Some of the world’s most powerful organisations invest millions in cybersecurity, physical protection, background screening and sophisticated security systems.
Yet one of the most consequential vulnerabilities may have nothing to do with technology.
It may be a person who has earned the trust of someone at the very top.
The controversy surrounding NVIDIA and its former CEO Jensen Huang is a reminder of why counterintelligence deserves a place in corporate risk management, particularly where sensitive information, influential relationships and national interests intersect.
What the NVIDIA case tells us
In November 2026, NVIDIA became embroiled in a controversy involving Jensen Huang and his relationship with Alice Huang, a former executive at Megaspeed International.
According to reporting by Reuters and other major news organisations, US intelligence authorities suspected Alice Huang of having links to Chinese intelligence. Those allegations have not been independently verified by Reuters, and China has denied them.
Jensen Huang subsequently resigned as NVIDIA’s CEO amid scrutiny of his personal relationships. NVIDIA’s board initiated an independent investigation into whether he had been targeted by a foreign intelligence effort.
Importantly, NVIDIA stated that it had no reason to believe national security or company secrets had been compromised.
The investigation has not established that espionage succeeded.
But the wider questions are relevant far beyond NVIDIA.
How does an organisation assess risk when access is established through personal trust? Who challenges a relationship when the person involved is exceptionally powerful? And how much can conventional security really tell us about the intentions of someone who appears entirely credible?
Counterintelligence is about people, not just secrets
Counterintelligence is the practice of identifying, assessing and countering attempts to obtain sensitive information, exploit trusted access or influence decisions for unauthorised purposes.
It is often associated with governments and intelligence agencies.
In reality, the underlying principles are equally relevant to multinational businesses, family offices, private wealth environments and high-profile individuals.
Consider a senior executive who develops a close relationship with an influential business associate.
The person has an established company, an impressive professional background and recognisable connections. Meetings are legitimate. There may be no obvious evidence of wrongdoing.
But conventional screening cannot necessarily establish whether someone has undisclosed loyalties, conflicting interests or intentions that extend beyond the stated business relationship.
A clean record is not proof of good intentions. Equally, an unusual relationship is not proof of misconduct.
Good counterintelligence distinguishes between the two.
The limits of traditional security
Cybersecurity teams are trained to identify technical vulnerabilities. Close protection teams manage physical exposure. Compliance departments assess regulatory obligations.
All have important responsibilities.
But who is responsible for understanding the human relationships surrounding the people with the greatest access and influence?
Not everyone who creates a security concern is an employee or formal insider.
An external individual may gain access through a professional association, social connection, romantic relationship or trusted intermediary.
Information may be shared during ordinary conversations. Influence may develop gradually. Decisions may be affected without anyone attempting to enter a restricted building or compromise a computer system.
This is why counterintelligence and insider-risk management need to consider both formal access and informal influence.
Why human intelligence matters
Open-source intelligence (OSINT) can establish a great deal about a person: corporate interests, litigation, professional history, public associations and financial connections.
But information collected from records does not always explain the relationships or behaviour behind them.
Human intelligence (HUMINT), when conducted lawfully and appropriately, can help establish context through informed human enquiries, corroborated sources and an understanding of the environment in which someone operates.
Neither method is sufficient on its own in every case.
The value lies in asking the right questions, testing competing explanations and understanding which findings are reliable enough to support a decision.
Counterintelligence should not be an exercise in suspicion. It should be a disciplined process of assessing exposure before assumptions turn into expensive mistakes.
How FRYA approaches counterintelligence
FRYA International supports organisations, family offices and high-profile individuals facing concerns involving trusted access, sensitive relationships, information leaks and potential manipulation.
Our work may include background and reputation enquiries, mapping relevant relationships and interests, examining potential conflicts, investigating suspected information exposure and assessing human risk.
Where appropriate, we combine open-source research, specialist investigative methods and human intelligence to understand what is happening and who may be involved.
We work alongside existing legal, compliance and security teams, providing independent investigative insight rather than replacing their responsibilities.
Findings are assessed in context, with a clear distinction between established facts, credible concerns and unverified allegations.
And where an investigation establishes a genuine risk, protective or organisational measures can be considered proportionately.
The question worth asking
The NVIDIA case does not prove that every powerful executive needs their personal relationships investigated.
Nor does it prove that more extensive background screening would have prevented the controversy.
What it demonstrates is the potential significance of trust, access and influence when sensitive information and high-level decisions are involved.
Organisations can invest heavily in keeping outsiders out while overlooking the people already close enough to be invited in.
Counterintelligence begins by understanding the difference.
Frequently asked questions
What is counterintelligence in a corporate environment?
- Corporate counterintelligence involves identifying and mitigating attempts to obtain confidential information, exploit trusted relationships, influence decisions or misuse access. It combines investigative methods, human-risk assessment and appropriate security measures.
How is counterintelligence different from cybersecurity?
- Cybersecurity focuses primarily on protecting digital systems and information. Counterintelligence examines the people, relationships, intentions and activities that may create exposure, including situations where no technical system has been breached.
Can a private intelligence firm investigate concerns involving trusted individuals?
- Yes, within applicable legal and professional requirements. The scope of any investigation should be based on a legitimate concern, proportionate methods and a clearly defined purpose.
Editorial note: the account of NVIDIA reflects reporting available at the time of writing, including Reuters and AP. The espionage allegations remain publicly unproven and the independent investigation is ongoing. This article will be reviewed as significant developments occur.