Interim Nvidia CEO Colette Kress announced during an earnings call that the semiconductor maker has halted all shipments of its licensed Hopper architecture H200 data center products to China. The guidance moving forward explicitly excludes licensed Hopper products for the Chinese market. Kress did not provide specific reasons for the sudden strategic pivot or clarify whether Nvidia will seek alternative business avenues in China.

The decision follows the resignation of former CEO Jensen Huang after reports surfaced regarding extramarital affairs, including one involving a suspected Chinese intelligence asset. Before Huang’s departure, Nvidia faced regulatory scrutiny over its vetting processes for Chinese partners handling the processors. Additionally, Reuters reported that approximately 600,000 H200 chips were recently held up in Hong Kong due to Beijing withholding large-scale import approvals in an effort to curb reliance on U.S. technology.

The withdrawal eliminates a significant potential revenue channel for Nvidia in Asia and underscores escalating geopolitical and regulatory friction surrounding foreign AI hardware deployments. For U.S. hardware developers and global enterprises, the move highlights ongoing supply chain volatility and tightening export environments for frontier compute infrastructure.

Why it matters

  • Expect heightened supply constraints and shifting geopolitical strategy for advanced chip distribution across Asian markets.

  • U.S. AI startups targeting global hardware deployments must prepare for persistent compliance shifts and potential market access restrictions.

  • Chinese domestic chipmakers may see accelerated demand as foreign hardware access contracts under local and international policies.

Source: theverge.com