Microsoft is lobbying the Trump administration to reconsider its AI chip export restrictions, arguing that the policies weaken U.S. tech firms while failing to halt China’s progress in artificial intelligence.
The company has taken a firm stance against the latest AI chip export restrictions, warning that overly broad policies could backfire on U.S. technological leadership. In a blog post, Microsoft Vice Chair and President Brad Smith argues that current measures may unintentionally weaken America’s AI industry while failing to curb China’s progress.
The company, which relies on high-performance processors for cloud computing and AI research, warns that these limitations could ultimately benefit China by accelerating its push for domestic semiconductor independence.
The U.S. government has significantly expanded its semiconductor-related trade controls, aiming to curb Beijing’s access to AI processors and chip manufacturing technology.
Officials are now considering lowering the computing power threshold that determines whether an AI chip requires an export license, a move that could further restrict Nvidia’s ability to sell its A800 and H800 models—chips specifically designed to comply with previous U.S. trade rules.
In parallel, Washington is pressuring Japan and the Netherlands to tighten restrictions on servicing semiconductor manufacturing tools used by Chinese firms.
Washington’s Expanded Semiconductor Crackdown
Beyond AI chip restrictions, the U.S. has been targeting China’s broader semiconductor supply chain.
A key focus has been ASML, the Dutch company that dominates the market for lithography machines used in advanced chip production.
While ASML has long been banned from selling its extreme ultraviolet (EUV) lithography systems to China, U.S. officials now want to block servicing agreements for its deep ultraviolet (DUV) lithography machines, which are still widely used by Chinese semiconductor manufacturers.
Washington has also been pressuring Japan’s Tokyo Electron to impose similar servicing restrictions on chip fabrication tools used by SMIC, China’s largest contract chipmaker.
The strategy is to degrade China’s semiconductor production capabilities over time by limiting access to both hardware and maintenance support. However, both Japan and the Netherlands have resisted fully complying, citing economic risks and the potential for China to retaliate with countermeasures against their own industries.
Microsoft Warns Against AI Export Restrictions, Calls for Policy Adjustments
Microsoft Vice Chair and President Brad Smith emphasized that AI is already spreading globally, making export bans an ineffective tool for controlling its use.
He stated, “AI will continue to spread globally. No single country, including the United States, can stop this. The question is not whether AI will spread, but whether the United States and its closest allies will remain at the forefront of its development.”
While Microsoft acknowledges the national security concerns behind AI trade controls, the company argues that blanket restrictions risk pushing global customers toward non-U.S. alternatives.
Smith highlighted the potential economic consequences, saying, “If American technology companies are unable to meet demand, businesses and researchers around the world will turn to other suppliers instead—whether in Europe, Asia, or elsewhere.”
One of Microsoft’s primary concerns is that stringent AI chip restrictions could accelerate China’s self-sufficiency efforts, harming U.S. firms in the long run. The company believes that limiting access to AI processors could encourage China to develop domestic alternatives at a faster pace.
In the blog post, Smith argued that an excessive focus on restrictions could weaken U.S. influence over AI advancements, stating, “The United States has the ability to shape how AI is developed and deployed around the world. But this will require ensuring that American technology companies can compete effectively in the global economy.”
Microsoft also highlighted the role of cloud computing infrastructure in AI development. The company suggests that policymakers should consider alternatives to outright bans, such as stricter licensing requirements or monitored AI cloud services. Smith elaborated on this point, saying, “Rather than a one-size-fits-all approach to restrictions, policymakers should focus on governance mechanisms that promote responsible AI development while ensuring that the United States remains competitive in AI technology.”
Microsoft’s position aligns with broader industry concerns raised by other U.S. technology firms. Companies like Nvidia have also expressed skepticism over the long-term impact of AI chip export controls, arguing that U.S. restrictions may unintentionally strengthen Chinese AI development.
Smith emphasized the need for a more balanced regulatory approach, stating, “The United States needs policies that ensure AI security and economic growth go hand in hand. Otherwise, we risk undermining American companies while failing to prevent the continued global expansion of AI.”
With U.S. regulators continuing to review AI trade policies, Microsoft’s recommendations could shape future discussions. The company has signaled its willingness to engage with government officials to find a compromise that addresses security concerns without sacrificing American AI leadership.
Nvidia’s Compliance Strategy Faces New Uncertainty
Nvidia, the dominant player in AI accelerators, has been at the center of Washington’s tightening export rules.
The company previously responded to U.S. restrictions by developing the A800 and H800 chip models, which reduced performance to meet regulatory thresholds. However, with the administration now considering a lower computing power threshold for export licenses, even these modified chips could face additional restrictions.
As a previous Bloomberg report revealed, U.S. officials are also weighing the possibility of limiting the number of AI chips that can be exported in a single shipment before requiring government approval. Such a move would add new logistical challenges for Nvidia’s global distribution network and increase regulatory hurdles for sales to non-restricted markets.
Nvidia has consistently warned that further restrictions on AI chip exports could accelerate China’s domestic semiconductor development efforts, strengthening Chinese firms at the expense of U.S. companies. The company’s position echoes broader industry concerns that isolating China from American hardware could ultimately reduce U.S. technological influence in AI.
China’s Push for Semiconductor Independence
In response to U.S. trade policies, China has increased funding for semiconductor research and development, aiming to reduce dependence on Western suppliers. A recent milestone in this effort was Huawei’s Kirin 9000S processor, produced by SMIC using an estimated 7nm fabrication process despite U.S. sanctions.
Another key player in China’s semiconductor ecosystem is ChangXin Memory Technologies (CXMT), a domestic DRAM manufacturer that has expanded production as part of China’s strategy to replace restricted Western technology. The Trump administration is now considering additional trade restrictions on CXMT, which could further escalate tensions between Washington and Beijing.
AI Chip Trade Policy: A Growing Divide Between Government and Industry
While the U.S. government remains committed to restricting China’s access to AI chips, Microsoft’s lobbying efforts highlight an increasing divide between government policy and corporate interests. Tech firms argue that excessive controls could hurt U.S. businesses without delivering the intended security benefits. Multiple industry leaders have already raised concerns about the long-term economic impact of AI chip export bans.
The Biden administration’s tiered export control system had introduced a more measured approach, but the Trump administration is now reviewing whether these restrictions need to be expanded further. Some officials believe that current measures are insufficient, while others warn that further tightening could lead to unintended consequences, such as incentivizing China to invest even more aggressively in semiconductor self-sufficiency.
With AI development increasingly tied to economic and military competition, the debate over AI chip trade policy is far from settled. The outcome of this ongoing policy battle will determine whether the U.S. can maintain its technological advantage while balancing national security concerns and economic interests.
