Nvidia Shares Fall to $162.85 After Huang Resigns and Report Says Chipmaker Paid for Customers That Bought Its GPUs

Nvidia's board accepted Jensen Huang's resignation as chief executive on November 11, and the company named Colette Kress, its chief financial officer, interim chief executive the next day. Kress keeps the finance role while the board considers internal and external candidates. The resignation came days after Nvidia opened an investigation into reporting that Huang "has been the target of a foreign intelligence effort." Huang admitted an affair with a woman whom reports describe as a suspected Chinese intelligence operative. Nvidia has not stated a reason for his departure.

On November 13 The Wall Street Journal reported, according to TechPowerUp and Wccftech, that Nvidia invested in AI startups and smaller cloud providers, which then bought its GPUs. The report says this happened during a period when TSMC output exceeded the number of buyers ready to pay. Huang reportedly pushed for more such investment. Kress reportedly challenged the arrangements on the grounds that some customers could not pay, and she urged a reduction in Nvidia's reported $250 billion commitment to OpenAI. The headline claim that Nvidia "engineered scarcity" is the outlets' framing. The underlying allegation is that Nvidia financed demand for its own chips. Nvidia's response was not in the coverage reviewed, and the original Journal article was not accessible for this report.

Investors reacted to the uncertainty over both leadership and demand. Nvidia closed at $162.85 on November 13, down 2.01 percent, on volume about 15 percent above its three-month average. HSBC cut the stock to hold and lowered its target from $375 to $170, arguing the shares will struggle until a permanent chief executive is named. A higher-than-expected inflation reading also weighed on chip stocks that day.

The financing question matters to engineers because it sets the price and availability of GPU capacity. Nvidia has committed to buy back unused cloud capacity from neocloud operators. One example is a $6.3 billion order with CoreWeave that runs through April 2032. If those guarantees shrink under new leadership, smaller providers lose the backstop that let them borrow against GPU fleets.