‹ Field Notes
Disagreeing upward

Nvidia said its CEO's interactions with an alleged spy recruit were 'not unusual'. The question that tests disagreeing upward.

On November 7, 2026, TechCrunch reported that a memo circulated earlier this year claimed technology executive Alice Huang was recruited by China's Ministry of State Security to spy on Nvidia CEO Jensen Huang, according to multiple media reports. The New York Times was the first to report on the memo, which TechCrunch said was reportedly based on U.S. intelligence, and the Financial Times subsequently reported that the memo also claimed Jensen Huang had a romantic relationship with Alice Huang, an executive at a Singapore-based company called Megaspeed. Every link in that chain is a claim rather than a finding, and the company's answer was short. An Nvidia spokesperson told the NYT that Jensen Huang's interactions with Alice Huang were "not unusual," and that the company does not share classified information with its customers. TechCrunch also noted a separate NYT report on New Matrix, a Chinese company authorized by the U.S. Commerce Department to buy advanced Nvidia chips. At the office address listed by New Matrix, reporters found only "a hollow gray shell," raising concerns that it could be a front for a blacklisted private equity firm.

Whether any of it holds up is not the point here, and the reporting hedges at every step. Look at the shape of the moment instead. A claim arrives about the person at the top of an organization. It is unproven and uncomfortable, and the official position is that nothing is out of the ordinary. Everyone below that position now has to decide, mostly in private, whether they agree. Most will. A few will not, and those people face a problem that has nothing to do with espionage: they hold a view that differs from leadership's, on a matter where leadership has more information, more authority, and more to lose. The distance between the official account and the independent view is the whole story, and it is also the behaviour a hiring manager is reaching for with one of the most frequently reported questions in senior interviews: "Tell me about a time you disagreed with leadership."

The interview question
“Tell me about a time you disagreed with leadership.”

Why they ask it

The question is not a test of courage. Any manager at the level where this question gets asked has disagreed with a decision from above; the interviewer wants to know what happened next. Three things are being measured. First, whether the candidate forms views independently of the room, rather than discovering their opinion after the senior person has spoken. Second, whether they can express that view in a form leadership can actually act on: specific, evidenced, and offered before the decision hardens rather than after. Third, and most important, what they do once the decision goes against them. Companies are full of people who disagree and then comply resentfully, or disagree and then quietly route around the decision. Both are more expensive than someone who never disagrees at all, because they corrode trust while looking like alignment.

The trap

There are two ways this answer usually fails. The first is the vindication story: the candidate saw the problem, leadership ignored them, the problem happened, and everyone learned a lesson. It is flattering, it is unverifiable, and it tells the interviewer that the candidate keeps score. The second is the diplomatic non-answer, in which the candidate raised some concerns, there was a healthy discussion, and the team aligned. Nothing was disagreed with and nothing was risked. A subtler failure is choosing a disagreement about taste, such as a naming convention or a slide order, where nothing material was at stake. A strong answer picks a disagreement where the leader had a legitimate case, names precisely what the candidate believed differently and on what evidence, describes how the view was delivered, and is honest about the outcome, including the possibility that the candidate was wrong.

Applying STAR-T

Situation. Keep it to a few sentences and make the leader's position reasonable. For example: a vice president wanted a rebuilt billing service switched on for every customer at once, so the launch could anchor a quarter-end announcement. The candidate ran the team that owned the migration and had sat in the planning meetings where the date was set.

Task. State the obligation and the conflict in one breath. The candidate was accountable for delivering on the date, and believed the reconciliation job, which had never run at production volume, made an all-at-once cutover an unacceptable risk to customer invoices.

Action. This is where the answer is won or lost, and it should be mostly about how the disagreement was delivered. The candidate wrote a short memo naming the specific failure they feared, the dry-run result that supported the fear, and an alternative that kept the announcement date by cutting over a small cohort first. They took it to the vice president privately, before the staff meeting, and asked directly what evidence would change their mind. When the vice president held to the original plan, the candidate asked for two things instead: a kill switch that could revert the cutover within minutes, and a dedicated on-call rotation for launch week. Both were granted.

Result. The launch went out on the announced date. The reconciliation job did fail under full load on the second day; the kill switch limited the damage to a few hours of invoices, which were replayed. In the retrospective, the vice president adopted a cohort rollout as the default for future migrations. Say the result plainly, including the part where the candidate did not get what they asked for.

Trade-off. Name the cost. The relationship with the vice president was cooler for a quarter, and the candidate spent credibility they could not spend again soon. Interviewers trust answers that admit disagreement is not free.

The follow-up that breaks weak answers

The follow-up that exposes a rehearsed answer is almost always some form of: and what did you do in the weeks after you were overruled? Weak answers were built to end at the moment of vindication, so they have nothing to say about the interval in which the candidate had lost and still had to lead the team through the plan they argued against. The strong answer has specifics for that interval: how the decision was communicated to the team without editorialising, what the candidate did to make the leader's plan succeed, and whether the disagreement was ever raised again. A second common follow-up is simply whether the candidate was right. The best answers have a case where they were not, and can say so without flinching.

Score your answer against the director’s bar

Q: Tell me about a time you disagreed with leadership.

Ready when you are

Bank this story with the moment you were overruled written down, then rehearse the follow-up until it is the part you are most ready for. Try it free →

Rehearse this in L8 Loop →