US Gov't Revokes Controversial AI Hardware Export Rule that Would Mandate Investments From Foreign Companies — New Export Rules Are Still in the Works, Though

Why It Matters

The withdrawal removes a costly barrier for global AI chip buyers, preserving supply chains, while the pending framework could still reshape international AI competition.

Key Takeaways

• Draft rule required foreign AI clusters to match U.S. investments
• Proposed tiered licensing differentiated medium and large installations
• Withdrawal avoids doubling hardware costs for overseas customers
• New export framework still under development, targeting security and dominance
• Internal disagreement influences future U.S. AI export policy

Pulse Analysis

The Commerce Department’s draft rule, often dubbed the “U.S. Diffusion Rule,” sought to attach a financial condition to every export of high‑performance AI accelerators. Under the proposal, any foreign entity operating a large‑scale AI cluster—defined as more than 200,000 Nvidia GB300 GPUs—would have been obligated to invest an equal amount in U.S. AI infrastructure. A parallel tiered licensing scheme would have granted faster approvals to medium‑scale installations while imposing strict transparency and on‑site inspection requirements. The policy aimed to leverage America’s technological edge to fund domestic AI development and limit adversary access.

The abrupt withdrawal of the rule eliminates what many industry analysts warned would double the cost of AI chips for customers in Europe, the Middle East and Asia. Companies such as AMD, Nvidia, Cerebras and other accelerator makers can now continue selling without the onerous investment clause, preserving existing supply‑chain relationships and avoiding potential delays in AI projects abroad. Investors have welcomed the move, noting that a sudden price shock could have slowed adoption of generative‑AI workloads and strained revenue forecasts for hardware vendors. At the same time, the decision underscores lingering uncertainty within the administration about how to balance national‑security concerns with commercial competitiveness.

Despite the retreat, the Commerce Department signals that a revised export framework is still in the pipeline. The next iteration is expected to retain a tiered approach but may replace the dollar‑for‑dollar investment requirement with more nuanced criteria such as end‑use verification, geopolitical risk scoring, and limited‑quantity licensing. For multinational chip manufacturers, staying ahead of evolving compliance obligations will be critical; early engagement with U.S. regulators can mitigate disruption and shape policy outcomes. Ultimately, the direction of U.S. AI export controls will influence global AI leadership, supply‑chain resilience, and the pace of innovation across the sector.