The Oldest Exploit in Intelligence: How a Honeypot Operation Compromised the CEO of a $4 Trillion Chokepoint

Nvidia's Jensen Huang resigned after alleged Chinese espionage via personal relationship, exposing how a single human vulnerability can cascade through a $4tn company, the AI supply chain, and US national security.

Jensen Huang, who built Nvidia into the world's most valuable chip company over 33 years, resigned this week after the New York Times reported he had a close relationship with Alice Huang, a Singapore-based datacenter executive whom US officials suspect of spying for China. He admitted to the affair in a letter to employees, and reports of two additional affairs with Nvidia employees followed. The stock cratered, the Nasdaq fell 4%, and Nvidia lost hundreds of billions in market value.

The alleged operation is textbook honeypot — a method intelligence agencies have used for centuries. Alice Huang ran Megaspeed, a Singapore-based datacenter firm already under US Commerce Department investigation for diverting AI chips to China in violation of export controls. The relationship appears to have facilitated the transfer of billions of dollars in advanced Nvidia chips to China, according to Bloomberg. A classified memo prepared under Secretary of State Marco Rubio reported the relationship based on electronic surveillance of Alice Huang. The scandal exposes a structural vulnerability in how the US tech industry concentrates irreplaceable strategic value in single individuals. Huang was not merely a CEO — he was an informal policy adviser to Trump on export controls and open-source AI, a White House dinner companion, and part of the president's state visit to China. The Center for Strategic and International Studies has documented at least 224 incidents of Chinese espionage in the US between 2000 and 2022, with over half targeting commercial technologies. A former Google engineer was convicted in January of stealing AI secrets for the Chinese government. Corporate intelligence firms say this is not an anomaly but a visible instance of a constant pattern. Don Aviv of Interfor International said his firm alone has conducted roughly a dozen honeypot-related investigations for major companies over the past three years. OpenAI has held training sessions specifically running through honeypot scenarios. Randy Watt of SRW noted the method "speaks to a consistent weakness in humans that is exploitable" and predicted executives across industries are now reviewing their personal relationships. The Trump administration faces its own exposure. Axios reported that Megaspeed was spared from a national security concerns list despite intelligence flagging problems with the company. Commerce Secretary Howard Lutnick is under pressure to resign over perceived inaction. Trump called Huang an "honorable guy" after the scandal broke. China's Ministry of Foreign Affairs denied involvement, calling the reports "nothing but a smear against China." The deeper structural problem is that the entire AI hardware supply chain runs through a single company whose governance was concentrated in a single person for three decades. Nvidia hired a top US law firm to investigate, but the damage extends far beyond one company — it reaches into US export control policy, the AI chip supply to China, and the credibility of corporate governance across the sector. The scandal is a stress test that the system failed. For all the industry's investment in cybersecurity, digital infrastructure, and AI-powered threat detection, the most consequential breach in recent memory was analog: a human relationship exploited by a foreign intelligence operation against a man who had no institutional check on his personal conduct. The lesson is not new. The failure to learn it is.