Analysts say enterprises may face higher GPU costs and delivery delays by late 2025 as Nvidia absorbs potential revenue losses. Credit: gguy / Shutterstock China has accused Nvidia of breaching its anti-monopoly law, a move that could disrupt the chipmaker’s global operations and heighten risks for enterprises dependent on its GPUs as US-China trade tensions escalate. The State Administration for Market Regulation said a preliminary probe found Nvidia had violated conditions linked to its 2020 Mellanox acquisition and has opened an investigation. Nvidia, which relies on Taiwan’s TSMC and South Korea’s Samsung to manufacture its chips, counts China as one of its biggest markets, contributing about 13% of revenue last year. Any prolonged restrictions on sales there could complicate its operations and ripple through the semiconductor supply chain. This comes as demand for Nvidia’s GPUs continues to surge worldwide, driven by AI and cloud workloads. Limited global supply has already pushed up costs for enterprises, with CIOs competing with cloud giants and startups to secure hardware. Rising risks for enterprises China’s probe may not block Nvidia from the market, but it signals a greater risk of regulatory pushback, price volatility, and fragmentation of technology ecosystems. For CIOs, the concern is less about access in China and more about the spillover effects on long-term availability of critical compute resources worldwide. US enterprises already struggling with high costs and limited supply now face added uncertainty in securing the GPU power needed for AI and cloud initiatives. “They may not see an immediate short-term impact, but by the end of 2025, they could face higher procurement costs, even for non-H20 GPUs, and delays in delivery as Nvidia manages the potential loss of a double-digit share of its China revenue,” said Danish Faruqui, CEO at Fab Economics. “Rising procurement costs are likely to remain a sustained challenge for CIOs, as China’s $50 billion AI chip market shifts away from Nvidia amid ongoing geopolitical tensions between Beijing and Washington.” This means that enterprises should expect both higher costs and ongoing supply uncertainty. If Nvidia faces sustained regulatory pushback in China, its global revenue mix may weaken, and the company could try to offset losses by raising prices elsewhere. “Export controls and delays in shipping China-specific variants also reveal how geopolitical frictions can cascade into longer lead times,” said Manish Rawat, semiconductor analyst at TechInsights. The risk of dependence also grows. Enterprises tied to Nvidia’s CUDA ecosystem face steep hurdles in shifting workloads to AMD, Intel, or emerging cloud-native accelerators. “Should CIOs wish to create alternatives that are remotely comparable to Nvidia solutions, they need to spend more resources in time and cost, as the engineering process required to build such a system is extremely complicated and the right talent can be hard to find,” said Lian Jye Su, chief analyst at Omdia. Impact on the supply chain Reduced sales of Nvidia’s H20 and other China-specific GPUs could stress the global supply chain, according to Faruqui. This would hit US-based Amkor, which packages the chip, South Korea’s Samsung, which supplies HBM memory, Taiwan’s TSMC, which handles fabrication, and Foxconn, which integrates the hardware, among others. “Nvidia has already signaled to its partners to halt H20 production, a move that will ripple across the entire value chain, impacting materials, equipment, and testing suppliers as well,” Faruqui added. For global enterprises, this underscores the growing fragility of technology supply chains, which are now vulnerable to sudden policy shocks from both Washington and Beijing. “The move shows China using antitrust as leverage in the tech rivalry, much like the US uses export controls,” said Rachita Rao, senior analyst at Everest Group. “Even if the US supply is untouched, procurement cycles can still face sudden cost and compliance shocks.” If restrictions ripple out, Nvidia is more likely to push GPUs via cloud partners rather than accept deep unbundling, according to Rao. For enterprises, this could make cloud platforms the most reliable channel to access Nvidia capacity during periods of constraint. CIO priorities Analysts warn that the AI hardware ecosystem is becoming increasingly fragmented, as geopolitical and economic pressures create isolated demand pools and disconnected supply chains. CIOs must be ready for scenarios where US vendors face restrictions on bundling products in China or come under pressure to adjust business terms globally under regulatory scrutiny. “At the same time, China is accelerating domestic substitution to reduce dependence on US suppliers, potentially shrinking their addressable market,” Rawat said. “This dynamic creates a twofold risk for enterprises, near-term supply disruptions and longer-term cost escalation as vendors attempt to recoup margins elsewhere.” Procurement strategies must now factor in regulatory volatility, diversified sourcing, and potential price pressures. “CIOs must design AI infrastructures with supply security in mind for non-dispensable components, and supply resilience for fungible components, while balancing roadmap schedules and total cost of ownership across compute, memory, connectivity, storage, networking, and other infrastructure elements,” Faruqui said. For enterprises, the message is clear: geopolitical tensions are not just background noise, but direct factors shaping AI infrastructure planning. More Nvidia news: Nvidia rolls out new GPUs for AI inferencing, large workloads Nvidia networking roadmap: Ethernet, InfiniBand, co-packaged optics will shape data center of the future Nvidia’s new computer gives AI brains to robots GPUsTechnology Industry SUBSCRIBE TO OUR NEWSLETTER From our editors straight to your inbox Get started by entering your email address below.