US Threatens Sanctions on Chinese Open AI Models Over Alleged IP Theft

US Treasury Secretary Scott Bessent said on July 21, 2026 that Washington may impose sanctions on Chinese open-weight AI models over alleged intellectual property theft. According to TechCrunch, the move widens the Trump administration's campaign to slow China's progress in artificial intelligence.

What Scott Bessent said

The head of the US Treasury raised the possibility of sanctions aimed directly at Chinese open AI models, meaning the products themselves, not just companies or hardware supplies. The stated reason is alleged theft of American developers' intellectual property. The secretary did not say which models or companies could be affected, and TechCrunch's report names none.

Scott Bessent has served as Treasury Secretary since January 2025. Through its Office of Foreign Assets Control (OFAC), the US Treasury administers sanctions programs, from asset freezes to bans on American companies doing business with listed parties. Bessent's words about models are therefore not abstract rhetoric but a signal from an agency that already has the tools to act.

• July 21, 2026: Scott Bessent's statement, reported by TechCrunch
• Potential sanctions target: Chinese open-access AI models (open models)
• Stated justification: alleged intellectual property theft (IP theft)
• Context: the Trump administration's campaign to contain China's AI development

"The US may impose sanctions on

Chinese open AI models over alleged intellectual property theft," is how TechCrunch relays the position of Scott Bessent, the US Treasury Secretary.

Why the US is pressuring Chinese AI

Bessent's statement continues Washington's systematic effort to contain Chinese AI, TechCrunch notes. Until now, the main tools were export controls on advanced chips, in place since 2022 and tightened several times, along with restrictions on Chinese tech companies. Sanctions against the models themselves would be a qualitatively new level: the target would no longer be hardware but a software product that is already freely distributed around the world.

Accusations of intellectual property theft against Chinese AI labs have been made for years. The most prominent episode came in January 2025, when OpenAI said DeepSeek may have used "distillation" of its models, that is, training on the outputs of another company's system. That argument could now serve as a possible basis for government sanctions.

What DeepSeek and open weights have to do with it

The Chinese open AI models in question are primarily systems such as DeepSeek and Alibaba's Qwen. Their weights are published for free, and any developer can run the models on their own servers. That price tag is what made Chinese models so widespread: after the release of DeepSeek R1 in January 2025, open models from China took hold in startups and corporate products worldwide, including in the US.

TechCrunch's report does not say what sanctions on freely downloadable software might look like. A ban on using open weights inside the US would have to be enforced at the level of companies, cloud providers and app stores, and the AI industry has no precedent on that scale. A similar dilemma arose with TikTok: banning a widely used digital product is legally difficult and politically costly. Open weights, unlike an app, are not tied to a single distributor at all.

What this means

The US is shifting from restrictions on chips to pressure on Chinese software itself. If Bessent's threat becomes policy, American companies will have to reconsider their use of DeepSeek, Qwen and other open models, corporate lawyers will get a new line in their compliance checklists, and the split of the global AI market into two incompatible spheres will accelerate.