Why the Tech Right Lost Big This Week
Kevin RooseCasey NewtonErin GriffithEzra KleinTheodore Schleifer
- AI’s political backlash now threatens the industry’s permission to build. Ezra Klein cites exit polling in which 56% called AI and data centers a crucial or major voting factor, 57% expected AI to hurt society versus 24% expecting benefits, and 76% opposed a local data center. The Republican position—“let them rip”—left an opening for opponents who understood the technology and rejected its trade-offs.
- Texas’s election result puts a substantial share of next year’s planned AI capacity at risk. Erin Griffith says 10–15 gigawatts of roughly 45 gigawatts expected online next year could be endangered because investors assumed Texas’s construction moratorium would disappear after the election. Continued restrictions could tighten compute supply, raise prices, and weaken growth across the “AI trade.”
- Slower construction could also protect investors from a capital-spending bubble. Klein argues that companies may be overbuilding because nobody wants to fall behind, while the Nasdaq’s Wednesday rebound contradicted investors’ initial catastrophizing. Externally paced investment could push labs toward economically useful deployment: “Not at all obvious to me that a slowdown in data center construction from where we are right now is bad for that.”
- Political spending produced sharply different returns, with Musk becoming an electoral liability. Theodore “Teddy” Schleifer estimates Musk spent $470–$500 million; roughly half, perhaps more, centered on Texas and Ken Paxton’s losing campaign. Klein calls Musk an “attentional super object” whose money amplified the accusation that he was buying the election, although Schleifer says Paxton probably would have lost without it.
- California’s wealth-tax result rewards targeted spending without removing national tax risk. Sergey Brin contributed over $100 million to an effort whose wealthy backers raised about $333 million against the measure. Klein nevertheless predicts a wealth-tax proposal will become “table stakes” for Democrats in 2028, arguing federal administration and jurisdiction make a national version stronger than a state measure.
- AI lobbying is encountering stronger resistance than the crypto playbook anticipated. The Brockmans delivered $25 million of a pledged $50 million to Leading the Future, then withdrew the second installment as Greg Brockman called the contribution a “distraction”; Meta also retracted roughly $20–$25 million in state political giving. Crypto’s deterrent suffered too: Sherrod Brown won despite $45 million in Fairshake attack ads.
- Schools and energy permitting are the next concrete tests of political consent. Griffith describes Google conditioning school Search access on acceptance of AI products—“You don’t want our AI? Well, fine, you don’t get our Search either”—before promising to work through the issue. Klein’s nearer legislative watch is a bipartisan permitting bill that could accelerate energy construction while making data centers bear more of their costs.
- The next funding cycle could pit AI acceleration money against newly liquid AI-safety donors. Schleifer expects Musk to remain involved—“I would be shocked if Musk did not support Vance”—while an anticipated Anthropic IPO could enable people in its community to contribute $1–$3 million apiece. His central limitation: increasingly opaque nonprofit funding means public donor rankings reveal only part of the contest.
1. AI became an electoral liability beyond Trump’s other problems
- Ezra Klein separates the tech right’s responsibility from Trump’s tariffs and war in Iran, but argues it helped produce his unpopular AI position. Exit polling he cites found 56% considered AI and data centers a crucial or major voting factor; 57% thought AI would hurt society, versus 24% expecting benefits; 76% opposed a data center in their community.
- The Republican prescription—“let them rip, build as many of them as you can”—created identifiable local targets, including Jon Husted in Ohio. Max Read’s chicken-and-egg question remains useful: was Trump dragging down AI, or was AI dragging down its political sponsors? Klein sees both an unattractive proposition and an untrustworthy rollout.
- Erin Griffith identifies the industry’s mistaken inference: because people use Facebook, Amazon, iPhones, and a ChatGPT she says has a billion users worldwide, investors assume opposition cannot matter. The tech right’s posture became, “We’re so powerful and rich that we don’t have to convince you.” The election challenged that equation between usage and consent.
2. The inevitability argument fails to answer whether people want AI
- Klein’s alternative begins with regulation as a “trustworthy” and at least “semi-democratic container” for managing trade-offs. His Abundance question—“What do we want more of, and how do we get it?”—exposes the missing premise: Silicon Valley knows how to fund more AI but has not convinced people to want it. That is more than a messaging problem.
- Jon Ossoff’s victory speech linked Silicon Valley’s bunker-building titans to their own warnings about human extinction. Klein says lab employees are taking concerns directly to legislators such as Ossoff. Politicians are therefore absorbing the industry’s internal doubts alongside its public promises, rather than simply responding to an outside caricature.
- El-Sayed’s discussion of distillation impressed the panel because it challenged the China-race argument technically. Klein’s skier-and-boat analogy: a boat keeps accelerating because the skier it is towing remains close behind. “Well, it’s because you’re pulling the skier.” Klein sees younger politicians acquiring enough expertise to dispute the industry’s premises.
- Klein reports insiders saying, “If I could, I would push the button. I would stop it for now,” while invoking China and competitors as reasons they cannot. Griffith distinguishes investors: they are less worried about existential risk and more worried regulation will constrain data centers and compute. Their concern itself contradicts technological inevitability.
3. The data-center revolt is concentrated where construction is concentrated
- Griffith preserves a salience caveat: data centers had ranked near the bottom five of roughly 30 issues in polling discussed the previous week. Yet a direct yes-or-no question produced emphatic opposition. That does not establish how many voters they brought to the polls; it does establish a continuing political obstacle.
- Klein calls the issue “spiky”: Texas, Ohio, Georgia, and Arizona had substantial construction and strong Democratic performances. Texas, which he describes as building the most data centers, was particularly exposed in its governor’s race. Ohio Democrats targeted Ramaswamy and Husted. The wedge could attract Republicans; it need not explain a race somewhere like Alaska.
- Griffith says investors were especially alarmed by Texas because 10–15 gigawatts of roughly 45 gigawatts expected online next year could now be endangered. They had expected the pre-election moratorium to lift afterward. Constrained capacity would mean more expensive compute and slower growth for frontier labs, chip suppliers, neoclouds, and cloud resellers.
- Read does not assume construction stops entirely: projects could take longer or move to states such as Louisiana or Oklahoma. Klein’s explanation for local resistance remains blunt: facilities are “big and ugly,” “don’t create jobs,” and face the ordinary difficulty of winning acceptance for industrial construction.
4. Restraining construction might improve the economics of AI
- Griffith heard predictions of a market collapse, economic damage, and China winning. Klein notes Nasdaq futures initially fell, but the market bounced when trading opened Wednesday. His contrarian explanation is a suspected “collective rationality problem”: each company spends to avoid being left behind, producing aggregate overinvestment. He floats Oracle’s demise as one possible consequence.
- Externally paced capital spending could prevent fragile entrants from launching or surviving, while pushing Anthropic and OpenAI toward diffusion and economic usefulness. Klein’s condition is alignment between construction, product value, and adoption. A capital-spending bubble is bad for markets; so is creating dangerous AI before useful AI. Slower building is therefore not obviously harmful to investors.
5. Musk’s visibility and the tech right’s candidates became weaknesses
- Schleifer estimates Musk spent $470–$500 million, with probably half or somewhat more concentrated on Texas and Paxton. Klein calls that spending “hugely counterproductive”: the money sought attention, but Musk’s status as an “attentional super object” made the closing message that a billionaire was buying voters’ election.
- Schleifer is more cautious about causality: Paxton probably would have lost even if Musk spent nothing. Read adds that Musk’s prominence in Texas and national right-wing politics might have made him a campaign character regardless.
- Ramaswamy’s final Ohio ad featured Amy Acton’s mother calling her daughter a liar after Acton described childhood abuse by her stepfather. Klein calls the tactic grotesque and dishonorable. His broader diagnosis is that Ramaswamy and JD Vance “speak Trump with an accent”: imitation produces cruelty without Trump’s political feel, like a “GPT-3-level LLM.”
- Klein argues the tech right worked better as a wealthy faction behind a charismatic coalition leader than as its public face. Its preferred candidates understand its intellectual agenda, but that does not make them appealing politicians. “It doesn’t understand and often doesn’t seem to like human beings”—including the voters it needs.
6. California rejected a particular wealth tax, not necessarily the category
- Klein was surprised the California wealth-tax proposition failed amid pervasive anti-billionaire sentiment. Griffith thinks voters distrusted its design: opponents called it poorly written, difficult to sustain, and a one-time measure. Both distinguish hostility toward billionaires from confidence that a particular ballot proposition would improve California’s finances.
- Schleifer supplies the spending mechanism. Brin put over $100 million into Building a Better California; wealthy opponents collectively raised about $333 million, while the union-backed proponents raised almost nothing. At recording, Schleifer said the measure seemed headed for defeat. “They were able to demolish this thing financially”—a clear counterexample to claims that money never works.
- Gavin Newsom’s opposition mattered to Klein, as did fears that billionaires would leave the state. He predicts wealth-tax proposals become “table stakes” for Democrats in 2028, including Newsom. A national measure would have the IRS and federal jurisdiction behind it, with relocation abroad a less credible threat than moving between states.
7. A political reset requires a more desirable technological future
- Griffith expects companies may hire lobbyists and cultivate Democrats, but doubts right-wing cultural loyalties will vanish quickly. Reid Hoffman is taking a victory lap while calling for money out of politics despite his own large donations. She does not expect the much-discussed “big forgetting” to happen rapidly.
- Klein sees an opening for “a normal tech CEO,” exemplified by Satya Nadella: someone visibly running a company rather than becoming a political influencer. He does not see a left-wing equivalent of the All-In group emerging to replace the right-wing network.
- Read contrasts the moment with 2008, when Democrats and Silicon Valley shared an optimistic story about the internet, social connection, green energy, and recovery. Klein asks what Democrats positively want from technology now. Drug development could provide one answer, but the party needs more than opposition to billionaires, job losses, and utility costs.
- Klein rejects a successful-AI future where humans become “well-kept pets” of superintelligent systems as intrinsically uninspiring. With children aged seven and five, he finds their future preparation for work unclear. The difficulty is substantive: people may reject even the future builders describe as going well.
8. Schools and energy permits will test whether the industry can negotiate
- Griffith highlights reporting that Google would turn off Search for schools unwilling to use its AI products. Chromebook penetration makes that consequential. Google subsequently said it would work through the issue, but she sees the original posture as a familiar refusal to accommodate customers: “I’m taking my ball, and I’m going home because you don’t like me.”
- Read expects keeping AI away from children and education to become a major issue for millennial parents over the next two years. A company appearing unresponsive gives campaign managers and legislators an obvious point of intervention.
- Klein is watching a bipartisan permitting bill involving Democrats Sheldon Whitehouse and Martin Heinrich and Republicans Shelley Moore Capito and Mike Lee. He acknowledges parts he does not understand. Broadly, it would ease construction of green and non-green energy while trying to internalize more data-center costs. Amendments could become a vehicle for the backlash.
- At his latest check, Senate movement was expected within a week or two, with negotiations extending over several months. The lame duck may be decisive: Klein says the theory is that failure then makes passage unlikely afterward. A Democratic Congress could make agreement harder despite Democratic sponsorship, particularly with more natural gas being built to power data centers.
9. Prediction markets became ordinary without proving predictive superiority
- Read’s surprise is how little drama Kalshi and Polymarket generated despite ubiquitous advertising and television visibility. They seemed broadly to get the election right, although he puts “right” in quotation marks because interpreting probabilities is an epistemological question. His main observation is normalization within the election-news apparatus.
- Klein pushes back with Polymarket’s Mississippi Senate market: actual information arrived, then the lines abruptly reversed. That demonstrated fast reaction, not necessarily advance insight. Griffith adds that markets often track polls. Read accepts the exception while retaining his narrower point about their increasingly routine presence.
10. Donor returns undermine the myth of automatic political conversion
- The top 20 donors supplied about $2 billion, more than $800 million connected to technology. Marc Andreessen and Ben Horowitz themselves spent $100 million. Schleifer questions his beat’s assumption that money reliably buys wins: Musk’s roughly $300 million in disclosed 2024 spending was perceived as successful, but that narrative did not establish how much it caused.
- Hoffman’s largest publicly identified investment was $17 million supporting James Talarico, including a competitive Democratic primary. Talarico’s unexpectedly large general-election victory vindicated Hoffman’s judgment about the candidate, while raising a counterfactual Schleifer keeps open: perhaps Hoffman did not need to spend anything in the general election.
- Fairshake’s 2024 success narrative followed fundraising of roughly $100–$150 million from Coinbase, Ripple, Andreessen Horowitz, and others. Its biggest November 2026 test was Brown: $45 million in September–October attack ads failed to defeat him. Brown had lost in 2024, partly, Schleifer thinks, because of roughly $40 million in crypto spending.
- The timing illustrates the threat model. Democrats blocked the Clarity Act in September; the next day came the punishment campaign against Brown. Super PACs can bank money as a “permanent looming threat.” Brown’s victory punctures that deterrent, although Schleifer stresses crypto won other races and should not be treated as universally defeated.
11. AI’s stronger public salience disrupted the copied crypto playbook
- Leading the Future reproduced Fairshake’s model with AI donors, notably Andreessen Horowitz and Greg Brockman. Schleifer identifies the crucial difference: people hold stronger opinions about AI. Crypto’s low salience can make accepting its support strategically attractive; AI money can bring a more substantial political cost.
- Greg and Anna Brockman pledged $50 million in 2025, matched by a $50 million commitment from Andreessen Horowitz. The firm delivered its commitment. The Brockmans gave the first $25 million, then backed out of the second installment a year later. Internally, Greg said his support had become a “distraction.”
- Meta also retracted approximately $20–$25 million from state political giving as the politics worsened. Read sees employee pushback returning after its suppression in recent years. The implication for executives is attachment risk: a public donation can become a story inside the company as well as outside it.
- Schleifer adds a major qualification: retreat from disclosed giving need not mean retreat from politics. He says less is disclosed than ever, especially through Democratic nonprofits. Newly formed intermediaries can suddenly fund $30 million super PACs. Fear of Trump’s threats and investigations helps explain secrecy, but leaves outsiders unable to identify much of the underlying money.
12. Musk’s next campaign may face a growing AI-safety donor base
- Schleifer traces Musk’s attempted lower profile to Wisconsin’s spring 2025 judicial election, when his spending and personal involvement made him central to Brad Schimel’s defeat during DOGE. Musk subsequently floated a third party without implementing or recanting it. This cycle, he said little publicly and deleted a “Vote red today” post after election night.
- Schleifer reads Musk’s warning about a possible Democratic trifecta in 2029 as a signal he will participate in the 2028 Republican primary. Musk’s relationship with Vance has remained good despite difficulties with Trump. “I would be shocked if Musk did not support Vance.” Other industry donors’ choices will depend more on candidates’ positions.
- AI-safety money could become “a force in the Democratic primary,” Schleifer predicts: donors around Anthropic, effective altruism, and parts of the Democratic world favor slower development or take existential risk seriously. Dario Amodei could appear on a future top-donor list. Eric Schmidt and Dustin Moskovitz already combine Anthropic investments with substantial Democratic giving.
- Read anticipates an Anthropic IPO creating more wealthy donors. Schleifer says influence need not require billionaire status: people in the Anthropic community could give $1 million, $2 million, or $3 million. Democratic fundraisers are watching for newly liquid people willing to host candidates such as Josh Shapiro. The political consequences remain prospective, and undisclosed giving will complicate any accounting.
Full transcript
What just happened to Elon Musk? Two years ago, the SpaceX founder and world’s richest man was riding high. He and the rest of the tech right of Silicon Valley had helped propel Trump into office and had grand plans to reshape the federal government. Remember DOGE? Those plans are now on hold. Musk, once again, spent hundreds of millions of dollars this election cycle but got almost nothing in return. Instead, his allies had among the worst losses of their lives, and he became one of the main villains of some of the cycle’s most important races. On today’s episode, we’re going to dive into the midterms and try to figure out what happened. Was this election a repudiation of Musk and his cadre of Silicon Valley billionaires? A referendum on AI build-out and the data centers cropping up around the country? Or an entire rejection of the accelerating AI future that the tech industry has been aggressively selling? How did Elon Musk lose so badly, and what will he do next? I’m Max Read. This is Hard Fork. Welcome to Hard Fork. I’m joined by Aaron Griffith and Ezra Klein. Aaron is a tech writer for The New York Times, and Ezra is, I believe, a podcaster who’s been sent to us from the 28th floor to talk about politics. It’s been a little while since the Democrats retook Congress, and the thing I really want to talk about—the thing I’ve been thinking about—is what we’ve called the tech right.
I’m talking about a highly visible group of Trump-supporting executives, founders, and investors from Silicon Valley: people like David Sacks, Marc Andreessen, Palmer Luckey, and, of course, Elon Musk. Two years ago, this felt like the most powerful faction within the Trump coalition. They were able to dominate news and attention, and most of all, campaign contributions. They arguably helped elevate JD Vance to the vice presidency. After the election, they were kind of put in charge of federal government budgeting and procurement, and now it feels like it’s all fallen apart.
Both of you are close observers of the tech industry, and I wanted to see how you would characterize what happened this week to this growing conservative movement of leaders. Ezra, what do you think happened to the tech right?
Well, it attached itself to Donald Trump. This is a place where, I think, at least in part, their success became a failure. The tech right is not the reason Trump layered the world in tariffs. It is not the reason he went to war in Iran, but it is at least part of the reason he has a hideously unpopular position on AI.
I was surprised in the exit polling by how important people said AI was to their vote. 56% said artificial intelligence and data centers were a crucial or major factor in their vote. 57% said AI would do more to hurt society; 24% said it would help society. 76% opposed a data center in their community.
You’re dealing with an issue that people were taking into account in a direct and vibesy way. Trump, because in this period he has really bought into the David Sacks, Jensen Huang, Marc Andreessen, and others’ view of the world, has taken a position on data centers that the Republican Party broadly shares: let them rip, build as many as you can; whoever wins AI wins. That put them sharply on the wrong side of public opinion on an issue that’s unlike diffuse issues like costs or the war in Iran.
Mm-hmm.
You could sometimes tie a specific politician like Husted in Ohio to data center backlash, and that gave Republicans who didn’t like them maybe a reason to cross party lines on them.
Yeah. Is this a chicken-and-egg question? Is AI getting dragged down by Donald Trump and the tech right, or does AI just naturally have bad politics, with its association with these figures dragging them down?
I think it’s a mistake to think that the only reason we regulate things is to slow them down, stop them, or because they’re bad. You regulate things to put them within a trustworthy, at least semi-democratic container, so that people can make peace with and have confidence in the trade-offs.
There’s a world where AI is rolling out in a way that makes people think the folks in charge have thought through the risks and the questions of jobs, are finding out what they need to find out, and maybe have more trust in it. In that way, I don’t think it’s being dragged down by Donald Trump’s popularity. I think the problem with AI’s popularity is that the basic sell is: it might kill everyone, and it probably will take your job. But if it doesn’t do those things, you’re going to have cheaper energy. That’s not a good description of a technology you want.
The problem it’s having on a secondary level is that it’s rolling out at a time when the people rolling it out just don’t seem trustworthy. They have not been subtle, agile, or thoughtful about how to manage its risks.
But that’s been the whole point—
Yeah.
—of this rightward shift, right? All the tech guys are so excited to be able not to have to actually care, to soft-pedal things and sell them to people. Hasn’t that just been the gleefulness of this Trump 2.0 tech-right thing, where they’re like, “Ha ha, we’re so powerful and rich that we don’t have to convince you. We’ll just do it, and you’ll like it—or you won’t, but we don’t care”?
And, as you can see, it’s working out great.
Well, I think we’re seeing the direct backlash to that, which is kind of what you’re saying, right?
Aaron, you were sort of addressing this in the last thing you were saying, but I’m interested in why this group of tech investors and executives went for Trump so hard in the first place.
There are a lot of layers to this. Obviously, a lot of them just don’t want regulation. You can go back to Marc Andreessen’s whole thing when he went on a media tour after coming out for Trump. He was telling this story of the deal, the broken deal, and how the elites are now under attack. And we had this meeting that may or may not have happened, where all the Democrats were saying they were going to kill AI; subsequent reporting has called that into question.
I do think it’s also a little bit of a cultural thing, where it just seemed more fun for them to be edgelord meme posters and not have to care about cancel culture and wokeness and all of that. On a high level, a lot of the billionaires and leaders of the industry went to the right, and everyone said, “Oh, well, the rest of the tech industry is still very liberal.” But over time, I think that very online kind of thing attracted way more people to that side. I’m going to be very interested to see how that plays out if people start backing away from that vibe.
One thing I also want to say, to add on to what I was saying earlier, is that this idea of ramming through the tech stuff regardless of whether people like it or what the sentiment is—and they’ll like it on the other side—is very prominent among a lot of the VCs I talk to. They say, “Well, people don’t like big tech,” or, “People don’t trust big tech, but look at their revealed preferences. Look at how much they use Facebook and Amazon. They love their iPhones. They’re addicted.” They’re looking only at the data and saying that because people use AI a lot, because ChatGPT has a billion users around the world, people are going to like this stuff. They might not have trust in it or whatever, but that doesn’t matter because people are using it. I think this election is showing us that that’s the wrong way of looking at AI because people are obviously very, very mad about it, to the point where they don’t want it.
This felt like a decidedly anti-tech election, but I think we can also separate some of the sentiments that contributed to the votes and the blue tsunami. I heard somewhere that in Japan, because you don’t want to call it a blue wave or a tsunami, they’re calling it “the national scolding,” which I think is so much better. So I’ll call it the national scolding.
The factors that went into the national scolding—let me separate the tech stuff into 3 different categories. One is the broad opposition to AI that you’re talking about, Erin. One is the specific anti-data-center sentiment that really captured a lot of attention this summer, which a lot of politicians ran quite hard on.
And then finally, a kind of general anti-oligarchy, anti-billionaire sentiment that focused in particular on Elon Musk, but I think in general on a lot of tech billionaires. I want to take these one at a time and start with the anti-AI sentiment. I’m going to play a little bit of Jon Ossoff’s victory speech from last night.
“And at our grave peril, humanity is fast forgetting the terrible lessons of two world wars and the Holocaust. Meanwhile, the titans of Silicon Valley dig bunkers for themselves and warn that the artificial intelligence they’re creating risks human extinction.”
That’s pretty strong. I’m not talking about those new biceps he’s got, folks. That is a real—
Oh, no.
Well, he is getting a lot stronger than he used to be. I thought that would play a lot harder, but maybe we can cut it out later. It’s a pretty specific, pretty aggressive speech. This is like a speech written by somebody who’s on Twitter, at the very least, who has a strong understanding of AI culture and Silicon Valley culture.
Ezra, what did you make of this part of Ossoff’s speech?
At the beginning of the book Abundance, which Dirk Thompson and I wrote, we have this thing about supply-side politics where we say the core question should be: What do we want more of, and how do we get it? The basic problem they have on AI is that they haven’t convinced people to want more of it. They understand how you get it, right? Push all this money to CapEx. But why do you want it?
First, you have all this energy going into that, and then the data centers are just big and ugly, and they don’t create jobs. People don’t like the construction of industrial facilities near them, even in normal times. It’s hard to build an apartment building. It’s very hard to build these gigantic systems.
So they seem to have come up with this view that it’s not going to matter because people enjoy using their chatbot. But it is going to matter. And the reason Ossoff—but not just Ossoff—is that El-Sayed had a great answer on an AI pause when he was talking to our former colleague Rush Douthat on 60 Minutes.
Yeah, he actually knows what distillation is.
Yeah, it was very impressive. A lot of the frontier research is just spoofing our models.
China’s research?
China’s research. It’d be like you’re pulling a skier on a boat, and you speed up the boat, and you’re like, “The skier’s still so close. We gotta speed up. The skier’s still so close.” Well, it’s because you’re pulling the skier. He knows what distillation is.
This is a thing I think younger politicians are really developing expertise on. I can tell you they all understand and believe it will be a central issue in politics going forward. And, like everybody else, they’re not buying the “just don’t worry about it” line.
Yeah.
Yeah.
Okay, one question on that, though. What do you make of their answer to how this happened—that AI has a messaging problem? To me, that misses the point a little bit. They’re reinforcing that, again, they believe this is what’s best for everyone, and if everyone just lets them do it, then we’ll see. But if they stick to that—
I don’t think they do believe it’s best for everyone. Some of the tech-right guys do, but I think one of the most interesting things when I talk to them—and I’d be curious how much this backs up your experience—is that a lot of people in AI say publicly and privately, “Oh, if I could, I would push the button. I would stop it for now.”
Oh, just stop it. Yeah.
“And I just don’t feel we can. There is this inevitability thesis. There’s China, there are the other companies, right? There’s this kind of spirit of technology. In some ways, it’s like when Elon Musk talks about it in a very teleological fashion: ‘I tried to stop it. It turns out I can’t, so it’s more fun to be part of it.’”
It is true that some of the tech-right people we’re talking about feel that way, and some of them also don’t care or don’t see it in existential terms. But what a lot of people are hearing from them—and correctly absorbing the lesson, as Jon Ossoff is—is that a lot of lab employees are going to Congress and talking to people like Ossoff and explaining their concerns. The message I think a lot of us are getting is that what’s pushing this forward is a belief that it cannot be stopped.
But even the people pushing it forward believe that maybe it could be stopped, at least for a while, and that figuring that out would be better. So if you don’t buy into inevitability—and most politicians don’t—then all of a sudden it’s like, “Well, why not try to pace it, slow it down, regulate it, you know?”
Yeah.
Change public opinion on it.
Well, I think there are two groups, though. There are the employees who are very worried, like you’re talking about, especially at the labs. And then there are the investors, who I spend a little bit more time with.
Mm-hmm.
They’re not so worried about the existential-risk piece of it. They’re more worried about regulation. They don’t think it’s inevitable because they worry about regulation, and they worry about not having the data centers and the compute. They’re kind of on the flip side of that a little bit.
Yeah. Let’s talk about data centers for a second. Tuesday night was a near-total rout for data centers on basically every level. Nearly every local referendum went against them, and a bunch of the states that have been building the most intensely saw Democrats take control, often on anti-data-center platforms.
But I’m still trying to figure out how big a deal it is, because it’s not necessarily as salient as tariffs, affordability, Iran, or immigration. But it also clearly animated voters all year. How big a driver do you think the data-center issue was of these results?
It was interesting because last week we talked about how it ranked in maybe the bottom five out of 30-some issues when voters were asked what they care most about. But it’s clear, as you’re describing, that when asked a binary yes-or-no question, the answer is absolutely not. Whether or not that drove people to the polls, Azara, you probably have a lot more info on that, but it’s obvious that it’s going to continue to be an issue.
Yeah, I think it was spiky. In states where there’s a ton of data-center build-out happening, like Texas, Ohio, Georgia, and Arizona, it was a really big issue.
Yeah.
I think you have to look at the Texas results. Texas is the place in the country that’s been building the most data centers. They were the most vulnerable. It was a huge part of the campaign, particularly in the governor’s race, and it worked.
Yeah.
It was just a huge wedge. And it’s a wedge that, importantly, doesn’t split or polarize normally. You can win over Republicans that way. You look at that list—Texas, Ohio, Georgia, and Arizona—and those are all states where Democrats performed really, really well.
In Ohio in particular, they really went at Ramaswamy and Husted, the Republican gubernatorial and Senate candidates, on the data-center issue. I don’t think data centers were a huge part of the Alaska race. But in the places where they’re building a lot of them, it really mattered.
Yeah. That’s interesting, and it does feel like there’s going to be a big shift in how we think about the build-out. It doesn’t necessarily mean a full cessation of data-center build-out, but it’s going to take more time, and it’s going to move to other states, maybe Louisiana, maybe Oklahoma.
Erin, I’m wondering how you think this is going to play out among the VCs and investors you talk to—how they’re changing their plans or thinking about it differently now.
I think investors are generally a little bit terrified by the results, particularly in Texas. I was watching the All-In feed last night, and the Texas governor’s race was the one they seemed most upset about.
As you do every Tuesday night.
There was this kind of cope moment: “Oh my God, they’re only reporting the liberal or blue districts so far. We still have a chance.” They know that, of the roughly 45 gigawatts of energy that’s supposed to come online next year, 10 to 15 gigawatts of that could now be in danger of not coming online because it’s in Texas.
Yeah.
They all expected that the moratorium on building data centers in Texas, which was in place through the elections, would just be lifted, and they’d all go back to building and everything would be great.
This is going to slow down a lot of AI development. It’s going to make compute capacity much more constrained and, ergo, much more expensive. It’s going to hurt the growth not just of the frontier labs that need the compute, but of the whole ecosystem of companies playing into the so-called “AI trade”: chip companies, neoclouds, and cloud resellers. That could potentially hurt the bull market we’ve been in for a long time.
Getting a little bit dramatic about that.
I’ve heard there was a little—
No, no investors getting dramatic, I don’t think so. That doesn’t sound right at all.
There was a little bit of catastrophizing, like, “The entire market’s going to tank. The economy’s in the shitter. China’s going to win.” All of that. I even saw a tweet from Beth Jesos, the internet personality, saying this is the end of Western civilization and it’s time to move to space, which is where the data centers were moving anyway.
You’ve got to get that space real estate. What I thought was interesting about this is I was hearing some of this from investors, too, and the Nasdaq futures were down. Then the market opened Wednesday morning and bounced up. So clearly, the market doesn’t think this is so bad.
A good contrarian take I’ve seen floating around is that this might be good for investors, because I think there’s a widespread suspicion that there’s a collective rationality problem happening right now. Everybody is investing too much in data centers because they don’t want to be the company left behind, but that might lead to the end of Oracle as an existing American corporation.
Yeah.
And a situation in which CapEx investment is externally paced, such that some of the probably more fly-by-night players who are beginning to emerge, and who could collapse easily, either don’t get off the ground or can’t keep going. Similarly, players like Anthropic and OpenAI might focus more on diffusion and making what they’re deploying economically useful, as opposed to racing each other toward the notional goal of superintelligence. It’s not obvious that that’s bad for markets. A CapEx bubble is really bad for markets. Creating dangerous AI before economically useful AI is also bad for markets.
In a slightly more sober sense, what markets need is an alignment between how fast the build-out is happening and how economically useful the product is, and thus how quickly it’s adopted. It’s not at all obvious to me that a slowdown in data center construction from where we are right now is bad for that.
Yeah. Let’s talk about tech billionaires. This is sort of the third category of anti-tech sentiment I’ve seen pop up throughout this election season, and it came up quite often in the victory speeches on Tuesday. We had Talarico effectively calling out Elon Musk, which is something he did many times over the course of the election cycle. Ezra, I’m wondering if the amount of money Elon poured into this election was counterproductive to him in some sense, because it gave politicians like Talarico a foil to play against.
Hugely counterproductive. Think about what that money is buying. It’s attempting to buy attention. But the sheer size of that money and its attachment to Elon Musk himself, who’s an attentional super-object, did get attention. And the attention it got was that Elon Musk is trying to buy the election from you, specifically in Texas.
Did Ken Paxton want one of the central themes of the closing weeks of the election to be “Elon Musk is trying to buy the election”? There’s broad anti-billionaire sentiment. You can poll this: people really don’t like billionaires. There’s specifically negative sentiment toward Elon Musk. I was a little surprised that the wealth tax proposition didn’t pass in California.
Yeah, I wanted to— Sorry, no, go ahead.
In California, there was a proposition to impose a wealth tax on the very wealthiest residents. It wouldn’t affect that many people, just a couple hundred, but it would be a hard tax to administer, and there was fear of billionaires fleeing California for, say, Texas. It didn’t pass. So, in this election, where anti-billionaire sentiment was very, very constant, the argument that driving some of them out of California would be bad for California actually won, which struck me as a little surprising. If you’d told me a week ago how badly Republicans would lose the election and asked what would happen to the wealth tax, I’d have said, “Oh, that’ll definitely pass.” It didn’t.
Aaron, I wanted to ask you about this because we talked about it last week, and you’ve done a lot of reporting on it. How do you account for this result in what was otherwise a very clearly anti-billionaire election?
It feels a little bit like a California thing, where people might just be tired of politicians throwing together a plan that doesn’t seem fully thought out and saying, “Just trust us; we’ll figure this out.” I think there was enough messaging that introduced doubts. The anti-Prop 40 groups were saying it was poorly written and they were just going to lose all the money. They need to do a better job. It’s not sustainable; it’s a one-time thing. I guess that resonated with people.
Honestly, it’s the one thing the billionaires I’ve seen and the tech right are clinging to this week. They’re saying, “If they can’t pass this kind of billionaire tax in a blue tsunami in California, it’ll never win anywhere in the country.” I don’t think that’s true, Ezra. I think there’ll be more attempts at this in the next few years. They’ll try to position it a little differently or find ways to make it resonate more with people. In general, I think people are tired of half-hearted propositions being thrown out there, and this one got a lot of criticism for that.
Yeah, I agree that the California proposition has the problem that Californians have begun to be skeptical of ballot propositions, full stop—which is correct, speaking as a Californian. But I do think the other issue was the specific concern that it would disadvantage California. I think it’s meaningful that Gavin Newsom didn’t support it.
Yeah.
As for the comfort the tech right is taking from this, I think it’s going to be basically table stakes in 2028 for Democrats to have a wealth tax proposal. The argument for a national wealth tax is a lot stronger than the argument for a state wealth tax, because they may say they will, but they’re not actually going to move to Ireland or wherever it might be. Argentina, maybe.
The long history of people moving to Argentina when things go south in their country.
Yeah. I think you’re going to have Gavin Newsom with a wealth tax proposal for that election. So the idea that this is gone, as opposed to it moving to the level at which it’s better administered—with the full force of the IRS and the full jurisdiction of the United States—is more the way I understand the situation.
Yeah. I want to talk about one more billionaire. Pour one out for Vivek Ramaswamy, who lost the governor’s race in Ohio. Vivek is more or less a tech billionaire. He was ascendant in politics, or seemed to be. He was one of the two faces of DOGE, and now he’s frankly humiliated. He probably should be ashamed, given that the final ad of his campaign involved putting his opponent Amy Acton’s mother on TV to call her daughter a liar after Acton had said her stepfather had abused her as a child.
I’m wondering if this race stuck out to you, and what you remember about his rise and fall as a figure—or if he was just a singularly bad candidate, regardless of the tech halo around him.
Look, Vivek and, for that matter, JD Vance are the kinds of MAGA politicians the tech right actually wants. They don’t want Donald Trump with all of his all-over-the-place-ness, right? They want these guys who are actually their guys and can go to their conferences and talk knowledgeably about the highly cerebral reactionary policies the tech right likes.
And they also like—
And the thing about—
You don’t have to convince him, right? They can just—
Yeah.
They already agree. They don’t have to dwell on it.
The thing about both of those guys is they’re bad politicians. Ramaswamy, I think, has a particular arc where he could have lost this election in an honorable way, but he decided to lose it in a maximally dishonorable way. It was a grotesque, awful, immoral thing to do, and it should cover Ramaswamy in shame.
I see this sometimes with Vance, with Ramaswamy, and with a bunch of Republicans: they speak Trump with an accent. Trump has this cruelty to him, this outrageousness, but because it’s natural to him, he knows where the lines are. He knows how to deliver it in a way where people don’t take it exactly that seriously. He’s done some pretty gross stuff in his time, quite a lot of stuff that is very, very gross. But he has a finesse to it that these guys, who are mimicking him as if they’re a GPT-3-level LLM, don’t have. They’re smart, but without any art to it whatsoever. They have no feel for the situation, and then it just comes out horrible and cruel. There’s a silent room where nobody’s laughing, and that’s that race.
Vance runs into this problem, too. The dogs-and-cats thing is an example of this for JD Vance. I think they believe they’re doing what Trump is doing, and they don’t understand that they’re not. To tie this a little bit to our main conversation, this speaks to a big problem the tech right is going to have. The tech right could only really survive as a factional player in a broader Republican ecosystem where its completely inhuman and bizarre views about how to treat other people weren’t dominant. It could have a lot of influence, and it was the smart player in the coalition. It had a lot of money. But ultimately, the thing was run by this guy with animal charisma and magnetic cunning.
When you reverse the foreground and background and put them and their people in front, it goes really, really badly. What the non-college white working class is looking for is not Vivek Ramaswamy. That’s just not the brief. I don’t think this is where politics is going. You see in the failure of some of these candidates, and in the problems I think JD Vance himself is going to have, how fragile the tech right is. It doesn’t understand, and often doesn’t seem to like, human beings—let alone the kinds of people who need to vote for things. It doesn’t want to listen to them. That’s not how you win in politics.
It’s kind of wild seeing them learn this in real time, because in 2024 they had this awakening where they realized, “Oh my God, we are so rich, and the amount of money we have, if you apply it to politics, is absurd. Our money goes so far in politics.” Fairshake, one of the largest PACs ever supporting crypto things, raised an amount that was just a drop in the bucket for some of these billionaires and the companies that donated to it. But now they’re starting to learn, as you’re saying, this is a great point: they’re starting to learn that it takes more than money.
Let’s pause for a second and look forward. I want to game out, or at least speculate about, what’s coming next. Two years ago, we were talking about a big vibe shift in Silicon Valley. I’m wondering whether we’re vibe shifting back again, and what the early signs are. I want to ask Aaron in particular: what are you seeing in the early days after the results? How are investors and other Silicon Valley figures reacting to this?
Well, first of all, we’re seeing a victory lap from the likes of Reid Hoffman.
Yeah, for sure.
It’s still a little bit confusing because he’s out there saying, “We should get money out of politics,” while he was one of the biggest left-wing donors in this last cycle. That’s a little odd. I think it’s going to take a little while to shake out.
To root.
I mean, maybe they’ll all hire their lobbyists, buddy up to people on the left, and figure out how to get their companies well-positioned for this next moment. But culturally, I think we’re still going to see people hanging out and espousing their right-wing ideas. I don’t know that this big forgetting everyone’s talking about is going to happen as rapidly. Ezra, I’m sure you have thoughts.
It’s not obvious to me that there’s going to be a vibe shift back in the other direction. Mark Zuckerberg takes off the chain—you have some things like that happen. But in the way that Andreessen and Musk and some of these other figures have become influencers, tech influencers online, do you have a group of them, maybe with Reid among them, who can do that in the other direction, given the headwind they’re up against? In some ways, I think this is a good moment to be a Satya Nadella.
Mm-hmm.
Just a normal tech CEO.
Just neutral.
Right? Someone who doesn’t seem to be out there trying to be a kind of comic-book villain or a podcaster—just somebody who runs a company in a seemingly normal way. I think that’s going to be the play. I don’t see a left-wing analogue to the All-In guys emerging.
Yeah, I think that’s true. I’d also suggest that part of what we’re seeing, as we’ve already been discussing, is a collection of liberal-to-left politicians who are pretty well-versed in tech and tech criticism beyond an anti-oligarchical stance. We already sort of swooned over AES' 60 Minutes interview, where he’s able to talk about diffusion but not in a way that suggests she’s fully bought in.
One of the really interesting contrasts between this blue-wave election and the one in, say, 2008 is that maybe I’m only speaking for myself, but I think we all associate Obama and the Obama administration with Silicon Valley and the tech industry of the 2010s.
Oh, yeah.
Now we’re looking at a Democratic Party that no longer has the same relationship to the tech industry. I’m interested not just in what CEOs emerge as normal or even liberal-leaning, but also in what politicians emerge and how their relationship to tech unfolds over the next few years. And it feels like—
I think the flip side for the Democratic Party—and this is going to be harder in this period—is: what do they want from tech, as opposed to just what they don’t want from tech? They don’t want billionaire and oligarchical influence. They don’t want AI to take all your jobs or kill the human race, or whatever it might be. They don’t want the effects on utility rates, water, and the environment.
If you go back a couple of years, and I thought a lot about this in terms of the work I was doing in this period, when climate was the issue, there was a natural union emerging between the Democratic Party and an area of tech. There’s some alternative universe where Elon Musk, who makes solar panels and Teslas, just remains a normie Democrat and is very influential from that perspective.
What a world that would have been.
The problem is that, particularly with AI, although not only with AI, the question of whether we want this is so salient. For most people right now, the answer is, “Not under these conditions, not if it’s going to do the thing you say it’s going to do”—you, the person who is building it, profiting off it, and in control of it. In that world, it’s very hard to have a positive politics of technology, because the technology is perceived, I think very correctly and understandably, first and foremost as a threat.
You go back to that 2008 world you’re talking about, and what tech meant in politics was the internet and social media, which were going to connect everybody. We were going to have more friends, and we were going to know about more things. And we had green energy.
And it was going to save the economy from the recession.
Mm-hmm.
It was going to save the economy, and there was green energy.
Yeah.
Right now, it has all cohered around AI, and the AI story is a much more mixed and frightening story for people at the moment. I don’t think you’re going to have any politicians running and saying, “What we need is way, way, way more AI.” I think you’ll have some saying, “There are things we can get from AI,” in terms of drug development and other domains, “and we should encourage that.”
But yeah, the problem with the tech industry right now is that they’re all in—no pun intended—on something people increasingly don’t think they want, and that’s not a messaging problem. I just wrote a column about this the other week. Max, you wrote about it, with this idea of, like, on We Risk. But even when you hear them talk about the world they’re trying to get to—a world where human beings are the well-kept pets of these superintelligent AIs—that world sucks.
Mm-hmm.
People don’t want that world. That’s not an inspiring vision. There’s a real problem here, not a messaging problem. Too many people in Silicon Valley are building toward a future that, if it goes well—if you’re not in the existential-risk scenario—people don’t want either. I have a 7- and a 5-year-old, and it’s completely unclear what it would mean to train them to have jobs.
They’re not going to be prompt engineers? That no longer seems to be a growth industry.
Okay, we’re going to take a break. When we come back, we’re going to do a segment that—we’re trying out a new name. We’re going to see how it plays. The new name is “Pacing the Frontier,” and we’re all going to take stories—
This is a really negative reaction so far. If you’re not watching this on YouTube, I’m getting the stoniest faces I could imagine. We’re going to talk about stories that we’re tracking ourselves, that maybe other people aren’t looking at, that we think are going to be important over the next few weeks, months, or years.
This is Hard Fork. I’m with Aaron Griffith and Ezra Klein.
Wait, I thought you were going to announce a name for the segment.
Pacing the Frontier.
I don’t get it.
Ugh. I know it doesn’t really make sense, does it?
Welcome back to Hard Fork. I’m here with Aaron Griffith and Ezra Klein. We’re doing a segment that, before the break, was called “Pacing the Frontier,” and after the break is no longer called “Pacing the Frontier.” We’re asking our guests about stories they’re tracking that we think are going to be important over the next few weeks or months, but that nobody—or not enough people—are paying attention to right now. Erin, what did you bring to show and tell today?
There’s a great story by our colleagues Dana Goldstein and Natasha Singer about Google announcing that it’s going to turn off Search—the main thing we know Google for—for schools that don’t want to use its AI products. This is notable to me because Google is in many, many schools. Google Chromebooks are probably the most-used laptops in schools, so kids are using Google Search a lot.
There are a lot of schools, including New York City schools, I believe, that don’t want their kids exposed to or using AI. Google is basically saying, “You don’t want our AI? Well, fine, you don’t get our Search either,” which is such a classic tech industry move: “I’m taking my ball and going home because you don’t like me.” Google had to come out and clarify, saying, “Okay, well, we’re going to figure out a way to work through this.” But essentially, any schools that don’t want AI also don’t get Google Search.
There’s definitely a backlash to this, but I think it also speaks to this moment we’re in, with schools and parents and everyone pushing away from tech in schools and education. Google is kind of going the other direction. So this story is something that’s going to continue and be something to watch.
Yeah, and it’s a good story to talk about on this episode in particular because it feels like this is an issue that’s going to weigh heavily on the electorate and politicians for the next 2 years. Speaking as a millennial parent, I get the feeling that a top issue for us voting millennial parents is going to be keeping AI the hell away from kids and education, and maybe away from kids entirely.
When the tech company doesn’t seem to be listening, if you’re a campaign manager or a chief of staff somewhere, you’re going to want to stick your crowbar in and use your leverage to make it a big issue.
I mean, I don’t know why they don’t just use DuckDuckGo, but—
I do think all classrooms should be very aware of surveillance and privacy issues. I’m happy to come and lecture at elementary schools about this stuff. Ezra, what story are you tracking this week?
There’s this permitting reform bill, and I think it’s relevant to this whole conversation in a couple of ways. This permitting reform bill was struck. It’s a bipartisan bill between, on the Democratic side, Senators Sheldon Whitehouse of Rhode Island and Martin Heinrich of New Mexico, and, on the Republican side, Shelley Moore Capito of West Virginia and Mike Lee of Utah.
It’s a big bill, and it would take a long time to go through it. Parts of it I don’t understand. But it basically makes it easier to build energy of all kinds—green energy, non-green energy—and it also has data center provisions about trying to internalize more of the costs of data centers.
This is a bill that’s kind of splitting the Democratic Party a bit. It’s there in the lame duck, but it’s really a movement around how easy it is to build energy in this country. And as it gets amended and worked through, the data center backlash can find a place to express itself.
So this bill is going to be a signal for how much bipartisan dealmaking the two sides want to do, particularly in the lame-duck session right now. How much Democrats are going to want to stick with that, now that they have so much power coming up, is a very open question. But it’s also a place for this question of how easy it should be to build energy, particularly at a time when a lot more natural gas is being built to power data centers, and what the national rules on building data centers should be.
This bill could become a vehicle where a lot of that gets fought out, or we could see that there’s not enough consensus for even a very, very bipartisan piece of legislation to move forward on this right now.
And what’s the sort of time frame you’re looking at? Is this over the next several months?
The last I heard, it was going to start moving forward in the Senate in the next week or two.
Mm.
But yeah, over the next several months.
Yeah, so we’ll get a good picture on how willing Trump and the Republicans are to make those deals pretty soon.
And the Democrats, right?
Yeah.
Because I think the theory right now is that if the deal doesn’t get made in the lame duck, it probably will not move forward. It’s actually going to be harder in some ways under a Democratic Congress, even though the bill does have Democratic support.
I see.
But we’ll see.
Interesting. My story is kind of a non-story. I was shocked that prediction markets didn’t make more of a splash this week. That’s not to say they weren’t everywhere, being advertised to constantly. You walk outside the Times offices, they’re all over Times Square. There are Kalshi and Polymarket ads.
But they seem to have largely gotten the elections right. I’m using air quotes because I think it’s a question of philosophy and epistemology—what prediction markets are actually doing in terms of getting things right. But unlike the Brazilian elections a few weeks ago, there weren’t any huge swings. We didn’t have any crazy stories about wins or losses.
It just feels like less of a coming-out party, and more a moment when prediction markets have been folded fairly thoroughly into the elections and newsmaking apparatus.
I don’t know. I’m a little skeptical of this whole thing. Go look on Polymarket at the prediction market for the Mississippi Senate race.
Okay, that’s fair. That one did swing.
Just look at what that— I’m going to be—
What happened there? I missed that. Were they really wrong?
Yes. But it’s more the way the thing all of a sudden—you get the actual information, and then the 2 lines go like this.
Yeah.
Yeah.
Right?
Yeah.
And it’s not like the prediction markets got something wildly right either. They’re willing to swing faster in the direction of the information we’re seeing, but—
Yeah.
They do kind of track the polls quite a bit in many other races.
Yeah. But really, more than anything, it just felt like this was an election where we all kind of nodded our heads and said, “Yeah, of course there are prediction markets, and they exist, and they’re there on every TV screen all the time.” I’m not saying that’s good, and I’m not saying it’s bad, though I have my suspicions. It was just an interesting feature of the week.
I mean, maybe it’s just that they’re becoming more sophisticated, and there are fewer suckers out there, so—
I mean, I think there’s smarter and smarter money.
You know, there’s less money to be made.
Yeah. Sadly for—
Sorry for you.
—degenerate gamblers. Not for me. Not for me.
Okay. Yeah.
I’m an ethical journalist. I don’t have any stakes in any bets on the election. Thank you, Ezra and Erin, for joining us. This was an unbelievably stimulating discussion. I had a really good time. Everybody else stick around. We’ve got Teddy Schleifer coming on to talk about who spent their money wisely and who spent their money poorly. I’m going to give you a hint about who’s in the latter category: his initials are E.M. Thank you, guys.
Thank you.
Thanks for having us.
Welcome back to Hard Fork. I’m here with Times reporter Teddy Schleifer to talk about winners and losers in tech from this week’s elections. Teddy, you’ve spent the last several months reporting on campaign money, especially as it pertains to the tech industry.
Yeah.
In this cycle, I know from your reporting, the top 20 donors injected about $2 billion into the midterms. More than $800 million of that came from the tech industry in one way or another. I want to start by identifying the big players. When we’re talking about the top tech donors, who are we talking about?
Sure. The number one tech donor is Elon Musk. He spent about $470 million to $500 million on the election. Other tech people folks may have heard of, Marc Andreessen and Ben Horowitz, the founders of the venture capital firm Andreessen Horowitz, spent $100 million themselves. There’s also the company Coinbase, which joined Andreessen Horowitz in backing a big crypto super PAC this cycle. And there are a couple of other individuals, such as Greg Brockman, the president of OpenAI, who spent a ton of money on MAGA Inc., the Trump super PAC, and has also been a huge proponent of all this money supporting so-called pro-AI politicians.
A couple of years ago, it felt like a sort of conventional wisdom was coalescing among at least some people in the Valley that they were so rich—maybe Elon in particular was so rich—they could just spend freely and get what they wanted out of politics. It feels a little bit like a big lesson from this election is that that’s not really true. Tell me a little bit about how spending has functioned here, how it’s maybe different from 2 years ago, and what kind of lessons we might take away from it.
Sure. Sixteen years ago, the Citizens United decision came down from the Supreme Court, which began this new era of anything-goes billionaire political spending. Now people can cut checks of unlimited sizes to political committees. Every cycle, the zeros keep adding up to more and more. In 2024, Elon Musk spent somewhere around $300 million, and that’s just in disclosed giving. We’re not even touching all the undisclosed giving that happens.
The narrative formed, I think somewhat unfairly, that it was a raging success. What Musk did in 2024 was perceived as successful, and I think the gospel of the entire world of money and politics is that money helps with politics—that you win elections by spending more money. This cycle—and this is a meta-critique of my own stuff over the last decade—I don’t know. There are times when that’s true, and there are times when it’s not true. Last night was a vivid illustration of the fact that you can spend, in Musk’s case, $500 million or $450 million, and this stuff doesn’t necessarily convert.
Yeah. So part of what I want to do in talking to you is create—I don’t want to call it a loser list. Let’s call it a return-on-investment power ranking. I want to know whose money was well spent and whose money wasn’t. I want to talk about some of the people you were just mentioning. These are the names that seem like they’ve done particularly well. Feel free to challenge me if you disagree. Reid Hoffman is basically the only Democratic donor among the tech group in the top 20. It’s got to feel like he did pretty well with his money.
Yeah. Reid’s single biggest investment this cycle, at least publicly, was $17 million supporting James Talarico, including in the competitive Democratic primary. The general election was competitive, too, and Talarico won by a shockingly large margin. In retrospect, it makes you wonder if Reid even needed to spend anything in the general. He’s somebody who knows a thing or two about political spending at this point. He’s been doing this for 10 or 15 years. He believed that Talarico was a generational talent and the sort of person who could win Texas, and he won.
So we can maybe put him somewhere near the top.
Sure. Let’s do that.
Another person I want to talk about didn’t come up before because he hasn’t been a huge donor nationally across the midterm cycle, but Sergey Brin did pretty well for himself.
Yeah. Sergey’s not on the list we mentioned because all of his giving is at the state level. Sergey Brin, the founder of Google, is a rookie, if Reid is a veteran. He spent over $100 million this cycle on a California political effort called Building a Better California. Brin basically started this group to push back on the proposed wealth tax in California, which, at least as of this recording, seems like it’s going to fail.
That’s in large part because Sergey and other wealthy people from tech spent $330 million on a political effort to cripple this thing, and it worked. They were able to demolish this thing financially. Not only did they raise $333 million, but the pro side—the union pushing this billionaire’s tax in California—basically raised bupkis. They were outspent tremendously, and it’s a testament to money in politics.
So these 2 guys, Sergey and Reid, we might put at the top of our list—
Yeah.
Let’s do it. Partly because they’re the only guys who got what they wanted out of the races last night. I want to turn to the crypto people, because this is a whole grouping that you’ve put together when we’re looking at the top 20 donors. We have the Winklevoss twins, who are big donors to Trump and Trump-aligned PACs, and there’s also this thing called Fairshake, a super PAC that’s crypto-aligned. Where are the crypto people spending their money? What are they looking for?
Yeah. This started in 2024, when Fairshake raised somewhere in the $100 million to $150 million range from a couple of key crypto companies, including Coinbase, Ripple, and Andreessen Horowitz. The narrative was that it was a raging success and that they were able to elect a very pro-crypto Congress. Even Trump ended up being seen as pro-crypto.
Yeah.
They wanted to keep it going in 2026. They had some successes in primaries, like a congressional race in Maryland, where they elected a pro-crypto person. They also had some losses in primaries, including the Illinois Senate race. But this November, their big bet was the same as in 2024: to defeat Sherrod Brown, the Ohio senator who used to be in charge of the Senate Banking Committee, which of course regulates crypto. It’s a big deal in Washington.
Fairshake spent $45 million on ads against Sherrod Brown in September and October, and Brown won last night. I think people felt there was a sense of invincibility or infallibility around crypto: if you were going to oppose the crypto industry, you were going to get demolished. That would resonate with Sherrod Brown, because he lost in 2024, in part because the crypto industry spent, I think, $40 million. But this time, he won.
The crypto industry has won some other races, so I don’t want to say they’re totally at the bottom of the losers list. But in their highest-profile race, it’s an L for them.
Yeah, there’s lots of room on the losers list.
Sure.
One interesting thing here is that a whole narrative got built around both the money crypto was spending and a sense that there was this demographic shift in 2024. Young men were becoming conservative or Republican. There was a sense that maybe Democrats had to appeal to crypto owners and crypto traders. This was a narrative pushed quite hard by the crypto industry.
It feels like the armor has been pierced a little bit on this. Maybe we’ll see Democrats being a little more aggressive with the crypto industry after something like that.
Yeah. The reason Fairshake got involved in such a significant way this cycle was that they encountered the pushback you’re talking about.
Basically, in September, it wasn’t clear whether the crypto industry was going to spend all this money it had. The way these super PACs tend to work now, which I think is kind of interesting, is they raise a ton of money, and then it sits in a bank account. They don’t necessarily always spend it. It’s like a permanent looming threat: “Shame if something happened to you.”
Democrats blocked a crypto policy bill called the Clarity Act in September, and the next day it was time to punish the Democrats. They spent all this money against Brown. If you believe the crypto industry is overrated politically, you’d obviously be emboldened by Brown staring down $45 million and winning.
Yeah. Let’s talk about another set of donors to the Fairshake super PAC.
Yeah.
Andreessen Horowitz. This is one of Silicon Valley’s most famous and prestigious VC firms, but also quite memorably a conservative one. Marc Andreessen is obviously one of the loudest pro-Trump voices in the Valley. They were big donors in this cycle. They donated to Fairshake, but they were also donating to an AI super PAC called Leading the Future. Tell me a little bit about the money they were spending and where it was going.
Sure. It’s fascinating if you’re into Andreessen, have read his biography, or are familiar with his political oeuvre, because he was an institution in Democratic fundraising in his 20s. Andreessen’s been around forever, and for a while, until Elon Musk entered the picture, he and his firm were the biggest political donors of the entire cycle.
I wrote a story about how they got into this earlier this year. The AI stuff they’re doing is the most interesting.
Yeah.
After Fairshake spent all this money in 2024, the crypto industry was seen as a political savant. They basically said, “Hey, let’s do the same thing with AI.”
Mm-hmm.
The AI industry was nascent in 2024. By 2025 and 2026, it was becoming this huge thing. It’s the exact same model. Obviously, it’s not crypto companies funding it. It’s AI firms, like OpenAI co-founder Greg Brockman, and Andreessen Horowitz, which has a ton of investments in AI. They basically ran it back with AI.
The difference, as was probably predictable in retrospect, or foreseeable, is that people have a lot stronger opinions about AI than they do about crypto.
Right. It’s funny to say that AI feels more toxic than crypto right now, but as we’re making our losers list, it feels a bit like that’s what we’re learning.
Yeah. I do think there’s some political upside to crypto, especially because the downside is that it’s not a salient issue for people. Taking money from crypto can be a net positive decision strategically.
Right. But let’s keep talking about AI, because perhaps even below Andreessen Horowitz on our list is Greg Brockman, co-founder of OpenAI and its president, one of the most prominent people at the company after Sam Altman, and a major donor to Leading the Future. You reported on the controversy that sprang up inside OpenAI. I found that story incredibly fascinating. Tell me about that.
I was going to say, he’s the biggest donor—one of the biggest donors—to Leading the Future, with the asterisk that it was pledged. This shows how quickly the politics of AI have changed. Greg Brockman and his wife, Anna, committed to giving $50 million to Leading the Future in 2025. That was matched by $50 million from Andreessen Horowitz. Only the latter actually happened.
The Brockmans gave the first $25 million in 2025, when AI was a policy issue, but not like it is today. Then, a year later, they backed out of the second $25 million they were going to give, so instead of two installments of $25 million, there was one. That was in part because Greg Brockman said internally at the company that he was upset his support for the super PAC was becoming a distraction.
We’ve also seen something similar with Meta. As part of its political engagement this cycle, it retracted about $20 million to $25 million from its state political giving. The timing was similar to the Brockman gift: as the politics were getting worse, they wanted to spend less money.
Right.
This is tough, and it just shows how quickly things have changed.
Yeah. It does seem like we’re seeing a revival of some of the internal employee pushback that was really common in the tech industry at the end of the 2010s, and then got really quashed during the Biden administration and the second Trump administration. It seems clear that if you’re an executive, you want to be careful about what causes your name is attached to and what kind of politics it’s attached to.
Yeah. That’s why we have the beauty of dark money groups: if you want to do something, there’s a way to do it. I mean, that’s a joke, but we’re talking about gifts to disclosed vehicles, and so much of the world of money and politics and billionaire influence now funnels through things that are undisclosed.
If you’re Greg Brockman and you’re listening to this and you want to get another $25 million into the political system, there’s a nonprofit out there—
Yeah.
Waiting to take your money.
Okay, so we’ve constructed a list here. We’ve got a couple of people who got what they wanted, and then a lot of people who really didn’t, who spent their money in different ways. I think at the very bottom of our list, no matter what—
Yeah.
—is Elon Musk, who spent $400-plus million and didn’t—I don’t know that he won a single race he spent money on or got a positive outcome out of it. Beyond that, even more than somebody like Greg Brockman, who became a story when he didn’t want to, Elon became an enemy, especially in Texas—
Mm-hmm.
—where he was constantly invoked by the foil in the Senate race. I don’t even know where to begin with this. What happened with Elon’s money here, and why did it turn out so poorly, in your estimation?
Yeah. Let’s roll back the clock to last spring, the spring of 2025, when there was this special election in Wisconsin. Musk—this was a random election people probably don’t even remember—spent a ton of capital, literally and in terms of his time, on this judicial race in a state where he again became the main character. This was when he was leading the Department of Government Efficiency.
After the guy lost in Wisconsin—his name is Brad Schimel—I think Musk realized, based on some reporting, that being so visible in his political support wasn’t working out. He wanted a much, much lower profile going forward. That election was basically why Musk, for over a year, wasn’t really doing much in the public eye. He was talking at one point about starting a third party. He never did that. He never recanted it, I should also add, so it’s still technically a live situation.
It’s so crazy that Elon Musk announced something he was going to do—
Ah, man.
—and then it didn’t happen. That’s probably the first time that’s ever happened.
He does delete tweets, though, and he has not deleted the tweets about the third party. This time, Musk was barely saying anything about it. He said one time, “Vote red today,” which he did delete later, at some point overnight on Tuesday night.
His spending was concentrated in a few places. We don’t have all the details yet because of campaign finance reporting, but probably about half of it, maybe a little more, was centered on Texas and Ken Paxton’s race. That’s why James Talarico was able to make Musk a foil at the end, including in his victory speech last night, where he basically took a shot at Musk.
Paxton lost by a lot, and ultimately, that sucks for Elon Musk, so he deserves the bottom spot here. But to what extent is he to blame? Paxton probably would have lost even if Musk had spent zero. Maybe the argument there, the return on investment, is not to spend in a race you’re going to lose, because at least that way you still have the $500 million, right?
Yeah. I think you could also argue that Elon Musk was going to become a character in this race no matter what, since he’s an important figure, a huge right-wing figure in Texas—
Yeah.
—and around the country.
Yeah.
Can we look forward a little bit? We’re still sort of—the dust is still settling on this week’s elections. We’re still figuring out what happened. Are you expecting all of the people on this list to be on it again in 2028? Are there figures we should be looking out for who weren’t on it this cycle?
Yeah, let’s start with Musk. The most extensive thing he said about the midterms was a couple of days ago, when he basically said his fear, and the reason he was so involved in the midterms this cycle, was about 2029. His fear is that there’d be a Democratic trifecta, with the White House and both houses of Congress, which they now do have, and that would lead to, in his view, the end of civilization. I think he said “game over for America” or something like that.
That means Democrats are going to let all the illegals in from Mexico, register them all as voters, and then it’s the end of civilization, blah, blah, blah. I read that as: Yes, Elon Musk plans to play in the 2028 Republican primary. Musk has a pretty good relationship with JD Vance, probably not a surprise. But that relationship has stayed pretty good, even with the ups and downs with Trump. I’d be shocked if Musk didn’t support Vance.
Mm-hmm.
The other people on the list we’ve mentioned, I don’t know, because some of the folks who are really tied to a specific industry are effectively doing pay-for-play lobbying. I don’t mean that in a nefarious way, but the crypto industry and the AI industry are really engaged on specific issues. It really depends on what politicians do on those issues.
We’re also talking about Republicans. In the Democratic primary, will there be a crypto candidate who gets money from the crypto industry? There will certainly be a range of opinions on AI, even in the Democratic primary. Will there be someone who is more pro-AI and gets some money there?
One interest group from the tech sector that I think is fascinating, and that we haven’t talked about, is the AI safety money.
Yeah.
That’s basically the backlash to the Andreessen view of the world. That is a huge source of cash. I think one of the big takeaways from this cycle is that there’s a ton of money in the Anthropic, effective altruist, lowercase-D, maybe uppercase-D Democratic world, among people who want the AI companies to move more slowly and are more concerned about existential risk.
That’s absolutely going to be a force in the Democratic primary. In 2028, Dario Amodei could be on this list—
Right.
—or even some people at OpenAI who are more partial to that community.
Yeah, I expect, given your beat, that you’re looking with some interest at the Anthropic IPO coming up soon—
Yeah.
—because you’re going to mint a bunch of billionaires who have exactly the passions you’re talking about and exactly the interests, and who are going to spend.
Yeah. They’re already on the cap table of the company, to some extent, and I think they’re on the campaign finance cap table, so to speak.
Yeah.
People like Eric Schmidt, who’s a big investor in Anthropic, are huge Democratic donors. Dustin Moskovitz, one of the earliest investors in the company, is also a huge Democratic donor.
The resources created by these IPOs are going to mint a lot of people who aren’t necessarily billionaires. It’s a very civically minded community of people at Anthropic, broadly speaking. These are people who could easily give $1 million, $2 million, or $3 million. I know Democratic fundraisers are also watching the Anthropic IPO closely, hoping that a lot of 30-year-olds are going to make a ton of money and want to host Josh Shapiro when he comes to San Francisco, for example.
Right. You mentioned dark money before. Is there a version of the next cycle where a lot of the people on the list right now are spending the same kind of money, but keeping it much more secret?
Yeah. That’s a huge caveat to everything I’ve said: I’m very curious, and hopefully self-aware, about how much we don’t know about what’s happening in this space. I’ve been writing about this on and off for 10 years, and there’s less disclosed than ever, especially in the Democratic world, where so much political giving now happens through nonprofits.
Reading a campaign finance report is to be inundated with an alphabet soup of nonprofits. There are groups called Americans for America Inc., all started two weeks before you’re reading the report. Suddenly there’s a $30 million super PAC that got a donation from Americans for Americans Inc., and it’s like—
Yeah.
—how did that happen? That’s an attractive way of engaging in politics now, especially in the Trump era, when lots of Democrats are fearful of investigations. Trump regularly threatens big Democratic and progressive donors, and he’s gone after Harvard and other Ivy League institutions for philanthropic gifts and tried to withhold research money.
There’s a wholly legitimate reason why I think liberals fear the Trump administration and want to have their political giving be more shrouded in secrecy. But the downside is no one really knows much about—
Yeah.
About this stuff now. None of the nonprofits made the list you’re referencing because there are so many of them, honestly.
Mm, yeah.
It’s like, the number of different nonprofits that started in August, each donating $10 million—you could have your own list made of just that.
All right. Well, on that very cheery note—
Yeah.
Teddy, I want to thank you for coming by and talking to us about this stuff. Thanks for coming on, Teddy.
You bet.