There is no doubt that Nvidia founder and CEO Jensen Huang is a visionary genius, whose vision as the ‘Godfather of AI’ has improved billions of lives around the world and contributed greatly to American prosperity and innovation. He has attended our Yale CEO forums and has been revered by CEOs across sectors and around the world for his bold, entrepreneurial vision and his revolutionary technological triumphs, plus he is a showcased contributor to our new, upcoming book, When Machines Act: The Promise and Perils of our Agentic AI Future, published by MIT Press and written with Two Sigma Co-Founder David Siegel.
Yet, amid calls for Huang’s immediate resignation from top GOP voices, including Sen. Kevin Cramer of the Senate Banking Committee and others, and demands that he answer to Congress, 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.
Late last month, the Times reported that U.S. intelligence believed that Alice Huang was a Chinese asset. A company spokesperson said the diversion claims had been debunked and described the relationship as nothing unusual. Ten days later, Jensen admitted he had been unfaithful to his wife, in a message that referred to the national security concerns in the press – a troubling sequence of events after all the repeated denials.
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.
All of the major risks that have now landed on Nvidia in such dramatic fashion were exactly the risks directors should have been most alert to.
First, and most importantly, every board entrusted with protecting minority shareholders should be alert to management conflicts of interest. A Megaspeed affiliate bought about $4.8 billion of Nvidia-powered equipment in under two years. A chief executive in a personal relationship with an executive of a customer that size has an apparent conflict that Nvidia’s own code of conduct would have likely required him to disclose. And Megaspeed was no ordinary customer: by October 2025 both Washington and Singapore were investigating it.
That conflict poisons the compliance record. When the questions first surfaced, Nvidia said it had visited Megaspeed’s sites and had no reason to believe chips were diverted. That assurance came from people who report to a CEO with a personal tie to the customer being cleared, a tie no investor had been told about. Bloomberg reports that Nvidia’s general counsel went so far as to contact a senior Commerce Department official about the government’s Megaspeed probe. Did the board authorize that call? Did it even know?
Then comes national security and export-control evasion, as “mission critical” a risk as food safety at an ice cream maker or flight safety at Boeing. Nvidia makes the most strategically sensitive product on earth; and The Information reported today that companies tied to Alice Huang helped Chinese buyers obtain $7.5 billion of Nvidia chips, including banned ones that were laundered through neutral parties. That report is unconfirmed, and Megaspeed denies wrongdoing. But the penalties for export violations run from fines to criminal prosecution to the loss of export privileges, and a board should not have been waiting for verdicts before it starts asking questions. According to Nvidia’s own proxy statement, the board committee responsible for trade compliance met three times last fiscal year, and the company has no chief compliance officer. How much did they know, and when?
The national security risk is larger still. According to the Times, U.S. officials concluded that Alice Huang was recruited by China’s Ministry of State Security to get close to Jensen and to shape his views on chip sales to China – which of course, Jensen Huang has been very vocal on in Washington and in the media, arguing that the U.S. should continue to sell chips to China so that China is not building their own chips. No one has alleged that Jensen knew Huang was a spy, and Beijing calls the claims a smear. But the president was reportedly briefed in late 2025, and a secret memo circulated among policymakers this spring, the same year Jensen was appointed to the President’s Council of Advisors on Science and Technology. Now three House committee chairmen want him in a closed-door briefing, and the president says the FBI is reviewing the matter. Yet not one Nvidia director, as Semafor noted, has a national security background; what questions were asked and what prudent diligence steps were taken by the board?
Then there is the risk every board dreads: that management did not tell it the truth. On Oct. 29, a company spokesman described Jensen’s interactions with Alice Huang as “not unusual or unexpected,” the routine customer contact of any CEO. The same week, Jensen called the Megaspeed dealings normal. Nine days after that he told employees he had been unfaithful to his wife, and tied the admission to the national security concerns in the press. If directors approved those October statements knowing what Jensen knew, they could be in trouble. If they never asked him, that is the Caremark problem.
Shareholders have already paid. Nvidia closed today at $168.89, down almost 10% in the two sessions since the admission, a loss of more than $400 billion in market value, and about 30% below its early-October record. Nobody can attribute this recent fall merely to just the AI sell-off, as the evidence shows it was an idiosyncratic move. On Monday, as Nvidia fell 4.3%, the semiconductor index jumped nearly 7%, and direct competitors all rose, including AMD.
The last risk is one the board manufactured itself. Jensen has more than 50 direct reports and no identified successor, and one prominent analyst estimates Nvidia would be worth 25% to 30% less without him. That reflects poorly on the board. Succession is a director’s first duty, and these directors let one man become the company with no obvious succession plan until it is too late.
Even now, the reflex appears to be deference. The same statement that launched the board’s investigation declared, before a single interview, that the board had “no reason to believe that either national security or Nvidia secrets were compromised,” and that it had confidence in Jensen. A board that announces its conclusion on the day it opens the inquiry, makes one wonder whether the board has already predetermined what their preferred outcome would be rather than following the trail of wherever facts lead them in a way that protects all shareholders.
No dual-class shares, and no shortage of entrenchment
Readers of Fortune will know that we have been critics of Vice Chancellor Travis Laster on the Delaware Court of Chancery, who wrongly attacked us in his Cloudflare opinion as “advocates for dual class shares” and for founder control more broadly. Vice Chancellor Laster did not apparently understand that we are not advocates for anything; rather, we call it as we see it as fact-based academics with five decades’ experience as a widely recognized scholar of corporate governance.
We have published extensively on dual-class shares, on when they work, and when they don’t, with extensive academic analysis and quantitative research into performance of companies with dual class share structures.
In our extensive careful research, published widely in academic outlets such as the Harvard Law School Forum on Corporate Governance as well as FORTUNE, Chief Executive Magazine and many other outlets, we have studied instances when dual class shares work well, such as when a founder or family with real money at stake uses the protection to invest through cycles and outlast short-term activists, as at Berkshire Hathaway, Alphabet and The New York Times.
Similarly, we have studied instances when dual class shares should be retired, such as when they shield controllers with ulterior motives from consequence, as at WeWork under Adam Neumann or Paramount under the Redstones.
Critics of founder control tend to treat dual-class shares as the bogeyman behind every entrenched founder, and one share, one vote as the cure. Nvidia is the case study that breaks that outdated, prosaic theory. Nvidia has a single class of common stock, not dual class shares. Jensen owns about 3.5% of it and casts 3.5% of the votes. The board is declassified, every director stands for election every year, and a simple majority of directors could replace him as CEO at any meeting. On paper he is as removable as any hired manager. In practice, no CEO in America has been more immune from proper scrutiny and oversight.
What entrenched Jensen was not the capital structure of the company; rather, it was the apparent idolatry of a board which was in thrall socially and interpersonally to Jensen Huang. He earned the adulation honestly, building a company from a booth at Denny’s into, for a time, the most valuable on earth. But the directors whose job it was to provide effective oversight to guard shareholders’ interests, apparently effectively joined the congregation. No charter provision did that, and no proxy adviser would have caught it. Nvidia has the independent majority, the lead director, the annual elections and the single share class that the checklists of ISS and Glass Lewis reward.
Nor has one share, one vote ever been a vaccine against scandal. Enron, Tyco, Wells Fargo and Boeing all had it. One of us argued in the Harvard Business Review in 2002, after Enron, that failed boards routinely passed every structural test, and that what set great boards apart was a culture of open dissent, in which directors challenge the chief executive and each other. That was true then and Nvidia shows it is true now.
When it comes to providing effective oversight, what matters is the oversight culture of the board and the quality of the board, not the ‘check the boxes’ yardsticks so often misleading propagated by governistas, such as ‘one share one vote’.
How Nvidia’s board could be exposed to potential liability
Delaware needs no new invented legal theory for measuring whether Nvidia’s board did its duty as a board in providing oversight. If shareholders bring Caremark claims against Nvidia’s directors, the court will ask what the board knew and what it did to find out. We have called out the Delaware Court of Chancery when it overreaches, and we will be again. But failing to question a celebrated CEO about a customer under federal investigation is no business judgment at all. Nvidia’s directors were paid to provide oversight over Jensen Huang, however brilliant. And in that duty, by all outside appearances, they apparently whiffed.
Aristotle famously warned, “There is no great genius without a touch of madness.” Psychiatrists call shared disorders “folie à deux.” It is up to the board to ensure a company’s owners benefit from the genius of founders but are protected by the madness.
The opinions expressed in Fortune.com commentary pieces are solely the views of their authors and do not necessarily reflect the opinions and beliefs of Fortune.
This story was originally featured on Fortune.com