The U.S. Commerce Department is considering a new set of export controls that could significantly reshape how advanced artificial intelligence chips are sold to foreign buyers. According to a draft proposal reviewed by Reuters, the rules would require certain overseas customers to invest in U.S. infrastructure or provide security guarantees before they can access large shipments of American AI chips.
If implemented, the policy would represent a shift in how the United States manages the global distribution of advanced computing hardware. Instead of relying solely on bans targeting specific countries, the new approach would tie chip exports to economic and security commitments.
The proposal is still in an early stage and may change before becoming official policy. However, it signals that Washington is looking for new ways to maintain control over the supply of the most powerful AI chips while strengthening domestic technology infrastructure.
At the center of the draft proposal are new thresholds that would determine when export approvals are required.
Under the current draft, foreign purchases of more than 200,000 advanced AI chips would require government approval unless the buyer commits to investing in U.S.-based AI data centers or provides security guarantees related to the use of those chips.
Even smaller shipments could face new oversight. Deployments involving fewer than 1,000 chips may still require export licenses depending on the circumstances and the buyer’s location.
These thresholds reflect the increasing importance of large-scale computing clusters in modern AI development. Advanced models require thousands or even hundreds of thousands of GPUs working together, making hardware distribution a critical point of geopolitical influence.
The proposed rules also place new responsibilities on chip manufacturers and exporters.
Companies such as Nvidia and AMD could be required to monitor how exported chips are used once they reach foreign customers. This monitoring would likely involve software tools designed to track deployment patterns and detect whether chips are being assembled into massive computing clusters.
Clusters refer to large arrays of GPUs working together to train advanced AI systems. These systems power everything from large language models to advanced scientific simulations.
Under the draft proposal, foreign buyers might also need to install software that prevents exported chips from being combined into larger computing clusters beyond approved limits.
The goal of these restrictions is to prevent countries from secretly building large AI training facilities using imported hardware.
One of the most notable aspects of the proposed rules is the idea of tying chip exports to investment in U.S. infrastructure.
The approach mirrors recent deals involving Middle Eastern countries such as Saudi Arabia and the United Arab Emirates. In those arrangements, the countries agreed to invest heavily in American technology infrastructure in exchange for access to advanced AI hardware.
The Commerce Department appears to be considering a similar framework for future export approvals. Foreign buyers seeking large shipments of AI chips could be required to support U.S. data center development or other domestic technology initiatives.
This model effectively turns access to advanced chips into a form of economic leverage.
Officials at the U.S. Commerce Department confirmed that discussions around new export rules are underway. However, they emphasized that the proposed framework would differ significantly from earlier regulations introduced under the previous administration.
In January 2025, the United States introduced the Artificial Intelligence Diffusion framework, which aimed to limit global access to advanced AI computing hardware through licensing quotas and restrictions.
The Trump administration rescinded that policy in May 2025, arguing that it placed unnecessary limitations on allies while failing to effectively slow China’s technological progress.
The current approach appears to focus on a more flexible strategy. Instead of imposing broad restrictions, the government is considering targeted requirements that combine security controls with economic incentives.
The global race to develop advanced artificial intelligence systems has made high-performance chips one of the most strategically important technologies in the world.
Training modern AI models requires enormous computing power, and most of that power currently comes from GPUs produced by American companies. This concentration gives the United States significant influence over the development of AI worldwide.
At the same time, governments are increasingly concerned that unrestricted chip exports could allow rival nations to build large-scale AI infrastructure capable of competing with U.S. technology leadership.
As a result, export controls have become a central tool in global technology policy.
For American semiconductor companies, the proposed rules could introduce additional compliance responsibilities. Exporters would need to monitor chip usage and ensure that customers follow restrictions related to cluster formation and deployment.
This could increase administrative complexity for companies shipping high-performance GPUs overseas.
However, the policy might also create new opportunities. If foreign governments and investors are required to support U.S. infrastructure in exchange for chip access, American technology companies could benefit from increased domestic data center investment.
Such investments could expand the U.S. AI computing ecosystem while keeping advanced infrastructure within American borders.
For foreign buyers, the proposed rules may create a tiered system of access to advanced computing hardware.
Countries with strong diplomatic ties to the United States are likely to receive easier approval pathways. Nations outside that circle could face stricter requirements, including investment commitments or additional oversight.
China, which already faces strict export restrictions on advanced chips, would remain largely excluded from the market under the proposed framework.
The rules also attempt to prevent circumvention strategies in which restricted countries obtain chips indirectly through third-party buyers.
The policy debate around AI chip exports is also moving into Congress.
In February 2026, lawmakers advanced bipartisan legislation that would give Congress the authority to review and potentially block certain export licenses involving advanced AI hardware.
Supporters argue that greater oversight is necessary because advanced chips play a critical role in national security and technological competition.
Critics warn that overly strict regulations could harm American semiconductor companies by limiting access to international markets.
The final shape of the export rules will likely depend on negotiations between the executive branch, lawmakers, and the technology industry.
The emerging export strategy reflects a broader shift in U.S. technology policy.
Rather than relying entirely on bans, the government appears to be exploring a model that combines security restrictions with economic incentives.
Foreign governments and investors seeking access to advanced chips may need to support American infrastructure development, effectively linking hardware exports to domestic economic growth.
If adopted, the rules could reshape global AI supply chains while strengthening the United States’ position in the rapidly expanding AI industry.
For now, the proposal remains in draft form. But as governments and companies race to secure the computing power needed for the next generation of artificial intelligence, control over AI chips is becoming one of the most consequential policy debates in technology today.
BrandCrowd built its name as a go-to logo tool for small business...
There is a specific kind of disappointment that comes from opening a b...
AI companion apps have moved well beyond novelty status. What started...
Let’s start with an uncomfortable truth. Most people want their photos...
Amazon Web Services has introduced Amazon Connect Health, a new health...
Vidful AI has built its reputation on speed and accessibility. You ope...
Discussion