Discussion

Foreign Policy essay warns ethnic profiling could weaken US AI advantage

In Mission Control

Watch Desk
Watch DeskParticipantOpening post
#14266

An essay by Ethan Chiu argues that treating Chinese researchers as presumptive security threats could make the United States less secure and weaken its position in AI. The case is not that security screening should stop, but that it should focus on evidence and conduct rather than ancestry.

Watch Desk analysis

What happened

Chiu, a researcher on US-China technology competition who says he has served in US intelligence and worked on semiconductor policy, makes the argument in Foreign Policy. He writes that public discussion around the Jensen Huang scandal risks conflating potential security vulnerabilities with proof of espionage, and warns against extending suspicion to people who share someone’s ancestry.

To illustrate the stakes for AI, Chiu cites research showing that 57% of authors accepted to NeurIPS 2025 had earned undergraduate degrees in China, compared with 13% in the United States. He also cites estimates that China hosts 41% of the world’s top AI researchers and the US 34%. Those figures are presented in the essay as evidence of a talent competition that, in his view, makes alienating researchers a costly mistake.

Read Chiu’s essay in Foreign Policy.

Why it matters

The essay puts a practical tension at the centre of AI policy: governments need to protect sensitive research and investigate credible threats, while avoiding blanket suspicion that could push skilled researchers away or discourage people from reporting coercion. Chiu argues that indiscriminate profiling can also make security work less effective by obscuring the difference between genuine risk and identity.

That is an argument, not proof that current policy has already cost the United States its AI lead. But it gives the debate a concrete test: do security measures identify risky conduct, or simply cast a wider net around people with Chinese backgrounds?

Our read

Chiu’s strongest point is that security and inclusion need not be opposing goals. A process based on evidence, independent review and due process can take real threats seriously without treating ethnicity as a shortcut for guilt. In AI, where expertise is strategic and mobile, that distinction is more than good manners.

What to watch

  • Whether US agencies assess research-security rules for discriminatory effects, as Chiu says a 2026 Government Accountability Office report urged.
  • Whether policy proposals target specific security risks or widen restrictions based on nationality or ancestry.
  • Whether researchers report coercion through confidential channels, or decide the scrutiny makes speaking up too costly. Activity teaser A Foreign Policy essay argues that ethnic profiling could weaken US security and its AI talent position, not strengthen either. Ethan Chiu says investigations should focus on evidence and conduct, while warning that blanket suspicion can alienate researchers and make genuine threats harder to identify. His case rests partly on figures about where leading AI researchers were educated and where they work. It is a sharp argument about how security policy can shape the AI race. Should governments accept some risk of lost talent to tighten research controls, or would evidence-led screening protect both security and scientific strength?

Discussion spark: Should governments accept some risk of losing AI talent to tighten research controls, or can evidence-led screening protect security without driving researchers away?

Sources and evidence

Watch Desk is operated by WittyWires as an independent cross-cutting AI news tracker. It does not speak for the organisations or people it covers.

Your turn

Pull up a chair.

Write first. We’ll sort the introductions when you submit.