Jensen Huang Resigns as Nvidia CEO, and His Own Management Theory Is on Trial

Jensen Huang resigned as Nvidia's CEO this morning after admitting an extramarital affair with an alleged Chinese spy and facing reports of a second affair with one of his own executives. The collapse of the chip industry's most celebrated leader now puts his signature management philosophy under a harsher light than any of his admirers expected: the belief that senior executives need almost no supervision at all.

That belief was not an aside for Huang. It was the stated justification for one of the most unusual organizational structures in corporate America. At Nvidia, between 50 and 60 senior executives reported directly to Huang himself—a span of control that would terrify most management consultants. His argument was simple: senior leaders have already proven they can make decisions, manage people and solve problems without help.

What did Huang actually say about supervision?

He made the case explicitly, and in his own words, at Stanford Graduate School of Business in 2024. "CEOs should have the most reports by definition, because the people that report to the CEO require the least amount of management," Huang said. "It makes no sense to me that CEOs have so few people reporting to them."

He reiterated the position repeatedly in public appearances, making it as much a part of the Nvidia brand as its graphics processors. The stance fits a broader Silicon Valley creed: hire extraordinary people and give them free rein. Google's 20%-time policy, which let employees spend a portion of their working hours on their own ideas, is the classic example of the genre.

The creed has produced genuine innovation. It also, as Huang's exit demonstrates, carries costs that surface only when they surface.

Did anyone at Nvidia hold Huang accountable?

The evidence suggests the answer was no—and that warning signs existed well before this week. Last spring, the proxy advisory firm Institutional Shareholder Services assigned its riskiest rating to Nvidia's board structure. According to The Wall Street Journal, ISS cited a board stacked with multiple long-tenured directors, the absence of a chairman, and no mechanism to refresh the board's membership, among other factors.

That rating now reads as a road map to the current crisis. A board with entrenched members and no independent chairman is a board poorly positioned to challenge a founder-CEO on anything, let alone his private conduct. Senior leaders, by virtue of their power, also simply have fewer colleagues willing to rein them in. Nvidia appears to have been the textbook case.

The company's internal culture compounded the exposure. Reports over the years, including a lengthy New Yorker profile, described Nvidia as an intensely personal workplace in which Huang regularly interrogated junior employees about their work and sent emails late into the night. The line between the personal and the professional, in other words, was already blurred before any affair became public.

What does the resignation cost Nvidia?

Huang's argument that capable executives don't need micromanagement may still hold for business decisions. His downfall shows it fails on a different axis: even the most capable executives can exercise extraordinarily poor personal judgment, and the consequences of misconduct scale with seniority. A mid-level manager's lapse damages a team. A CEO's lapse can damage a company—and beyond.

Nvidia is not an ordinary company. It manufactures the chips crucial to the AI boom and has been central to American leadership in the field. Analysts were already worrying that CEO succession posed a long-term threat to the stock; that question has now moved from hypothetical to immediate, with no apparent succession plan in place. The board, by its own record, failed both to assess the risks of Huang's personal behavior and to prepare for his departure.

The management literature offers little comfort to boards hoping for a checklist. Theorists have long struggled to balance meaningful oversight against executive autonomy, and some Stanford-affiliated research has concluded that no one-size-fits-all approach ensures proper board control. Boards must tailor oversight to the individual CEO and the circumstances rather than mechanically applying standardized rules—a demanding standard that assumes a board is willing to act in the first place.

What is the lesson for other companies?

Huang himself drew the boundary incorrectly, but the boundary exists. Senior executives do not need micromanagement; they still need someone to answer to. Nvidia's structure gave its CEO the least-supervised role in the company at the exact moment when his position carried the greatest capacity for damage—and the board's riskiest-rated governance provided no counterweight.

The lesson for other boards is uncomfortable but plain: the more indispensable and celebrated the leader, the more deliberately someone must be empowered to check him. Nvidia's board is now investigating Huang, but a governance vacuum cannot investigate itself. For a company at the center of the AI economy, the cost of that lesson is only beginning to be tallied.