According to a Bloomberg report citing multiple people familiar with the matter, Chinese regulators are expected to grant export licenses for Nvidia’s H200 AI accelerator chips as early as this quarter. The approval would allow a limited number of local enterprises to purchase H200 processors for defined commercial applications, renewing Nvidia’s access to one of the world’s largest AI markets. Industry estimates suggest that China could account for as much as $1 billion in additional annual revenue from these sales, contingent on U.S. government fees being recognized below the revenue line as a surcharge rather than a direct cost. Analysts at Stifel have reiterated a Buy rating on Nvidia, highlighting that any uptick in China demand would be “materially additive” to the company’s earnings trajectory.

At the January Consumer Electronics Show, Nvidia unveiled its next-generation AI hardware platform, codenamed Vera Rubin, comprised of six distinct chip designs now in full production. Nvidia CEO Jensen Huang projected that Vera Rubin will begin shipping to hyperscale cloud providers in the second half of 2026, offering up to 10× lower inference token costs compared with the existing Blackwell architecture and up to 4× faster training for mixture-of-experts models. The company forecasts that the combined Blackwell and Rubin ecosystems will underpin roughly $500 billion of AI infrastructure demand over the next several years, cementing Nvidia’s lead among the so-called Magnificent Seven technology giants.

In early January, Nvidia confirmed plans to test its own Level 4 autonomous driving platform, DRIVE Thor, in a private robotaxi pilot by mid-2027. The announcement marks the company’s strategic shift from pure chip supplier to vertically integrated AI systems provider in the mobility sector. DRIVE Thor integrates Nvidia’s custom Alpamayo family of chain-of-thought models with specialized high-performance sensors and in-vehicle networking, targeting a 20% reduction in end-to-end perception latency versus current third-party solutions. Jensen Huang has characterized the move as the next phase of “physical AI,” positioning Nvidia to compete directly with established robotaxi players and unlock new recurring revenue streams from both hardware and per-ride software licensing fees.