I’m thrilled to sit down with Desiree Sainthrope, a legal expert with a deep background in drafting and analyzing trade agreements. With her extensive experience in global compliance and keen interest in intellectual property and emerging technologies like AI, Desiree offers a unique perspective on the complex intersection of law, technology, and international trade. Today, we’re diving into the controversial deal announced by President Trump, which allows Nvidia and AMD to sell AI chips to China in exchange for a 15% cut of their revenue. Our conversation explores the legal implications, national security concerns, and the broader impact on U.S. policy and the semiconductor industry.
What was your first impression of President Trump’s deal to allow Nvidia and AMD to sell AI chips to China while taking a 15% revenue cut for the U.S. government?
Honestly, I was taken aback. This arrangement feels like uncharted territory in the realm of U.S. government and business relations. I’ve spent years studying trade agreements and export policies, and I can’t recall a precedent where the government directly claimed a percentage of private companies’ export revenue in this manner. It’s a bold move, and while creativity in policy isn’t inherently bad, this raises immediate questions about legality, fairness, and long-term consequences for the tech sector.
How does this deal fit within the framework of existing U.S. laws, particularly regarding the government’s authority to impose what looks like an export tax?
That’s a critical point. If we look at the U.S. Constitution, specifically Article I, Section 9, it explicitly prohibits Congress from imposing taxes on exports. Now, this deal isn’t a congressional act—it’s a presidential arrangement—which muddies the waters. There’s no clear legal precedent for the executive branch to unilaterally claim revenue from private exports like this. It could potentially be challenged in court as an overreach of authority, but the outcome is uncertain. Past cases involving executive actions under the Trump administration have sometimes been upheld, often due to broad interpretations of national security or emergency powers. So, while it’s questionable, it’s not guaranteed to be struck down.
In what ways does this deal diverge from previous U.S. policies on exporting advanced technology to China?
It’s a stark departure. Historically, U.S. policy has been heavily restrictive on exporting advanced semiconductors to China, primarily due to national security concerns. The fear has always been that such technology could enhance China’s military capabilities or be repurposed in ways that threaten U.S. interests. Past administrations, including Trump’s earlier policies and the Biden administration’s stance, placed tight controls on these exports through the Commerce Department. This deal flips that approach by prioritizing economic gain over those established safeguards, which is a significant shift in strategy.
Even though President Trump has described these chips as older models, like the H20, do you think there are still national security risks at play?
Absolutely, there are risks. Even if these chips are not the latest generation, they’re still advanced enough to potentially contribute to military applications or other sensitive uses in China. The challenge lies in enforcement—once technology is exported, it’s incredibly difficult to monitor how it’s used or prevent it from being diverted to unintended purposes. China’s opaque systems and history of dual-use technology development amplify these concerns. So, while the argument about them being older chips might lessen some worries, it doesn’t eliminate the underlying risks.
President Trump has argued that China already has access to similar technology, so this deal isn’t a big leap. How do you evaluate that justification?
I’m skeptical of that reasoning. While it’s true that China has been aggressively developing its own semiconductor capabilities, assuming they already have equivalent technology oversimplifies the issue. There’s a difference between having similar tech and having direct access to U.S.-designed chips, which might still offer unique advantages or insights. Moreover, using this logic to justify sales risks normalizing a reactive policy—essentially saying, ‘If they have it, we might as well profit from it.’ That could undermine long-term U.S. technological leadership and set a dangerous precedent for future export decisions.
How do you see this 15% revenue cut affecting the U.S. semiconductor industry over time?
It could have a chilling effect. Companies like Nvidia and AMD operate on long timelines—designing chips and planning fabrication can take a decade or more. They need stability and support to innovate, not additional financial burdens. Taking a 15% cut of their revenue might discourage investment in R&D or alter how they strategize future designs, especially if they feel the government could impose similar terms down the line. Instead of taxing their sales, I’d argue the government should be bolstering these firms with grants or incentives to maintain their competitive edge globally.
What is your forecast for the future of U.S.-China technology trade policies given the precedent this deal might set?
I think we’re at a crossroads. If this deal holds and isn’t successfully challenged, it could usher in a new era of transactional trade policies where economic gain trumps security concerns. That’s troubling because it might encourage a ‘pay-to-play’ model, not just in semiconductors but across other high-tech industries. On the flip side, if legal or political pushback mounts, we could see a return to stricter controls or even bipartisan efforts to codify clearer boundaries on executive power in trade. The next few years will likely reveal whether this is a one-off experiment or the start of a broader shift in how the U.S. navigates technology exports to China.
