Jensen Huang Resignation Gives Nvidia Its First Board Chair in 33 Years

The Jensen Huang resignation announced by Nvidia on 12 November ends a 33-year arrangement in which one founder ran the company whose chips power much of today’s artificial intelligence (AI), while the board around him never had a chair. Huang stepped down as president, chief executive and director with immediate effect, and the board used the same announcement to create the role it had gone without since 1993.

Chief financial officer Colette Kress becomes interim chief executive while keeping her finance job, and Harvey C. Jones becomes the first chair in Nvidia’s history. Splitting Huang’s authority across three people is the board’s answer to a question that customers, investors and governments had little reason to ask while the founder was in place, which is how much of Nvidia depended on his judgement alone. TBM looked at the gap between a public exit and its terms when it covered an announced resignation at Goldman Sachs, and the same questions apply here.

What Nvidia’s announcement says and what it leaves out

The board’s announcement, filed with the Securities and Exchange Commission (SEC), says Kress joined Nvidia as finance chief in September 2013 after nearly 25 years in finance roles at technology companies including Cisco and Microsoft. The board is searching for a permanent chief executive among internal and external candidates, and it gives no timeline for that search. Jones is described as an entrepreneur and venture investor who previously ran Synopsys as chief executive and executive chair.

The document is brief on several points that matter to shareholders, since it contains no separation terms for Huang and makes no reference to the investigation that preceded his exit. It confirms that Nvidia will report its third-quarter results on 17 November as previously scheduled, which means Kress will face analysts five days into the role. Tom’s Hardware reported that Nvidia shares fell more than 10% in early trading after the news.

How the allegations reached Nvidia’s board

The sequence began with a New York Times report that a classified memo, circulated among senior Trump administration officials earlier this year, cited intelligence indicating that China’s Ministry of State Security had recruited Alice Huang, an executive at the technology company Megaspeed, to gather information on the Nvidia founder. The Times described the memo through unnamed sources who had been told of US surveillance of Alice Huang, and no outlet has reported seeing the document itself, so its contents and the evidence behind it remain undisclosed. China’s foreign ministry called the allegation unsubstantiated.

Three Republican committee chairs in the House of Representatives sent Huang a letter requesting a joint closed-door briefing, made public on 5 November, on his interactions with Alice Huang and on Nvidia’s protections against foreign espionage. The chairs proposed 18 November for the briefing, the day after Nvidia’s results. The letter describes her as a publicly reported Chinese intelligence asset, which is the committees’ characterisation of an allegation that no court or public document has yet tested.

Huang published a message to employees on 8 November in which he admitted being unfaithful to his wife, took full responsibility, and said he supported an independent review. His message mentioned the espionage reports only to say that the board had opened an investigation, and it did not respond to their substance.

The board’s statement the same day opened that review under a committee of independent directors and said the board had no basis to believe national security or confidential company information had been compromised. The board also expressed confidence in Huang at that point, a position it held for four days.

The independent directors then retained outside counsel on 10 November, choosing Joon H. Kim, previously acting US Attorney for the Southern District of New York, and the law firm Wachtell, Lipton, Rosen & Katz to run the investigation. Huang’s resignation came two days after that appointment, before the investigation had published any findings.

Why Nvidia had a lead director instead of a chair

Nvidia’s 2026 proxy statement explains that the board chose an independent Lead Director rather than a chairperson, on the grounds that this gave every director an equal voice. The Lead Director, Stephen C. Neal, who joined the board in 2019, prepares board agendas together with the chief executive, while also leading quarterly sessions of independent directors without him. Oversight therefore existed, and this week’s events show that it acted once the independent directors decided to use it.

The arrangement placed the founder at the centre of the board’s work, since the person being overseen helped decide what the overseers discussed. The proxy also shows that Jones, the new chair, has been a director since 1993, which makes him one of the people who served alongside Huang from the company’s founding year. TBM found a related pattern of founder authority when it analysed founders’ control in Anthropic’s listing documents, where seven co-founders hold a majority of voting power through a single Class F share.

What the split means for customers and investors

Nvidia’s role has grown over the past year to include financing, which is why its leadership matters well beyond its own shareholders. TBM reported in August that Nvidia agreed to guarantee payments on an Ohio data centre campus if its tenant defaults, an OpenAI backstop that makes the company a financial counterparty as well as a vendor. The tenant’s lease runs for 20 years, which ties the value of Nvidia’s guarantee to the company’s stability across that whole period.

Governments carry a similar exposure, since Nvidia’s UK skills collaboration with the Department for Science, Innovation and Technology places its training courses inside government-funded apprenticeships in England. That public programme now continues under an interim chief executive and a chair appointed this week.

Businesses testing Nvidia’s open models for workplace AI rely on the company’s product roadmap continuing on schedule, which is the commitment the new chair made in the announcement.

Large customers have also been building alternatives, as Microsoft did when it said its Azure Maia chips would become the main system for running AI in its own products and services from 2027, so the largest buyers already have their own alternatives in development.

Retail investors hold part of this exposure too, sometimes through apps rather than traditional brokers. Binance began offering tokenised fractional Nvidia shares from $5 to its registered users outside the US, so the governance of one US company now reaches people far from Wall Street.

Those holders sit at the end of a chain that runs through fintech platforms, brokers and index funds back to a single board. That board gained a chair able to lead it independently of management only this week.

A board that now answers for itself

Nvidia’s board said operations, product plans and customer commitments will continue without interruption, and the authority Huang held since 1993 now sits with three people. The board’s first decision without him was to create the position that would have stood between him and the company, which shows that Nvidia itself no longer treats one person’s judgement as a sufficient safeguard for a business this many others depend on.

Editor’s note
Every piece goes through careful verification, but mistakes can happen. Readers who spot an error or have additional information can write to rosalia@thebrightminded.com.