The continued debate over whether Jensen Huang’s lover Alice Huang, the controlling shareholder of a Nvidia Singapore customer, Megaspeed, was a Chinese spy is a diversion from evidence ofserious abuses at Nvidia.1 The Wall Street Journal has just published important evidence of Nvidia’s and in particular Megaspeed’s chip laundering operations in Southeast Asia. The Hill makes clear that not just the State Department, which appears to have earlier leaked other damaging information to the New York Times, has the knives out for Huang, but also other major Administration departments, including Defense.
In the meantime, in a sign of delusion, Jensen Huang is plotting a return, despite the filing of shareholder suits against Nvidia’s executives and board and a lack of D&O insurance.2 This comes with reports of employee exits and a big fall in Nvidia’s competitive position.
We will also start to look at the seemingly overwhelming topic of Nvidia’s circular financing. We suspect that most have been deterred by the MEGO-inducing yarn diagram of inter-connections among Nvidia customers and AI infrastructure players, such as data center funders. One way to pull on these strings is to look at underlying frauds. One is channel-stuffing, as in accelerating the recognition of sales revenue by pushing more product onto buyers than they want or need (as in beyond what one might consider to be prudent inventorying). That has the effect of moving sales that would otherwise occur in future quarters to the current one. While not legally classified as a stand-alone fraud, it has been successfully prosecuted, with Enron and Bristol-Myers as examples.
Another may be whether Nvidia has so heavily subsidized some of its nominal sales that they cannot be classified as true sales. True sales is not a legally well-developed question. It has been fleshed out a bit in the context of securitization, as in whether a bank that made a loan and sold it to a securitization trust really truly had no more interest in it.3
First to the Wall Street Journal’s lead story, How Jensen Huang Built the Backdoor Route That Keeps Beijing in the AI Race. Key sections:
An hour outside of the Malaysian capital…a Chinese company plans to use Nvidia’s advanced artificial-intelligence chips. Thanks to Chinese customers, data-center industry leaders say Malaysia could become the second-largest data center hub in the world, after Virginia, packed with top-of-the-line Nvidia chips China’s leading AI companies can use to challenge American dominance…
Huang built a multibillion-dollar business to feed the ravenous appetite of Chinese customers for American chips, cultivating a network of Asian partners who could serve the Chinese far from Washington. He made a friend of President Trump and lobbied Washington hard to keep the backdoor to China open.
China jumped to take advantage of the avenues Trump didn’t close. Its AI developers exploited the Nvidia chips they desperately needed to catch up with America’s leaders, either using smuggled chips or accessing them remotely in places like Malaysia.
Let us stop here. Nvidia was both enabling violations of export restrictions and using permitted loopholes. Many Jensen defenders will highlight the latter and ignore that smuggling was also taking place. And that’s before getting to the fact, as we will see soon, that China hawks appear to be winning the war and will succeed in restricting now-permitted indirect sales to China.
Back to the Journal:
Nvidia reported $12.4 billion in China revenue in the six months through July 26 of this year, or 7% of its global revenue, and analysts say that countless more billions of dollars ultimately trace back to Chinese customers. By the estimate of Culper Research, the short-selling firm that flagged Nvidia’s China issue in May, more than 20% of Nvidia’s revenues last fiscal year were driven by China…
Huang’s downfall has set off a groundswell in Congress to halt the workarounds that allowed him to get Chinese business…
Records seen by The Wall Street Journal show Megaspeed—the company of the former Nvidia CEO’s onetime mistress, Alice Huang—arranged to send nearly $12 billion worth of top-of-the-line Nvidia chips to companies in China between March 2024 and June 2025…
The shipment records show orders for several hundred thousand chips, the very ones that the U.S. wanted most to keep out of Chinese hands through export controls, though the records don’t show how much of the hardware was actually delivered or where it originated.
Ed Zitron, who watches this beat like a hawk, was gobsmacked by the $12 billion in Megaspeed orders in a bit over a year:
Holy shit. Alice Huang alone sent $12bn of chips to China. Jensen did more than cheat on his wife (and also his mistress), this is a massive illegal smuggling ring with direct financial and carnal benefits for NVIDIA’s CEO
— Ed Zitron (@edzitron) November 15, 2026
Back to the Journal:
Then-Commerce Secretary Gina Raimondo crystallized the U.S. policy in December 2023: “We cannot let China get these chips, period,” she said. “We are going to deny them our most cutting-edge technology.”
Chinese companies quickly figured out the best two routes—smuggling and remote access—to sidestep that policy. By physically bringing Nvidia chips into China, Chinese AI developers keep their computing power nearby, making it easier to test, train and fine-tune powerful AI models, engineers say.
Confirmation of the value to key China AI players of getting Nvidia chips:
DeepSeek attracted investment from companies promising access to smuggled Nvidia chips:
Companies linked to Alice Huang through her family and past business connections invested in DeepSeek’s 105.5 billion yuan funding round, which raised $15.8 billion and valued the AI firm at ~$92 billion.
DeepSeek employees reportedly introduced Century Litong and its parent company, Lettall, to investors as major suppliers of black-market Nvidia AI chips, arguing that they deserved an allocation in the heavily contested funding round.
The investments were reportedly intended to build closer relationships with DeepSeek and help the companies secure a steady supply of restricted Nvidia Blackwell chips.
During DeepSeek’s second funding round, held between July and October, Lettall invested 95 million yuan, ~$14 million, through an investment vehicle established by Monolith Management.
Separately, Century Litong reportedly imported 240,000 Nvidia chips worth at least $7.5 billion into China since 2023 and continued securing Blackwell chips despite tighter US export controls.
US export controls reportedly did not stop Nvidia chip smuggling into China.
— Global Markets Investor (@GlobalMktObserv) November 15, 2026
Continuing with the Journal:
A case in March gives a glimpse of smuggling tactics. The U.S. alleges that a co-founder of Super Micro Computer, a longtime Nvidia partner, and others participated in a conspiracy to ship servers to a Southeast Asian company, then onward to China.
When inspectors from Super Micro asked to pay a visit to the Southeast Asian company’s facilities to ensure the servers it ordered were in its possession, the alleged conspirators had a problem, according to a federal indictment: The servers had already been sent to China.
To pass their audits, the conspirators allegedly arranged to build a Potemkin server warehouse that supposedly contained Nvidia chips. To fool a visiting U.S. government inspector, a hair dryer was used to remove and affix serial number stickers, according to the indictment, unsealed in March in New York federal court….
Data centers outside China can freely import Nvidia chips and rent the computing power to Chinese customers. In May 2025, the Trump administration opened the floodgates to the tactic by revoking a Biden administration plan to limit how many chips countries in Southeast Asia and elsewhere could import. The administration said it would put in place alternative curbs—but 18 months later, it has yet to do so.
To build the most advanced AI models and chase the U.S., Chinese developers say they still need Nvidia’s top-of-the-line chips. When using the chips in overseas data centers, they sometimes carry physical hard drives with the training data back to China from Southeast Asia. The data centers can also serve customers outside of China who use Chinese-developed AI….
Data center companies in Southeast Asia found that once they landed Nvidia chips, Chinese demand was insatiable…
Chai, the Singapore-based researcher, said in a recent report that Malaysia has attracted more than $34 billion in data-center investments, mostly from U.S. and Chinese firms…..
Momentum in Washington is growing to crack down on the China workarounds.
A new story in The Hill gives a sense of how the various key players are lining up. From Bessent, Congress zero in on Lutnick’s role in Nvidia-Megaspeed scandal:
Commerce Secretary Howard Lutnick is facing heat in Washington, including from Treasury Secretary Scott Bessent, over his approach to AI chip export controls amid Nvidia’s national security scandal.
Reports emerged this week that Lutnick took a softer line on Nvidia over the tech giant’s AI chip sales to China and failed to put the Singaporean firm Megaspeed on the trade entity blacklist….
“Treasury, War, USTR [U.S. Trade Representative] and State … were in favor of putting Megaspeed on the entity list for months,” Bessent said Friday on CNBC’s “Squawk Box.”
“But Commerce has the pen. Commerce BIS [Bureau of Industry and Security] has the pen, and you will have to ask them about their internal processes,” Bessent added.
Lutnick is set to appear before Congress. Get your popcorn ready.
Nikkei Asia describes how China is reading the handwriting on the wall in China sees Nvidia chief Huang’s swift downfall as political hit. The subhead: Commentators say a spying smear campaign aims to purge US tech of Chinese influence. From the article proper:
The strong pushback mirrors deepening mistrust between Washington and Beijing over technology and national security issues. While Nvidia’s internal probe was triggered by reports that Huang had been targeted by a foreign intelligence effort, many voices in China have framed the episode as a pretext to sideline ethnic Chinese tech leaders and build political momentum to tighten export controls aimed at slowing China’s AI development.
The Lutnick/Trump versus State, War, Treasury turf war over Nvidia and sales to China is still in play. From the Wall Street Journal in Trump Administration to Name Nvidia Among AI ‘National Champions’. Note this is a Commerce Department, as in Lutnick, initiative:
Nvidia will be one of more than two dozen companies designated artificial intelligence “national champions” as part of an effort by the Trump administration to help AI firms win business overseas…
The announcement, which is expected within the next week, indicates the administration is barreling ahead with its tech strategy despite the turmoil surrounding Nvidia. The national champions initiative is part of an effort to get foreign companies and countries to adopt U.S. AI technology over Chinese rivals.
With the “national champion” designation, the companies can receive expedited approvals for licenses to export chips and other technology, government support for their work overseas and access to federal financing, according to the Commerce Department’s website for the program. The program was initially announced about a year ago, but the national champions hadn’t been named.
We had said that the Jensen ouster would be more deeply destabilizing, and its effects would become visible sooner, than the Hank Greenberg exodus from AIG, which was the ultimate cause of its collapse. AIG might still have blown a $20 billion securities lending hole on Greenberg’s watch, but that would not be enough for it to have failed and pulled so much of the global financial system down with it.
Jensen if anything was more central to Nvidia operations than Greenberg was, if nothing else due to its vastly more dynamic environment, plus its lack of steady cash flow (the joy of insurance). And as many have pointed out, Nvidia has made itself a linchpin AI ecosphere funder.
High profile departures have started:
I’ve decided to leave NVIDIA. There are a few things I want to say.
As a robotics engineer at NVIDIA, I have seen Jensen as much more than the company’s CEO.
I admire his willingness to invest in a future that was not yet obvious. He committed to CUDA long before the deep learning boom and positioned NVIDIA for AI before it became everyone’s priority. Those decisions helped build the infrastructure that powers today’s AI.
I have seen that same vision in physical AI: investing in robotics, sharing open-source tools and projects, and building computing platforms that bring AI into the physical world. This is the work I do every day—and a future I believe in.
But what matters most to me is the culture he helped build.
Engineers are encouraged to explore, take risks, and learn from mistakes. Jensen reads employees’ Top 5 emails, follows their work, and engages directly. In a company this large, that connection matters. It makes an individual engineer feel that their ideas can still be heard.
That is why this has been so disappointing.
I do not have all the facts surrounding the allegations, and I cannot prejudge the investigation. My admiration for Jensen does not put him above scrutiny.
What I find difficult to accept is having to come to terms with his departure before receiving an explanation that helps me understand it. What this means for our direction, our culture, and the people doing the work remains unclear to me.
I have felt that engineers’ perspectives matter here. On this decision, I have not felt heard. That gap has been more painful than I expected.
I understand that some information cannot be shared immediately. But that does not erase the feeling of being left outside a decision that affects the company we help build every day.
A company’s greatness comes from its leadership, its governance, and the people who turn its ambitions into reality. Employees’ trust and sense of belonging are part of that, too.
Jensen’s vision and the engineering culture he shaped were important reasons I wanted to stay. His departure led me to decide that my own chapter here should also come to a close.
I leave with sadness and gratitude. The colleagues I have worked with, and the work we believed in together, will always matter to me.
These are my personal views.
— Darrell Dai (@darrell_dai) November 15, 2026
It’s a pretty good sign of how out-of-touch the board was when they made the decision to get rid of Jensen. The nearly certain downstream effect is the collapse of NVidia:
1) 60 direct reports
2) New CEO is not going to be able to do 60 direct reports
3) Puts in traditional management structure
4) 80-90% of those reports get layered
5) They don’t like it
6) They are all getting recruited hard by competitors
7) Most of them will leave (or retire, lol)
8) Their reports will also leave
9) Optional: replace them with second-tier players
10) Company collapses
If they are lucky, they will get a Marissa Mayer who comes in and engineers a $4B sale to Verizon for the remaining IP.
— Yishan (@yishan) November 14, 2026
While the “60 direct reports” shows why no one can simply be slotted in, the underlying issue is even bigger than that. Does anyone but Huang know how top level decisions were made, as in what information he relied on (management information systems, external data, intel from outside contacts, reports from or solicitation of input from staff) and if and how he sanity-checked before going ahead? Formal reporting structures are just one, albeit important, facet of how Nvidia operated.
Jensen and his allies seek his return. The lead story at the Financial Times:
From the article:
Jensen Huang is considering a comeback as Nvidia’s chief executive, as allies in Silicon Valley pressure the company’s board to reinstate him days after the tech executive resigned over reports he had affairs with a suspected Chinese spy and a subordinate…
One Huang ally told the FT they were putting together a “war room” of Silicon Valley executives to lobby for the former Nvidia chief’s return and would try to garner support from the White House.
“Everyone that matters in tech outside the frontier labs loves Jensen,” they said. The efforts had Huang’s support, they added.
A public campaign to reinstate Huang has gathered steam over the weekend. Tesla and SpaceX chief Elon Musk and Epic Games chief executive Tim Sweeney were among executives backing a public letter demanding that its board “Bring Jensen Back”.
Huang continues to command fierce loyalty from some employees, several of whom have gone public in calling for his return in the days since he resigned and have signed the open letter.
But the board is at risk from its past blind support of Huang. Yale professors By Jeffrey Sonnenfeld
and Steven Tian explain why in Fortune in Nvidia board’s idolatry of Jensen Huang could open the door to shareholder lawsuits in Delaware Chancery Court. This is a must-read article, and the big footnote below only skims the surface of the arguments. If Sonnenfeld and Tian are correct, it will be hard for the board to reverse its decision on Huang.
And as mentioned above, suits have been filed. A Financial Times story that is mainly about the lack of Nvidia directors’ and officers’ briefly mentions a suit seeking class action certification. From Nvidia exposed to possible lawsuits after Jensen Huang’s downfall:
While Nvidia has in place up to $900mn in coverage protecting the personal assets of its executives and board members, it lacked directors’ and officers’ (D&O) liability insurance covering the company itself against damages and legal costs stemming from the conduct of its executives, according to two people familiar with its policy….
Nvidia has faced at least one such case, with shareholders this week suing the company — as well as Huang — in a proposed class action accusing them of making false and misleading statements regarding the founder’s relationship to Alice Huang, the alleged Chinese spy.
As we said at the top, the sex and spy scandals are a useful way for the board to act as if the forced exit was due to personal conduct, which conveniently shields the board and remaining executives, and not chip-smuggling and potentially financial and accounting abuses. Keep in mind also that even though Trump still backs Huang, he’s already bleeding out power as a result of the midterms. His position vis-a-via Nvidia has weakened further thanks to a breaking Washington Post article, Trump’s stake in Nvidia grew to as much as $62 million before scandal broke (no archived version yet). Huang’s other big Administration ally, Lutnick, is compromised by virtue of how Cantor Fitzgerald, now run by his sons, has what CNBC politely calls “extensive commercial ties” to Nvidia.
Another reason the China hawks may remain ascendant on this issue is that geopolitically, this is the least bad US-dominance-preservation strategy available. The realists among the imperialists ought to recognize that the US has badly overextended its military. It needs to exit Middle East adventurism and Russia-baiting in order to rebuild. Touting China as a threat will still keep money flowing to arms-makers even if the US dials down its active commitments.
In addition, despite the media hype about Nvidia being an invincible colossus, its competitive standing has eroded considerably in the last year, making it look like the decay was a significant spur for the China sales push. From a must-read JP Morgan report, online at https://assets.jpmprivatebank.com/content/dam/jpm-pb-aem/global/en/documents/eotm/the-icarus-files.pdf. The key section:
Because this post is already long, so as to not unduly tax reader attention, we will give only a very very short intro to the two possible financial frauds mentioned: the cluster of abuses colloquially called channel stuffing, and some booked revenues falling short of “true sales” due to the extent and terms of Nvidia financing.
Ed Zitron has found evidence of channel stuffing but because Zitron can do only so many thing, and fighting AI touts eats a lot of his time, he has incorrectly assumed that channel stuffing, even though not exactly kosher-seeming, is legally permitted. It isn’t, although it is typically prosecuted (as in criminally!) as securities and mail fraud.
Zitron’s smoking gun is that he has ascertained that Microsoft had bought Nvidia chips way way way in excess of its use, as in it has a ton warehoused. There are lots of reasons why this is very bad, the biggest being that these chips get stale. They are soon surpassed by newer ones that run faster. Given the AI arms race, the most recent ones would be deployed as soon as an industry incumbent could get their hands on them. The older chips might never be used and should at least have their value written down when newer models roll out.
The question regarding channel stuffing is whether the Microsofts and Amazons and Metas of the world so vastly overbought due to FOMO, as in falling victim of their own hype machines, or whether they over-bought due to threats by Huang. And even if the biggest boys could have or could be expected to be able to stand up to Huang, one can imagine that smaller fry would be vulnerable to threats, particularly if Nvidia also offered generous financing as a carrot.
Zitron discussed his Microsoft find a bit over a month ago on Tech Report; he went into much more detail in a September article, Where’re All The AI Chips? We expect to mine these later but must leave you with only a teaser for now.
Here is the cleaned-up part from a machine transcript where Zitron speculates (based no doubt with some supporting industry gossip) that Huang might been able to muscle even the likes of Microsoft:
Nobody on God’s green Earth is sitting here wondering if Microsoft added gigawatts of CPU capacity or cloud storage, nor are analysts desperate to hear about aggregate numbers — they’re asking where is all the money you’re spending on AI going, and the answer, it appears, is “into a warehouse” or “into a data center without power.”…
It appears that Microsoft is being vague with its numbers to hide the very obvious truth: that it’s spending hundreds of billions of dollars to buy GPUs that sit in warehouses or unpowered data centers, likely years in advance….
Significant for Microsoft, but also for the rest of the world and also begs a few really worrying questions. So, Microsoft was the first company, I believe, to build an AI supercomputer…
Now, they have the most experience. They have the most logistical experience. They have the most money to do this. Amazon technically has done more cloud for longer, but Microsoft is way more in AI.
And Microsoft has what looks like less than half of their purchase GPUs installed.
This means that I believe personally that this is an across the industry phenomena. And if that’s the case, we likely have hundreds of billions of dollars of Nvidia GPUs in warehouses waiting to be installed. Which begs the question, why are people still buying them? And how is Jensen Hang doing it?
Now, big boy Jensen Hang, CEO of Nvidia, might just be fending them. He might just be saying, “You pig, if you don’t buy these today, I won’t sell you the next ones.” But this is fundamentally insane.
This means that GPU capacity is coming on extremely slowly and I believe personally that there is an industry-wide attempt to lie about available capacity….
Is Nvidia running the largest Kickstarter of all time? Is Nvidia just selling GPUs years in advance? How many GPUs have come online? Because if it turns out that everyone is like this, everyone is just hoarding GPUs, that is actually a that should be an SEC investigation because put it like this, I don’t think it’s illegal. But it is something that I think shareholders need to know because there’s a big difference between Nvidia is selling a bunch of GPUs because of demand for data centers that getting built and Nvidia is selling a bunch of GPUs because everyone’s worried that if they don’t buy them now, they won’t be available in the future. And I must be clear, Microsoft is so experienced at doing this. If Microsoft’s got less than half of their GPUs installed, do you think Meta’s much better? Do you think Amazon? Google? I don’t think so.
Let us be clear. This is illegal. It’s securities fraud. That does not mean it could be proven without some heavy lifting. But coercing customers to buy way ahead of their needs and booking the resulting sales as revenue in the current period is a fraud.
And again, there are many players besides these tech titans, and they would have been even less able to stand up to pressure from Jensen. Smaller fry are awfully dependent on Nvidia largesse:
“Some founders and venture-capital backers of companies with little to no sales said that if Nvidia did pull back from future equity investments, they could struggle to find capital to keep their cash-hungry businesses going, let alone secure a lucrative exit.”
It’s joever
— Ed Zitron (@edzitron) November 16, 2026
We will stop here for now. We are sure to return to these topics in the coming days.
____
1 It is to Nvidia’s advantage to depict the Huang ouster as a result of sex scandals (oh, big man felled by Puritans, so nothing else to see here) as opposed to chip smuggling and rising concerns about so-called circular financing; this is akin to Harvard depicting the faculty no-confidence vote against president Larry Summers as the result of his derogatory comments about women’s academic potential. In fact, the morning of the vote against Summers, every faculty member had had an explosive, extensively documented Institutional Investor article, How Harvard Lost Russia, stuffed into their inboxes. Harvard had paid over $30 million in fines in a settlement with the federal government yet (among other things) Summers had continued to defend one of the key grifters, Harvard professor Andrei Shleiffer.
2 Nvidia is more than rich enough to self-insure executives, but the lack of sufficiently comprehensive policies points to at best sloppy management and at worst, a lack of willingness to undergo adequate underwriting. I am not a lawyer but the insufficient D&O coverage may also make it easier to pursue derivative shareholder suits, where lawyers and shareholders argue that the company execs and board cannot or have persistently failed to execute their duties as stewards of the best interest of the company, and they step in to assert those powers.
3 A key test is bankruptcy remoteness, as in if the securitization trust fails, can the lender somehow lay claim to any of the loans it made? The analogy here would be if a Nvidia customer where Nvidia financed the sale failed, was the financing structured so that Nvidia claw back the chips? The securitization industry was so keen to persuade mortgage securitization investors that no way, no how could that happen that the typical securitization structure had the loans be transferred in and out of not just one but two intermediary entities before getting to the securitization trust.
4 A key section from the article:
….there were some blatant red flags in Huang’s recent leadership that were apparently missed by the Nvidia board.
In October 2025, the New York Times reported that Megaspeed, a little-known Singapore firm, had bought billions of dollars of Nvidia chips amid suspicion that it was helping China evade U.S. restrictions by laundering banned chips on the U.S. export restriction list to China via neutral countries. Nvidia said a review found no evidence its chips were diverted, but that reporting was followed by an explosive report by a short seller, Culper Research, alleging “uncomfortably close ties” between Alice Huang, the head of Megaspeed, and Jensen Huang, in May 2026. Jensen described the Megaspeed dealings as normal, and apparently the Nvidia board was content to leave it at that.
The fiduciary duty no genius nor idolatry can get around
The Caremark doctrine, established in 1996 in Delaware Chancery Court, establishes that corporate directors have a fiduciary duty of oversight, and that directors breach their duty of loyalty when they make no good-faith effort to monitor the risks central to the business, or when they consciously ignore red flags.
Under this doctrine, and derivative variants of it, several boards have run into trouble in Delaware Chancery Court for similar sins of omission when it comes to oversight. In 2019, Delaware’s Supreme Court revived a suit against the directors of Blue Bell Creameries, an ice cream maker whose board had no system for overseeing food safety before a deadly listeria outbreak. In 2021, the Court of Chancery let claims proceed against Boeing’s directors, who had no board committee responsible for airplane safety before the 737 MAX crashes. That case settled for $237.5 million. The lesson of both: on risks that can sink the company, the board must watch for itself: the board cannot simply trust management, even when those risks were unforeseeable or unanticipated.
The risks Nvidia directors should have been asking questions about
So plaintiffs’ lawyers will ask the questions directors should have asked many months ago. Who conducted the review that cleared Megaspeed? Did it report to independent directors or to management? Did any director know the CEO had a personal relationship with an executive of the customer under review? Did the board ever discuss Megaspeed at all, or Jensen Huang’s relationship with Alice Huang? Did any board members even care to ask any questions about these topics?
The answers are in the board meeting minutes, and alarmingly for Nvidia’s board, shareholders can now get those minutes with nearly complete transparency. Nvidia knows this better than most companies for a very ironic reason. In 2022, it fought a books-and-records demand all the way to Delaware’s Supreme Court. Even after Delaware narrowed those inspection rights last year, board minutes and board materials remain within reach for plaintiffs litigation. That is how the Boeing plaintiffs built their case.
