Editorial illustration for Scott Bessent Takes Aggressive Stance on Chinese AI
Bessent Pushes Hard Line on Chinese AI Threat
Scott Bessent Takes Aggressive Stance on Chinese AI
Treasury Secretary Scott Bessent has picked a side in Washington's fight over Chinese AI, and it's the hard line. As Beijing's open-weight models close the gap on American frontier systems, Bessent has pushed for tighter controls, putting him at odds with colleagues who favor a lighter touch. He's one voice in a crowded room.
Commerce Secretary Howard Lutnick controls export licensing through the Bureau of Industry and Security and has been talking to lab leaders about building US alternatives to Chinese open-weight models. Other officials across the National Security Council, the Pentagon, and various regulatory agencies are weighing in too, each with a different read on how fast China is actually catching up and what Washington should do about it.
There's no single AI policy shop coordinating these views, no memo everyone signs off on. That means the eventual rules on chips, model exports, and open-weight releases will likely come down to whoever has President Trump's ear at the right moment. Bessent's push for aggression on China is one entry in a much longer list of competing agendas now shaping that outcome.
With little interagency coordination in the administration, policy will almost certainly end up being set by the people with the most influence with the president.
Why this matters
Bessent's "IP theft" framing tells us where the administration is headed on distillation, but it also tells us the White House still doesn't agree on how to get there. Ten sides to one argument means ten different sets of rules that developers and founders could be building against by next year. If Treasury wins the internal fight, expect export controls or legal threats aimed squarely at teams fine-tuning on outputs from Chinese open-weight models, a practice that's become routine in research labs chasing cheap performance gains.
That's a real compliance risk for anyone using DeepSeek-style checkpoints or similar releases as a base. We'd also flag the obvious tension: US firms have spent years justifying scraping and training practices that look a lot like what Bessent now calls theft. Watch whether Commerce or the NSC pushes back on Treasury's framing, and whether any of this turns into actual policy rather than a strongly worded post.
Right now it's rhetoric with teeth still to be determined, and builders should plan for volatility, not certainty.
Related Reading
Policy & Regulation Grammarly faces class-action suit over AI ‘Expert Review’ feature
Policy & Regulation Claude Mythos highlights EU AI safety gaps, says researcher Caroli
Policy & Regulation After ditching AI fitness apps and a Fitbit, I return to Peloton classes
From Archives Hugging Face Deploys Open GLM 5.2 After Closed AI Blocked Forensic Analysis
From Archives U.S. Considers Targeted Bans on Chinese AI Models Over Security
Common Questions Answered
What is Scott Bessent's position on Chinese AI and open-weight models?
Treasury Secretary Scott Bessent has taken an aggressive hard-line stance on Chinese AI, pushing for tighter controls as Beijing's open-weight models continue to close the gap with American frontier systems. He frames the issue as intellectual property theft and advocates for stricter export controls and legal measures against teams fine-tuning on Chinese model outputs.
How does Scott Bessent's approach differ from other administration officials like Commerce Secretary Howard Lutnick?
While Bessent advocates for aggressive controls and restrictions on Chinese AI, Commerce Secretary Howard Lutnick controls export licensing through the Bureau of Industry and Security and has been discussing building US alternatives with lab leaders, suggesting a lighter touch approach. This disagreement reflects the broader lack of interagency coordination on AI policy within the administration.
What could happen if Treasury Secretary Bessent wins the internal policy fight on Chinese AI?
If Bessent prevails in the internal administration debate, developers and founders can expect export controls or legal threats specifically targeting teams that engage in fine-tuning on outputs from Chinese open-weight models, a practice that has become routine in the industry. This would represent a significant shift in how the US regulates AI development and international model usage.
Why does the lack of interagency coordination on AI policy matter for developers and companies?
With little interagency coordination and multiple competing voices shaping policy, developers and founders could face ten different sets of rules by next year depending on which administration officials gain the most influence with the president. This uncertainty makes it difficult for companies to plan their AI strategies and compliance measures in advance.
Further Reading
- Bessent Says US Will Scrutinize Chinese AI Models for IP Theft - Bloomberg
- Chinese AI firms that rip off US models could face sanctions, Treasury Secretary Scott Bessent warns - New York Post
- America’s ‘biggest risk’ on AI is China getting ahead, Bessent says - South China Morning Post
- U.S. can hold AI talks with China because ‘we are in the lead,' Bessent tells CNBC as nations plan safety protocol - CNBC
- Bessent: U.S., China to Start Discussing AI Best Practices - The Wall Street Journal
