Nvidia CEO Huang sees strong demand for Blackwell chips
![Item 1 of 2 Jensen Huang, CEO of Nvidia, reacts during TSMC’s annual sports day in Hsinchu, Taiwan November 8, 2025. REUTERS/Ann Wang [1/2]Jensen Huang, CEO of Nvidia, reacts during TSMC’s annual sports day in Hsinchu, Taiwan November 8, 2025. REUTERS/Ann Wang Purchase Licensing Rights, opens new tab](/_next/image?url=https%3A%2F%2Fwww.reuters.com%2Fresizer%2Fv2%2FNMNPVKU4HBN3ZAVMBB2E4XBFVY.jpg%3Fauth%3D200f616f38ac4ecd630282292b81392894c4717a7e5499d93aa268012b4c6c28%26width%3D1200%26quality%3D80&w=1920&q=75)
News Summary
Nvidia CEO Jensen Huang stated that the company is experiencing "very strong demand" for its state-of-the-art Blackwell chips, and its appetite for wafers from TSMC is growing. Huang, speaking at a TSMC event, noted that Nvidia builds not only GPUs but also CPUs, networking, and switches, all associated with Blackwell chips. TSMC CEO C.C. Wei confirmed that Huang had "asked for wafers," with the specific number remaining confidential. Huang emphasized that Nvidia's success would not be possible without TSMC's support. Nvidia recently achieved a $5 trillion market valuation, earning Huang the moniker "five-trillion-dollar man" from Wei. Regarding potential memory shortages, Huang indicated that business was growing strongly and there would be shortages of "different things." However, he highlighted that SK Hynix, Samsung, and Micron, three excellent memory makers, have scaled up tremendous capacity to support Nvidia, and Nvidia has received the most advanced chip samples from all three. He deferred questions about memory price increases to the suppliers. Furthermore, Huang reiterated that there are "no active discussions" about selling Blackwell chips to China, as the Trump administration has prevented such sales, citing concerns they could aid the Chinese military and the country's AI industry.
Background
Nvidia is a leading global manufacturer of graphics processing units (GPUs), dominating the field of artificial intelligence (AI) computing. Blackwell represents Nvidia's latest generation of AI chip architecture, considered pivotal for advancing AI technology. Taiwan Semiconductor Manufacturing Co (TSMC) is the world's largest contract chipmaker, leading in advanced semiconductor manufacturing processes and serving as a crucial supplier to many tech giants, including Nvidia. The close partnership between Nvidia and TSMC is essential for its production and market supply. Since 2022, the U.S. government, including the incumbent Trump administration, has imposed and continuously tightened restrictions on exporting advanced AI chips and related technologies to China. These measures aim to curb China's progress in military and AI domains, significantly impacting the global sales strategies of U.S. chip design companies like Nvidia.
In-Depth AI Insights
What are the deeper strategic considerations behind Nvidia's assertion of strong Blackwell chip demand? - Nvidia's move, against the current geopolitical backdrop, aims to solidify its industry leadership and investor confidence by emphasizing the immense market demand for its latest AI chips. - This could also be a tactic to strengthen its bargaining position in wafer capacity negotiations with TSMC, ensuring it secures sufficient advanced process support to meet the anticipated explosive growth in the AI market. - Furthermore, this statement indirectly addresses market concerns about an AI bubble and potential supply chain bottlenecks by showcasing close collaboration with key memory suppliers. What do the Trump administration's chip export controls on China mean for Nvidia's long-term strategy and the global AI ecosystem? - While potentially limiting Nvidia's revenue growth in the Chinese market in the short term, in the long run, these controls might push Nvidia to further diversify its global market strategy and supply chain footprint, reducing reliance on a single market. - The restrictions could also accelerate the development of China's indigenous AI chip industry, leading to a more complex global competitive landscape and potentially a divergence in technology standards and ecosystems in the future. - For the global AI ecosystem, persistent geopolitical tensions could lead to a fragmentation of innovation pathways and increase operational complexities and costs for multinational corporations in R&D, production, and sales. How will TSMC and memory manufacturers' bargaining power and strategic positions evolve within the supply chain amid sustained strong AI chip demand? - TSMC, as the core of advanced chip manufacturing, will see its bargaining power continuously strengthened by its scarce capacity and technological edge, further cementing its strategic bottleneck position in the semiconductor value chain. - Memory manufacturers, especially SK Hynix, Samsung, and Micron in the High Bandwidth Memory (HBM) segment, will become increasingly critical due to their essential role in the AI chip stack, gaining greater pricing power and profit margins. - This supply-demand imbalance will drive deeper strategic collaborations between chip design companies (like Nvidia) and manufacturing and memory suppliers, potentially involving upfront payments or equity investments to secure stable access to critical resources.