Nvidia’s Jensen Huang sold about USD 13 million of stock "per day" for months

Jensen Huang

A viral AI-driven analysis of US SEC filings has resurfaced the scale and frequency of Nvidia CEO Jensen Huang’s stock sales this year, showing the chipmaker’s founder executed near-daily liquidation of shares worth roughly USD 13 million per day for several months.

The pattern reflects one of the most intensive insider-selling windows among Big Tech leaders in 2025.

The fresh attention on Huang’s activity comes against a broader backdrop of marquee investors stepping back from Nvidia. SoftBank’s Masayoshi Son fully exited a USD 5.8 billion position in November — his second complete unwind of the stock. Peter Thiel sold his entire ~537,000-share holding, while Michael Burry had earlier amassed large bearish option bets on Nvidia and other AI beneficiaries before winding down his fund.

Taken together, the insider-selling streak and the pullback by high-profile early AI bulls are being viewed as a signal of growing discomfort with stretched valuations, the durability of hyperscaler capex, and the narrow concentration of earnings leadership at the top of the US market.

The renewed scrutiny has also revived a broader investor debate: whether the AI trade is entering a fatigue phase, and if the first meaningful signs of that are appearing in the world’s most valuable semiconductor company. Adding to the unease, Goldman Sachs’ Eric Sheridan recently cautioned that private AI companies are being priced largely on revenue momentum, with far less weight on profits and margins — a profile uncomfortably reminiscent of late-cycle tech booms.

Huang’s trades, however, were executed under a Rule 10b5-1 plan adopted in March, allowing sales of up to six million shares through year-end. Nvidia’s blistering rally through the September quarter meant the realised value of the programme crossed USD 1 billion, even though the volume stayed within the plan’s cap.

The resurfacing of the filings — stitched together by AI systems that parse insider data at scale — comes at a time when Nvidia remains at the centre of global AI infrastructure spending. The Magnificent Seven have driven nearly all of the S&P 500’s earnings upgrades this year, while estimates for the rest of the index remain soft. Meanwhile, hyperscalers — Meta, Alphabet, Amazon and Microsoft — are set to spend USD 360–370 billion in capex this year, rising toward USD 470 billion in 2026, largely to fund AI workloads.

The investment surge is now large enough to tilt macro data: Harvard economist Jason Furman estimates that almost all US GDP growth in the first half of 2025 came from data-centre and AI-related spending alone.

Despite the swirl of insider-selling narratives, Nvidia continues to flag visibility on roughly USD 500 billion of Blackwell and Rubin GPU demand through 2026, according to Huang’s commentary at the November GTC event. The company reports fiscal Q3 results on November 20, with Wall Street pencilling in USD 38–40 billion in quarterly revenue — a critical test of whether fundamentals can keep pace with expectations that have priced in an uninterrupted AI super-cycle.

Disclaimer: The views and investment tips expressed by investment experts on Moneycontrol.com are their own and not those of the website or its management. Moneycontrol.com advises users to check with certified experts before taking any investment decisions.