Advanced AI chips are not just a hot commodity for enterprises, but also a source of great geopolitical tension. After months of back and forth over export terms and restrictions, the White House announced this week that chips such as Nvidia’s top-of-the-line product, the Blackwell chip, would not be made available too China — and may soon be restricted from all non-U.S.companies as well.
It’s this second threat that could have serious repercussions for global enterprises, even those with U.S. headquarters.
In initial comments aired on CBS’ 60 Minutes program on Sunday nov. 2, President Trump declared that “the most advanced [AI chips] — we will not let anybody have them other than the United states.” white House Press Secretary Karoline Leavitt also addressed the topic on Nov. 4, stating that the Blackwell chip and other advanced chips would not be sold to China at this time, even though she did not comment on other global exports.
Specialized AI semiconductor chips — GPUs and TPUs — fuel the immense computing power required for today’s most advanced AI models, wich are incre
Table of Contents
The global landscape of artificial intelligence is undergoing a notable shift, driven by increasingly stringent regulations surrounding the export of advanced AI chips. These restrictions, primarily aimed at limiting access for specific nations, present a complex challenge for Chief Details Officers (CIOs) seeking to deploy and maintain AI capabilities across international operations.
The Shifting Regulatory Landscape
Recent developments, including expanded restrictions announced in October 2023, have tightened the controls on exports of high-performance computing chips – including those from Nvidia and AMD – to countries like China and Russia. These regulations aren’t simply about preventing access to technology; they’re about controlling the potential for these chips to be used in applications with national security implications.
The new rules extend beyond simply prohibiting sales to specific entities. They now include:
* Broader definitions of “covered” chips: Expanding the scope of chips subject to export controls.
* Increased scrutiny of chip designs and software: Regulations now cover not just the physical chips, but also the designs and software used to create them.
* Requirement for licenses: Even for chips not explicitly prohibited, licenses might potentially be required for export to certain destinations.
The Impact on Global AI Operations
These restrictions are already having a tangible impact on businesses with international AI deployments. Key consequences include:
* Supply chain disruptions: Difficulty sourcing the necessary chips for existing and planned AI projects.
* Increased costs: The need to find option suppliers or redesign systems to use compliant chips can significantly increase costs.
* Performance limitations: Using less powerful, compliant chips may result in reduced AI performance.
* Operational complexities: Managing different hardware configurations and software stacks across different regions.
* Two-tiered operational capacity.
* Impeded performance in markets where the chips were formerly permitted.
If CIOs want to maintain performance levels across international jurisdictions, while adhering to new regulations, they will need to be ready to act — fast.
When AI is Confined by Geographical Boundaries
Restrictions over AI chip exports are not new, with the Biden administration first limiting certain exports through the U.S. Department of Commerce in September 2021. But the main target has always been China, with looser terms and loopholes for allied and neutral countries.
“Politically, the position has already been that the U.S.won’t sell China its most advanced chips — or even the second line for a while,” said Alexander Harrowell, principal analyst in advanced computing at Omdia. “Over the summe
The potential for White House restrictions on the export of advanced AI chips could significantly challenge global enterprises, impacting international operations. should limitations be implemented, companies will need to adapt to maintain compliance and competitiveness.
Working around, Not Against, AI Restrictions
CIOs facing this evolving legislative landscape have several options, ranging from short-term fixes to long-term investments. These strategies aim to navigate restrictions without halting AI initiatives.
Two-Tier Operations
One approach is to adopt a mixed-capacity model, similar to strategies previously used with the Chinese market. This involves utilizing advanced chips like NVIDIA’s Blackwell for domestic operations, while leveraging second-tier chips, such as H20s, for international operations.
“A two-tier market already exists and has done for a while — this was the point of products like NVIDIA A800, H800, H20, L20, and AMD MI308,” said Richard Harrowell, CEO of Pulsar Ventures.
However, Harrowell cautions that this strategy may become less viable. The ban on exporting even the H20 chip to China during the summer demonstrated a lack of demand for these second-tier options outside of China – “Why would you, when you could buy the real thing?” he asked. Consequently, NVIDIA and other manufacturers have limited incentive to produce these chips, particularly given the higher profit margins associated with Blackwells.
Fung described the production of two chip tiers as being a “significant risk” for chip manufacturers, since the rules are changing so rapidly and yet companies need to make these production decisions on much longer