Four thoughts on the inevitable, but faster than expected, fall of Jensen Huang.
First, it has been obvious for more than a year that something was very, very wrong with Nvidia and its CEO. As I wrote last September, in More Like Jensen Wrong, Amirite?:
Nvidia and its CEO, Jensen Huang, have consistently said things that are, well, just not true, in an obvious, “we know they’re not true, and you know they’re not true, and you know that we know, and we know that you know, and yet you’re still saying them” sort of way.
And then, in Break Up Nvidia:
Nvidia’s efforts at corrupting American politics are at least as concerning. … Nvidia of course has the right to petition the government to redress its grievances, in the language of the First Amendment, or to sic its lobbyists on the Capitol and the White House, in modern parlance. But along with that right, as the beneficiary of and participant in a liberal democracy, Nvidia has the obligation to use facts that are true and arguments that have logical coherence. Its failure to do so, preferring instead to rely on contradictions, outright lies, nonsensical name-calling, and efforts to ruin careers, has raised more than eyebrows—it raises the question of how a political system should treat a company that thinks it’s a mob boss.
The program here is not to advocate or persuade but to leverage and threaten. Misbehaving with impunity is itself an assertion of power, intended to demonstrate that the rules do not apply. Note also the silence from Nvidia customers as Congress considered legislation that would put them ahead of China in line to buy chips—a policy in which they would have an enormous stake. Either these firms no longer care about buying as many AI chips as they can get their hands on, they’ve decided to disavow political advocacy on policy issues of particular concern, or they are terrified to say anything that might antagonize their supplier. One of these explanations seems more likely than the others.
By the time I got to The Geniuses Playing Chinese Checkers, I could not help observing that Jensen was “just delivering supportive platitudes about the CCP.” And in Will AI Worsen Affordability Woes?, I wrote, “the line between vigorously pursuing profit for your company and lying on behalf of the Chinese Communist Party is not a thin one. We should reasonably expect American executives to remain on the right side of it.”
Honestly, I never considered that Huang was literally in bed with the CCP, but I didn’t have to. Just listening to the absurdities pouring forth—
On U.S. firms doing business in China: “They want companies to come to China and compete in the marketplace, and I believe that. … I do hope because they say it—their leaders say it. And I take it at face value.”
On the AI race: “Does it really matter who gets there first? I think in the final analysis, I don’t think it does.”
On chip smuggling: “There’s no evidence of any AI chip diversion. These are massive systems. The Grace Blackwell system is nearly two tons, and so you’re not going to be putting that in your pocket or your backpack anytime soon.”
On education: “Basic math is being forgotten. Does it matter? … I don’t think it does. I don’t think it does.”
—and applying some basic critical thinking to the situation made clear that this person was incapable of participating in good-faith reasoning and should not have been given the time of day by the public officials responsible for representing the interests of the American people. The company’s efforts to get people fired from think tanks for disagreeing with its nonsense provided a helpful warning, too.
Huang bears responsibility for his personal conduct, but every one of his employers—the Nvidia board of directors—should be thoroughly humiliated as well. So should everyone who took the propaganda about chip exports as serious arguments about public policy. The many commentators and competitors, who chose to stay quiet lest Nvidia train its ire on them, enabled the whole thing.
In short, don’t be these guys, having to “pivot” to the answer you knew was right all along: “OpenAI has privately told congressional offices that it supports three bills aimed at making advanced semiconductors and the tools used to make them harder for China to access, according to two people familiar with the matter granted anonymity to discuss private details. The voice of support for the trio of measures, first reported by Punchbowl, is a pivot for the AI giant, which toned down its previous advocacy for tightening export controls in talks on Capitol Hill this year, a shift that came as the company deepened its financial ties with the U.S. chip giant Nvidia.” (Politico)
And for God’s sake, do pass those three bills as part of this year’s NDAA (American Compass has a policy brief on the details). For once, it is not hyperbole to say that the argument on the other side turns out to have been a pile of craven lies.
Second, can we please stop conflating technical prowess and wealth with wisdom and commitment to the national interest? Not to say these things are incompatible, or that one is disqualified from commenting usefully on public affairs by dint of leadership at a large Bay Area company. But we should at least insist on some evidence that a given CEO is doing more than talking his own book (whether on behalf of shareholders or girlfriends is beside the point) before crediting his point of view. As between a man on the street who would probably say, “no, we should not sell advanced AI chips to China” and a man in a leather jacket who says, “yes, we should sell my advanced AI chips to China,” it’s not hard to see whose insight would ideally be more welcome in the White House.
Beyond the basic and timeless problem of self-interest, there is also something more deeply rotten in our tech sector’s leadership class. You can see this in the many prominent people coming forward to argue that “Jensen is irreplaceable for America’s AI leadership” and “an incredible human” and “he’s gotta stay in that seat at all costs.” You get “journalist cancellation cannot stop technologist acceleration.” These are reasons for people in important leadership positions to take their responsibilities seriously, not reasons to excuse a failure to do so. There is a straight line from the notion that elites must get a free pass because of their excellence to an irredeemably corrupt elite. (Indeed, a not-corrupt elite would never dare make such a foolish argument in the first place.)
Third, China is not our friend. Can we please dispense now with the notion that American companies should be doing business in China? It’s bad enough when we hear it from the business leaders who want to cash in quickly, but we still have policymakers like Janet Yellen saying things like “we gain and China gains from trade and investment that is as open as possible.” Stop.
David Ignatius has a good column on this in the Washington Post: The Nvidia spy scandal was hiding in plain sight. None of this is new, but hopefully this is sufficiently high profile, as COVID shortages were in the context of supply-chain resilience, to wake people up.
Fourth, the investigations need to stay focused on the economic and national security dimensions of the debacle. One can already sense the push to make this story about “Jensen the playboy.” The actual story here is about entirely unacceptable personal conduct that led to corruption of both the world’s most valuable corporation and the federal government’s policymaking apparatus. Good riddance to Huang, but what is happening to our country?
GOOD READS FOR YOUR WEEKEND
America Called. It Wants Its Government Back. (Jacob Weisberg, New York Times)
One of the better attempts to articulate the frustration so obviously present in the country, motivating yet another wave election after yet another failure of representative government. This line caught my eye, given its resonance with our work on Reclaiming American Citizenship:
“The Democrats offered a more hopeful vision of what Americans could accomplish together, rooted in mutual obligation.” Whether they have the principles or agenda to deliver on that vision is another matter.
The Amnesty Crowd Shouldn’t Get Too Excited. (Mark Krikorian, National Review)
What does the next two years hold for immigration policy, and how to translate the president’s latest remarks on the matter? Always read Mark.
An American Furniture Maker Tried to Bring Jobs Back Home. It Didn’t Work. (Wall Street Journal)
A very interesting story on reshoring gone wrong. The tariffs have their desired effect, pushing production out of China, and then prompting an effort at domestic production. But much also goes wrong, including a disconnect with the workforce and, most importantly, a failure to make a serious investment in the kind of production that might be competitive. No, you can’t just “[buy] the factory from its previous owners for zero dollars” and expect to succeed. You end up with workers completing one-third as many sofas per day at wages five times higher. Regardless, there’s no free trade here. Only a broken industrial basis, a loss of know-how, and labor arbitrage in place of genuine comparative advantage.
BAD READS
It Was the Economy, Not the War (Mark Penn and Andrew Stein, Wall Street Journal)
If you want to know whether there’s a serious case that tariffs, rather than the Iran war, were driving American frustrations and affordability concerns, turn to the WSJ editorial page, which surely will make the best case it can. So let’s check in on their effort:
According to the Harvard CAPS/Harris poll, 58% of voters were unhappy with the economy when they showed Democrats the door in 2024, compared with 64% late last month. They were happier with the economy when Donald Trump was elected, but that faded month after month, showing the biggest decline when tariffs were issued and then when Democratic campaign ads were in full swing.
Striking! The Journal provides no link, probably a good move in this case, because here is the actual data for approval of Trump on the economy:
Approval in late March 2025, just before Liberation Day: 45%
Approval in late February 2026, just before Iran war: 45%
Approval in October 2026, just before election: 34%
Huh! Approval of handling inflation actually goes up from 43% to 44% (obviously within the margin of error, but still), and then down to 33%.
As for the war in Iran, here’s the pitch that voters actually like it and want it to continue:
True, when the Fox News Voter Poll asked if going to war with Iran was the right or wrong decision, roughly 6 in 10 voters said it was the wrong decision. But the Harvard CAPS/Harris polls asked follow-up questions that suggest voters don’t want to abandon the war—they want to finish it the right way. Asked if the Iranians were negotiating in good faith, 65% of respondents said no. Should the Iranians be allowed to control the Strait of Hormuz? Sixty-seven percent said no, including a majority of both parties.
The poll also asked if Iran should be permitted to obtain a nuclear weapon, and 85% said no. People agree with the objectives of the conflict, and those objectives may not be achievable without conflict. Asked whether the U.S. is justified in using force if the Iranians refuse to yield on these key issues, 63% said yes. The war has dragged on when voters wanted it to be won—and they still do.
See, none of those results suggest “don’t want to abandon the war—they want to finish it the right way.” The results suggest Americans like good things and don’t like bad things. The problem with Team War is that they continue to confuse things we wish would happen to things we can achieve. The American people, fortunately, are smarter.
There’s No Such Thing as a Stock-Market Bubble (Jeff Yass, Wall Street Journal)
I just really enjoyed this one. “What is a bubble? The bubble callers mean a price that hype has pushed above fundamental value. Notice what that assumes: that there is a true price, that the market has lost sight of it, and that when prices come down, reality is correcting a delusion. This totally misunderstands prices under uncertainty and volatility. There is no true price to lose sight of.” See, value is whatever markets say is the value, therefore markets are always right. Who could argue with that?
CHECKING IN ON REINDUSTRIALIZATION
Tired: Tariffs on Canadian lumber or making housing affordable.
Wired: Can Hotels, Libraries and Apartment Towers Absorb America’s Timber Glut? (Wall Street Journal). “Proponents of mass timber say it creates a new market for the glut of wood that has depressed prices and put U.S. timberland at risk of being cleared for more profitable uses, such as shopping centers and solar farms.”
Also wired:
Honda to Invest $3 Billion in New Ohio Plant to Boost Hybrid Vehicle Production (Wall Street Journal).
Claude for Manufacturing: AI plugin for plant teams (Anthropic).
Mexico says crackdown on Asian steel shipments is paying off as US tariff talks advance (Reuters).
AND YOUR CANADA GOOSE NEEDS NO INTRODUCTION…
…nor does this really require any comment. Prime Minister Mark Carney on his upcoming meeting with President Xi: “We’ll discuss our economic partnership which is something that is deepening… In terms of the security risks of a partnership, they are much much much much much less than the security risk, for example, of the United States.”
Enjoy the weekend!