warehouses needed to store all the specialized hardware that’s required to run AI models—have grown in salience as a political problem in recent months.

Building out and using AI models at scale requires enormous computational demands; data centers, which consume large amounts of energy, are necessary to power these models. A huge data center could create demands that power grids weren’t built for and can’t effectively handle, leading to service outages for residential areas. It’s also possible that costs of improving grids—to serve AI companies—could be passed on to households via rate increases paid by all customers.

That said, if we want more AI applications, and believe them to potentially be transformative, we’re going to need more computing power. And we’re going to need more data centers. (But you can perhaps understand why some people, when met with the “AI might render work as we know it obsolete” line, don’t feel especially accommodating.)

“If I were the mayor of a town or the governor of a state, and I had a chance to get a big plant in an AI plant or a data center, I would absolutely want it because the jobs are enormous, and the money paid, the taxes paid are just enormous,” Trump said last week.

Contrast this with how Republican politicians are talking about them. Texas Gov. Greg Abbott announced a moratorium on data centers within his state earlier this month, saying his regulators need to determine whether the power grid can handle the new facilities. (Right now, Texas appears to have 335 data centers, second in number only to Virginia.) The Wisconsin gubernatorial race has featured Rep. Tom Tiffany, a Republican, repeatedly arguing that his opponent, Democrat David Crowley, is soft on data centers, and that Tiffany will protect residents of his state from the scourge. In Michigan, Mike Rogers—who is running for Senate as the Republican against Democrat/socialist Abdul El-Sayed—has called for a moratorium on new data center construction (a reversal of how he had been talking about them). In Florida, Rep. Byron Donalds, who won the Republican primary for governor, says he’ll “protect” the state from data centers, also a reversal of how he had formerly approached the issue.

“The Senate GOP campaign arm, in a private memo to top AI companies, warns that toxic views of U.S. data centers are killing the party’s chances of holding a vital seat in Ohio,” reported Axios last week. “In the memo, obtained by Axios, the National Republican Senatorial Committee says Democrats have made data centers a ‘centerpiece’ of their campaign to defeat Sen. Jon Husted (R-Ohio)—and that it’s working. ‘If he loses and data centers get the blame, politicians across the country will take notice—and they will not go near the next one,’ the memo says. ‘This has become a sleeper issue for the entire election cycle.'”

So why are so many Republicans so willing to break with Trump on this issue, when fealty to El Jefe has for so long been table stakes?

“Trump just personally cares about this topic,” an anonymous source familiar with his thinking told The Washington Post. “You just can’t change his mind about it.” But polling data indicates other Republicans are doing a better job of reading the room: Per Pew data back in March, “around one-in-five Americans or more say they aren’t sure of data centers’ impact in each of the five areas we asked about. Others say these facilities have neither a positive nor negative impact in each area or haven’t heard about data centers at all.” Now, voters seem to have heard an awful lot more about them, and made up their minds: An August poll from The Economist/YouGov found that 61 percent of Americans would oppose a data center being built in their community, including 47 percent of Republicans. More than half of Americans disagree with Trump’s perspective on data centers when asked about it.

Scenes from New York: Rep. Alexandria Ocasio-Cortez (D–N.Y.) just renewed her lapsed membership in the Democratic Socialists of America (DSA) after a seven-year hiatus. What could this mean? Is she gearing up for a presidential run? A Senate run? Something more nefarious?

QUICK HITS

• “A Russian barrage of missiles and drones hit a children’s hospital, a school, apartments and warehouses around Kyiv last week, killing at least 16 people. Ukraine, lacking Patriot interceptors, didn’t shoot down any of Russia’s ballistic missiles. Ukraine’s response this week included a drone strike that caused a massive fire at one of Russia’s largest oil refineries, near Nizhny Novgorod, on Wednesday. Meanwhile, in the central Russian region of Tambov, a warehouse belonging to e-commerce giant Wildberries burned,” notes The Wall Street Journal. “The air war between Russia and Ukraine has entered a new phase. With expanded aerial arsenals and inadequate air defenses, both sides are bludgeoning each other in a bare-knuckle fight that is causing a widening swath of destruction and a rising civilian toll.”
• “Hispanics make up about 15% of the U.S. population. The percentage of Hispanic Americans identifying as religiously unaffiliated rose steadily from 18% in 2013 to 23% in 2025,” reports Axios, per Public Religion Research Institute data. “Among Catholic Latinos, the declines were historic. An Axios review of Pew Research Center data found that the share of Latino adults identifying as Catholic plummeted from 67% in 2010 to 43% in 2022.”
• Tim Carney has a great thread calling into question government-produced walkability indices. He has better ideas for how we could measure this:

This, btw, would be revolutionary: Make a walkability index. Then empirically measure if people are walking more in the "more walkable" places. https://t.co/CQ1cDW9eHL

— Tim Carney (@TPCarney) August 26, 2026

• Yayoi Kusama, the pop artist who was obsessed with both polka dots and depicting the phallus (as a means of overcoming earlier traumas), died earlier this week. I loved Kusama, who was helped along by Donald Judd, king of minimalism.
• Imagine using Italian-made AI for your antifa group. They’re not sending them their best!

Breaking: The US State Dept has connected Portland's Rose City Antifa to a foreign tech company that provides infrastructure to violent far-left extremist organizations.

The Oregon-based terror group has been using Italy-based A/I Collective to disseminate target information,…

— Katie Daviscourt (@KatieDaviscourt) August 26, 2026

The post Data Center Savior appeared first on Reason.com.

The AI arms race heats up: “The U.S. is preparing to ‌tell dozens of countries they must pick sides in the artificial intelligence race with China, warning they will be excluded from a U.S.-led AI coalition if they also sign up for Beijing’s competing framework,” reports Reuters. “Washington last year launched the Pax Silica initiative aimed at securing supply chains for AI models, semiconductors and critical minerals, amid a fierce technology rivalry with ​Beijing….By pressing ​countries to choose sides the U.S. hopes to starve China of resources in a race to make the most sophisticated AI, which could be used for military or economic dominance.”

At the same time, the Trump administration appears to be set to financially benefit from the growth and sophistication of Chinese AI models. World Liberty Financial, a Trump family–backed crypto firm, offers tokens that are then used by WorldClaw, which Reuters calls “a Hong Kong–based venture offering artificial-intelligence models developed by Chinese companies.” According to Reuters, “43 of the 90 models available through WorldClaw’s website, or nearly half, were developed by Alibaba, Baidu, Z.ai and other Chinese technology companies the Trump administration says pose risks to national security and intellectual property.”

Chinese AI products are rapidly gaining popularity in the U.S.: Mozilla’s chief technology officer “switched to Chinese AI startup Moonshot’s Kimi K3 for many of his day-to-day activities within days of the new, powerful model’s launch more than a week ago,” reports the Associated Press. “Earlier, he had been using another high-performing Chinese model, Z.ai’s GLM-5.2, for routine tasks such as managing his calendar, documents and email….U.S. companies like cryptocurrency exchange Coinbase said they’re switching to Chinese AI models to help trim costs.” Some of the value-add from Chinese models is that they’re often open-weight or open-source, not closed.

“AI systems exist along a spectrum,” explains John Villasenor at Brookings. “On one end are closed models, such as the various tiers of Anthropic’s Claude and OpenAI’s ChatGPT. Users can submit queries to these models and get responses, but the detailed inner workings of the models and how they were trained are proprietary. At the other end of the spectrum are open-source models, in which information about the training data, the code used to generate weights from the training data, and the resulting weights are publicly disclosed. In between these two endpoints are ‘open-weight’ models in which the weights that ‘determine how the model interprets input data and generates outputs‘ are public, but the methods that were used to generate those weights are not.” Globally, China is establishing dominance as a provider of both open-weight and open-source models, which some companies prefer as they’re a lot cheaper to run, with more customization, control, and data privacy available.

“If American AI is dominated by a small handful of closed model providers, the result will be an overconcentration of power,” warns Villasenor. “That raises many concerns, including the risk of regulatory capture and a set of resulting policies that impede new market entrants.”

Starving China of resources is one path, I guess, but beating China at their own game would be another possibly worth choosing.

Dose of sanity: Finally, someone says what much of the left needs to hear surrounding jobs, pay, and inequality! “If men disproportionately hold dirty or dangerous jobs, paying them more may widen the earnings gap between men and women even when no woman is formally excluded (of course, one could also frame this as widening the longevity gap in women’s favor),” writes the eminently reasonable Kelsey Piper for The Argument. “Differences in opportunity, choice, and achievement can accumulate into durable disparities between groups defined by immutable characteristics like sex. But—while such a disparity might be a warning sign to investigate further for injustice—we have to keep in mind that our objective is to identify and redress injustice, not difference itself. Sometimes unequal outcomes are evidence of sexism and will vanish in a better world. Sometimes they are the mechanism by which society persuades people to haul away our garbage.”

Scenes from Texas: I was back in Texas with my family last week and shocked by the amount of Talarico Fever (in an admittedly blue city).

Democratic state Rep. James Talarico, who is running for Senate against the state’s attorney general, Ken Paxton, believes “it’s time to start flipping tables”—referencing Jesus flipping the tables of merchants outside the temple—and reclaim talk of Jesus and the gospel from the right.

Democratic strategists, per a long Washington Post piece on the matter, seem to believe the Talarico playbook will win over ultra-religious Bible Belt voters. I am less sure. Talarico drew plenty of scorn back in March for an appearance on Joe Rogan’s podcast in which he claimed that the story of the Annunciation—when the angel Gabriel came to Mary and told her she would bear a child, to be called Jesus—lays the foundation for Christians to accept abortion as morally permissible. “To me, that is an affirmation in one of our most central stories that creation has to be done with consent. You cannot force someone to create,” Talarico told Rogan. He further elaborated that Christian opposition to abortion is not rooted in Scripture. “All I’m asking is that for Christians who are pro-choice and who respect the bodily autonomy of women, that we be given the space to make our theological argument, because I think there is a lot of biblical evidence to support that opinion,” he added.

I don’t think this is theologically sound in the slightest, but more to the point: I don’t think most Christian voters—like evangelicals and Catholics, both groups strongly represented all over Texas—will find his logic persuasive. (Talarico also said that “God is nonbinary” during a debate over trans youth inclusion in single-gender sports. My point being: I’m not sure most Texas evangelicals will appreciate his theology. “It is blasphemous,” one Republican told the Post.)

QUICK HITS

• Are people tiring of the pursuit of artificial perfection, in the form of botox, fillers, and veneers? asks Amanda Mull at Bloomberg. Some people “have moved on to some kind of artfully messy style or are asking their cosmetic injectors to let a few wrinkles remain. Ultimately, though, it’s probably too optimistic to regard this potential shift as the coming triumph of more realistic beauty standards or domestic ideals. If big white veneers are too popular, those with means can move on to even more expensive cosmetic dentists, armed with the latest techniques and materials to ensure their perfect new teeth look like the kind that might actually have grown out of their jaw. To avoid a too-puffy face full of filler, they can opt for a new injectable like Sculptra, which subtly stimulates your body’s own collagen production over time. Perfection, after all, isn’t some set point in the distance. It moves when you do.”
• Semi-related: Rep. Nancy Mace (R–S.C.) has new ink, and I kind of love it, though it’s admittedly a little comical. She’s going for the Liz Wolfe look!
• “Democrats have a habit of enacting abortion laws that are extreme by any reasonable definition, and then feigning shock if you care to point this out,” writes Emma Camp for The Wall Street Journal. “This week Massachusetts Gov. Maura Healey signed an expansive new law that effectively eliminates the state’s restrictions on postviability abortion. Under current law, abortion is unlawful after 24 weeks unless the mother’s life or health is at risk or the unborn child has been diagnosed with a fatal birth defect. The new law, which takes effect in November, allows abortion at any time, for any reason, subject only to the ‘professional judgment’ of a physician that it is necessary….Democrats want it both ways—to pass laws with the explicit intent of allowing elective abortion at any point in pregnancy, and to deny that these laws could possibly have that effect.”

The post Open-Weight appeared first on Reason.com.

Artificial intelligence (AI) systems may outperform doctors at diagnosing patients. In a recent series of experiments led by researchers at the Beth Israel Deaconess Medical Center and Harvard Medical School, a preview of the OpenAI large language model (LLM) known as o1 bested human physicians in multiple tests of clinical and diagnostic reasoning.

“We tested the AI model against virtually every benchmark, and it eclipsed both prior models and our physician baselines,” said Harvard Medical School professor Arjun K. Manrai, one of the study’s senior authors. Using case studies, the researchers had o1—an older model since usurped by o3—generate a list of possible diagnoses. It included the right diagnosis 78 percent of the time, while physicians were successful about 30 percent of the time.

In another test, o1 was given five “clinical vignettes” drawn from real cases and asked about next steps. Two physicians judged its responses. The average score for o1 was 89 percent, compared to 34 percent for human doctors given the same test.

The researchers also looked at o1’s abilities to diagnose emergency room cases, where clinical data might be lacking. “Overall, o1 outperformed both [an earlier LLM] and two expert attending physicians, as assessed by two other attending physicians who both were blinded to the source of the differential diagnosis,” notes the study, which was published on April 30 in Science. The LLM especially outperformed human doctors when it came to diagnosis at the initial, triage stage, identifying “the exact or very close diagnosis….in 67.1% of cases,” while the two human doctors did so in only 55.3 percent and 50 percent of cases, respectively.

These results track with other recent research. AI-assisted mammograms could be better at detecting breast cancer, per a Swedish study published by The Lancet in January.

Using abdominal C.T. scans from patients who were eventually diagnosed with pancreatic cancer, an AI model developed by the Mayo Clinic detected this deadly disease an average of 475 days and up to three years earlier than clinicians did. “Attaining such early detection would substantially augment the probability of cure and improved survival,” researchers led by the Mayo Clinic’s Sovanlal Mukherjee noted in the medical journal Gut.

Robo-docs are not likely to take over healthcare anytime soon. And “humans should be the ultimate baseline,” as Peter Brodeur, one of the authors of the Harvard study, put it in a press release. But experiments like these suggest that AI models could competently assist in a variety of diagnostic and medical management contexts, and perhaps even produce better patient outcomes—if we let them.

Nevada now bans AI systems from saying anything that “implicitly indicates” they are “capable of providing professional mental or behavioral health care” and from providing any service “that would constitute the practice of professional mental or behavioral health care.” An Illinois law passed last year says AI cannot provide therapy services and therapists cannot use AI to “make independent therapeutic decisions” or “detect emotions or mental states.” Several states—including Ohio, California, Minnesota, and Kentucky—are considering similar laws. (At the same time, some state bills would mandate that AI chatbots be capable of detecting and responding to mental health issues.)

Measures like these could hamper AI’s potential to diagnose diseases with more accuracy than human doctors alone can.

The post AI Is Already Beating Human Doctors in Medical Tests appeared first on Reason.com.

The Federal Communications Commission (FCC) recently imposed a ban on new foreign-made robotic devices, drawn so broadly as to even include robot vacuums. It’s an especially devastating blow, since one of the most prominent robot vacuum manufacturers has already been the victim of overzealous regulators.

“Today, the Federal Communications Commission updated its Covered List to include two new categories of devices,” the agency announced last week, including “‘advanced robotic devices’ (defined as mobile robots, such as humanoids and quadrupeds).”

The Secure and Trusted Communications Networks Act of 2019 directed the FCC to “publish and maintain a list” of “communications equipment or services from a company that poses a national security risk to U.S. communications networks.”

Items on the list are “prohibited from getting FCC equipment authorization,” effectively preventing their sale or use in the U.S.

The new ban goes much further than just humanoid robots: A National Security Determination cited in the announcement includes any automated software-controlled devices weighing more than 4.4 pounds that are “capable of locomotion, obstacle avoidance, navigation, or movement on the ground.”

The ban won’t apply to any devices already on the market, and it allows the government to grant exemptions “after finding that such device or class of devices do not pose such unacceptable risks.” But it still seems remarkably arbitrary, especially when it’s sold as a means of protecting national security.

“The FCC isn’t actually targeting companies with poor security. It’s targeting basically every robovac company, regardless of their security practices—because basically every robovac is manufactured outside the US,” Sean Hollister writes at The Verge.

In fact, “the government isn’t asking these companies any questions about security—not one—to get the waiver that lets them through,” Hollister adds. “The FCC only wants to know where they’re designed and made and assembled and tested and influenced, and get a specific commitment to start manufacturing them in the US instead. Even then, it appears the FCC will simply preference US companies without any new commitments to security or US manufacturing.”

Earlier this year, the FCC banned all foreign-made wireless routers, and it’s currently considering a proposal to ban foreign-made drones, even those it had previously approved.

But this new ban is worse, especially considering how it will hurt a company that has already endured its share of abuse from government regulators.

Amazon announced in 2022 that it would acquire iRobot, the Massachusetts-based company that makes the Roomba robot vacuum. While it was once a pioneer in the space, iRobot had fallen behind its competitors: The New York Times‘ Wirecutter no longer recommends Roomba products, noting they “have not performed well in our testing in recent years.”

An acquisition by a company with pockets as deep as Amazon’s could have been a much-needed lifeline, allowing iRobot to invest in research and development.

Instead, busybody regulators in both the U.S. and Europe opposed the merger on antitrust grounds.

In a letter to then-Federal Trade Commission Chair Lina Khan, Sen. Elizabeth Warren (D–Mass.) and five other lawmakers said the Federal Trade Commission (FTC) “should use its authority to oppose the Amazon–iRobot transaction”—which, they warned, “could harm consumers and reduce competition and innovation in the home robotics market.”

The European Commission, the European Union’s governing body, also objected, fearing that “because Amazon’s online marketplace is a particularly important channel,” it “may have the ability and the incentive to foreclose iRobot’s rivals.”

Amazon abandoned the deal rather than try to placate regulators on both sides of the Atlantic. As a result, iRobot laid off one-third of its workforce and soon faced bankruptcy. It also reported in a bankruptcy filing that tariffs on goods from Vietnam, where it made most of its products, added $23 million to its costs in 2025.

In December, the company announced it would instead be acquired by China-based company Shenzhen Picea Robotics, its primary contract manufacturer. At the time, iRobot owed Picea nearly $100 million.

Considering, as Reuters reported last week, that “the FCC is expected to exempt many non-Chinese suppliers from the restrictions,” this means iRobot will have borne the brunt of two successive presidential administrations’ regulatory actions.

The post FCC Bans Foreign-Made Robots, Including Roombas appeared first on Reason.com.

Democrats have a union problem. They have tied their political fortunes to unions, public and private, in ways that exacerbate intraparty conflict while stymieing both progressive and moderate goals. That’s not just a problem for the party. Democratic deference to unions creates economic hurdles for much of the country—almost everyone who isn’t in a union, and even some people who are—because today’s unions function as roadblocks to the future.

To understand the party’s problem, consider that the two most powerful words in Democratic politics in 2026 are abundance and affordability. Both represent rival party factions.

The abundance faction is a predominantly young, educated, and urban cohort of policy wonks, journalists, and activists who generally believe that the Democratic Party has overemphasized identity politics while moving too far to the left on economics, particularly when it comes to urban governance and housing policy. They tend to favor technological progress and measurable gains in efficiency. One of the movement’s core insights is that Democrats, who essentially have uniparty control of America’s biggest, most important cities, need to deliver better governance and a better standard of living for their constituents—and that to do this they must embrace deregulation, private business, market mechanisms, and high-quality policy analysis.

That starts with housing, which has been overregulated to the point of strangulation, resulting in shortages that have dramatically increased home prices in and around the country’s most desirable cities. The core issue is one of supply and demand, and Democratic politicians have often chosen to restrict supply and subsidize demand. In the abundance view, what those cities need (and what Democrats need to deliver) is more housing, primarily via market mechanisms that allow developers to build. More supply will ease demand and lower prices.

From this insight—that markets are an efficient and necessary way to deliver more goods at lower prices—an entire theory of policy flowered. What if market mechanisms are a good way to deliver all sorts of goods beyond housing? What if markets depend on developers, entrepreneurs, and businesspeople who are often quite wealthy? What if building new things, and even new labor-saving technologies, was the path to widely shared prosperity? What if there was an entire policy agenda devoted to delivering these goods via these processes? Perhaps you could call it an abundance agenda.

The abundance advocates are not libertarians. They tend to favor some form of social welfare spending. They sometimes say they want government to be good at doing and building things, like high-speed rail. But even this is a welcome acknowledgment that, in general, government is not great at doing things. Much of the movement amounts to the reinvention of Adam Smith from observational data, the primary observation being that Democratic Party governance has been a disaster. Better late than never.

On the opposite side of the party are the affordability mongers. Yes, abundance types will sometimes use the word too, but affordability has been more associated with politicians like Zohran Mamdani, the young member of the Democratic Socialists of America who is also now the mayor of New York City. Mamdani’s social media–savvy mayoral campaign helped boost affordability into political buzzword status.

Affordability is partly a response by the party’s left flank to the abundance movement’s market-friendly critique, and sometimes it comes with a grudging acknowledgement that the construction of new homes might require actually letting developers build some homes. But mostly it’s a rhetorical dodge, a marketing gimmick that allows the socialist-curious and their fellow travelers—the Bernie Sanders/Elizabeth Warren axis—to disavow responsibility for the party’s many policy mistakes while pushing further interventions into the economy.

Abundance is how you get a one-bedroom apartment that costs less than an aircraft carrier within walking distance of a cute matcha shop. Affordability is how you get subsidized daycare and city-run grocery stores.

While these two factions are at odds, they share a key premise, which is that modern life is too expensive for ordinary people, especially in American cities. One reason life is so expensive is an obstacle that neither side is well-equipped to deal with: labor unions. Democrats say they want abundance and affordability for America, but deferring to union demands will mean they can’t deliver either.

Driverless Cars, Deadhead Ubers, and Expensive Groceries

Public sector labor unions are key drivers of bloated spending, low performance, and all-around inefficiency in state and local governments. Private sector labor unions have repeatedly attempted to block labor-saving technologies, create legal requirements that slow new construction, and push tax policies that would undermine innovation and entrepreneurship, specifically in the tech sector.

For the affordability faction, which favors big-government solutions and top-down control, unions are an affirmative good, a source of both high-paying working-class jobs and partisan political muscle. For the abundance faction, labor is a longstanding part of the Democratic coalition, an interest group to be criticized occasionally but ideally kept inside the tent. Some abundance-aligned writers have taken a more aggressive line against union-driven economic sclerosis, but few high-profile Democratic politicians have been willing to cross organized labor in significant ways, especially in deep-blue urban districts and states.

The party’s capture by labor interests is particularly visible in debates about the role of robots and automation in the economy. Take Massachusetts, where in 2023 there were more than 134,000 car crashes, an average of more than 367 a day. Many of these are minor accidents, but some are serious, and some—too many—result in fatalities.

In 2022, more than 430 people died from traffic accidents in Massachusetts. Bay State traffic fatalities have dropped somewhat since, but vehicle accidents still kill hundreds of people in the state every year. Many fatal accidents probably would not have happened were it not for a driver under the influence of drugs or alcohol. Most years, roughly a quarter to a third of the state’s auto fatalities involve an alcohol-impaired driver.

Driving has obvious public benefits for both personal transportation and shipping. It is also a risky activity. The danger comes primarily from poor human judgment: impairment, distraction, or inability to think and act quickly in an emergency situation.

Until recently, there was no way to eliminate human judgment from the act of driving. Only human beings, imperfect as we are, could operate a vehicle.

That is no longer true. Waymo, an autonomous vehicle company owned by Google’s parent company, Alphabet, operates self-driving cars that are increasingly capable of navigating streets on their own. The company began limited operations in the San Francisco area in 2021; it now runs fleets of taxis in at least 10 cities, with further expansions planned. Other companies, such as Tesla, are also building self-driving cars, with varying degrees of success.

The technology isn’t perfect, but it’s rapidly advancing. The clearest sign of its success is in the safety data—Waymos are far, far safer on every major safety metric than human-driven vehicles. The more that people substitute robot cars for human-operated cars, the safer our roads will be.

But after Waymo began mapping Boston’s streets in preparation for a service launch in the city, the local Teamsters union helped form a coalition to stop it. Saying they had concerns about autonomous vehicle safety, the Teamsters supported a city ordinance that would subject self-driving cars to expansive bureaucratic oversight. Waymo says the ordinance would effectively prohibit the company from operating in the city, calling it an “unprecedented ban on a technology.” Given that the group the Teamsters joined was named Labor United Against Waymo, it’s reasonable to assume that a de facto ban was, in fact, the goal.

“Driverless cars and trucks pose a serious threat to public safety, our communities, and the livelihoods of the countless dedicated men and women across the Commonwealth who work as professional drivers,” Tom Mari, President of Teamsters Local 25, said in October 2025. Given Waymo’s extensive and well-documented safety track record, it is hard to take the group’s safety concerns seriously. They were worried about preserving jobs. And to preserve those jobs, they were willing to oppose a new technology that is demonstrably safer than human drivers.

They’re not the only ones. In California’s gubernatorial primary, both former state Attorney General Xavier Becerra, widely viewed as a more moderate (for California) party stalwart, and Tom Steyer, a left-wing billionaire backed by the party’s progressive wing, supported union positions opposing automated trucking.

“AI shouldn’t put California truckers out of work to pad Big Tech’s profits,” Steyer posted on X. “As governor, I’ll reverse the DMV’s autonomous trucking rules and keep human drivers on the road.” Similarly, Becerra posted that “when it comes to automation, jobs and safety come first. Trucks still need drivers.”

In March, a top official for the California Federation of Labor Unions, an affiliate of the AFL-CIO, posted gleefully about her expectation that Becerra would support “our position on Autonomous Vehicles” and that she appreciated “the near unanimity of Dem candidates in this issue.” Here was a union leader publicly celebrating that Democrats were practically united in supporting labor’s crusade against new technology. This, she wrote, was “progress.” Considering the promise and potential of autonomous trucking, one might reasonably argue it was the opposite.

Waymo does not yet operate an autonomous long-haul trucking fleet. But already there are demonstrations and pilot programs, like one in Texas, that show promise in making semitrucks fully self-driving. Large trucks are involved in a disproportionate share of auto fatalities, and autonomous versions of those trucks would almost certainly be safer than the human drivers Steyer wants to keep on the road. Yes, autonomous vehicles can make mistakes. But they cannot fall asleep, drive drowsy, become distracted, or operate a vehicle under the influence of drugs or alcohol.

Autonomous trucks wouldn’t just be safer. They would be cheaper and more efficient, and they would help lower the cost of consumer goods, especially groceries.

Even—especially—the safest human drivers need to sleep. But an autonomous truck could operate for 24 hours a day without needing to stop for food or rest, cutting transit times dramatically. Autonomous trucks are also expected to be more fuel efficient, reducing not only the cost but the environmental impact of shipping. They could also help solve staffing problems. The American Trucking Associations claims the industry is short 60,000–80,000 drivers. There are reasons to be wary about taking that number at face value. But it’s clearly true that fewer drivers (human or bot) mean more expensive transportation costs, which can be passed on to consumers. By 2035, self-driving trucks could generate $9 billion in consumer savings annually, under an optimistic scenario, according to a 2026 report conducted by the consultancy Steer and commissioned by the autonomous trucking company Aurora.

Self-driving trucks would also threaten the employment of many unionized truck drivers; a self-driving truck running a 24-hour shift might replace as many as three union jobs. That’s why the Teamsters Union, which represents truckers, wants mandates that require a human driver even in trucks that are nominally driving themselves. Steyer’s opposition to self-driving trucks amounts to a one-way transfer from ordinary consumers to organized labor in the form of higher prices—and a net loss for everyone in an automobile, in terms of road safety.

This pattern appears everywhere: Unions oppose robots that would otherwise deliver broad-based cost and efficiency gains to the general public.

The United Food and Commercial Workers union, which represents grocery checkers, opposes automated self-checkout systems. Those systems help prevent retail theft, and they lower grocery costs; the union’s demand is effectively to make grocery prices higher.

A fall 2024 strike by dockworkers in the International Longshoremen’s Association (ILA) was nominally about wage disputes but was really about automation. The strike increased costs for retailers and threatened the larger economy, with J.P. Morgan estimating a likely impact of $3.8 billion to $4.5 billion per day. But despite the economic impact, then-President Joe Biden supported the union, reportedly producing research supporting the Longshoremen in negotiations and inviting shipping industry representatives to a meeting so that the White House could make the case for the union’s side.

More than anything else, the union wanted a total ban on automation. “The ILA is steadfastly against any form of automation—full or semi—that replaces jobs or historical work functions,” the union wrote in an unsigned October 2024 statement. They had been offered a 50 percent pay increase, but that wasn’t enough. “We will not accept the loss of work and livelihood for our members due to automation. Our position is clear: the preservation of jobs and historical work functions is non-negotiable.”

Even for a union communication, this was unusually blunt. But the preservation of jobs and historical work functions—which is to say, prohibitions on cost- and labor-saving technology—is a primary goal for many unions. If it had been up to a horse-and-buggy union, Americans would still be traveling by horse and buggy.

Instead of horses and buggies, today’s unions merely insist on expensive Ubers that are burdened by artificial, legally enforced scarcity.

In Seattle, the city’s ride-share union, which represents drivers of Uber and Lyft, pushed for a mandatory minimum wage for ride-share drivers. After the law went into effect, Seattle became the most expensive city in the country to hail an Uber, with the price of some trips increasing by 50 percent or more. As demand for ride-sharing suddenly dropped, unionized drivers ended up spending more time driving without a passenger, which is known as deadheading. In response, the union called for “a pause in onboarding new drivers until a reduction in unnecessary deadheading miles is achieved,” along with “rules to maintain a balanced market where increases to driver supply don’t continue outpacing trip growth.”

The union’s initial minimum wage advocacy hurt its own members. Rather than admit that the wage law backfired, the union chose to back restrictions on new drivers—pulling up the ladder for new workers while making ride-sharing both less affordable and less abundant for everyone.

Closed Shop

Unions often make it hard to operate nearly any kind of business, small or large. Achilles Heel, a well-regarded bar in Brooklyn, closed three days after its employees announced they had formed a union, citing financial problems, according to The New York Times. Similarly, Barboncino, a pizza shop in the borough, closed after its workers voted to unionize, citing “rising costs and diminished sales.”

Unions make it more expensive and more difficult to run small businesses with low margins. They also make it difficult to own and operate large businesses.

In California, the Service Employees International Union (SEIU) proposed a one-time wealth tax on billionaires in the state. Union-friendly groups estimate that it would generate about $100 billion in revenue over five years, though some policy experts say this likely dramatically overstates the effect. The tax “has the potential to completely destroy California’s economy,” warned the University of California, Berkeley, economist Enrico Moretti.

Residents will have the opportunity to vote on the tax later this year. If it passes, it will go into effect retroactively for the year 2025. Some billionaires have already exited the state, though not all explicitly cite the tax as the reason. Critics have warned that because the tax would hit paper wealth, such as ownership of a company that has not yet experienced a liquidity event, it could effectively force some founders to sell their companies, all but taking their businesses away from them.

California construction unions, meanwhile, have long used environmental laws—particularly 1970’s California Environmental Quality Act (CEQA)—to slow or stop construction, a practice known as “greenmailing.” The goal of these campaigns isn’t to protect the environment. It’s to force costly delays and legal fees on developers in hopes of pressuring them into hiring union labor. Environmental concerns get deployed as a legally actionable, potentially expensive pretext for more nakedly self-interested union demands.

This ruse was taken nearly to the point of self-parody in May, when a union representing state attorneys and administrative law judges threatened a CEQA suit in response to a return-to-office demand. Perhaps the union actually cared about the environmental impact of their commutes; more likely, they just didn’t want to go back to working at the office.

Fighting unions in drawn-out court proceedings is costly. But so is acceding to their demands. Unionized projects are more expensive to build; a 2021 RAND study found that project labor agreements, which specify (typically pro-union) hiring rules, add about 15 percent to the cost of affordable housing projects. The added expense is by design: Unions demand higher wages for the same work, which means union demands make affordable housing less affordable.

Blue State Blues

California, New York, and Massachusetts are longtime bastions of Democratic Party governance, especially in their major metro areas. They are also among the most expensive places to live in the United States, with significantly higher housing, transportation, and energy costs than red states in the Sun Belt. These are the states that need abundance, or even just something like real affordability, the most.

Yet prominent Democratic politicians in those states have been borderline obsequious to unions. In New York City, Mamdani is arguably the most pro-union New York mayor in decades, the sort of person who specifically talks about fighting inequality with “union density,” which, as New Labor Forum notes, is a term more typically associated with union organizers. In 2024, Massachusetts Gov. Maura Healey gave a speech at the Massachusetts Building Trades Unions’ annual convention, where she called herself “the proud daughter of union members” and announced the creation of a Labor Advisory Council, essentially giving unions a direct voice in state governance. A press release from her office noted that it’s the state’s first labor council in more than 50 years. In California, Gov. Gavin Newsom signed Assembly Bill 1228, which increased the minimum wage to $20 an hour for fast-food chains with more than 60 locations. The bill was heavily backed by unions, and Newsom held the signing ceremony at the Los Angeles offices of SEIU Local 721.

Some prominent abundance advocates have argued that Democrats need to break with unions on some issues. The writer Matt Yglesias has argued that unions would in some cases actually benefit from abundance, and thus union enthusiasts should support an abundance agenda. The Argument, a publication that often takes abundance-aligned positions, has chided blue-state politicians and activists for rejecting or slow-walking robot taxis; it also recently published an extensive piece by Nicholas Bagley and Robert Gordon on how public sector unions prioritize jobs above all else, rejecting merit and performance reviews and dragging down public services in the process. The core problem, they wrote, “is that public sector unions generally fight to minimize differences among employees, including both standouts at the top and weak links at the bottom.” Their recommendation is in the title: Democrats should fire bad teachers and bad cops.

The same dynamic often plays out with private sector unions. While it’s encouraging to have writers and intellectuals making the case for better policy, in practice, prominent Democratic politicians rarely stand up to union demands.

The party’s most recent president, Biden, declared himself “the most pro-union president” in history. At one point he proposed what amounted to a $400 billion SEIU slush fund as part of an infrastructure bill. The proposal, which did not pass, would have funded home health workers; in order for states to get the money, they would have had to allow the federal government to dictate wages and let home health workers unionize.

Some moderate Democrats, such as Virginia Gov. Abigail Spanberger, have broken with unions on key issues, such as collective bargaining laws. But most rising Democratic stars, especially those hailing from deep blue territories, appear deferential.

Newsom looks to be gearing up for a presidential run. Mamdani is the great young hope for the Democratic Party’s progressives and socialists. Both have pushed policies intended to streamline the construction of new housing. These are concessions to the abundance crowd. At the margins, they might even have a positive impact on the housing market.

Yet it’s hard to imagine a Democratic Party that is willing to push back meaningfully against union demands—demands that almost always value union jobs over technologies that save time, money, and lives. Unions aren’t interested in affordability, and the only abundance they want is in the form of union jobs for a select few. Until Democrats learn this, they won’t achieve either goal.

The post Labor Unions Hate Robots—and It's Stopping Democrats From Delivering on Affordability appeared first on Reason.com.