Layer3Labs warns Chinese AI models now handle much of U.S. traffic

Layer3Labs says Chinese-developed AI models are now processing as much as 46% of token traffic on a major U.S. model router, up from about 30% a year ago. The Miami firm says the shift is often hidden inside vendor products and shadow AI, creating data-sovereignty and compliance risks for businesses.

Why it matters: - Layer3Labs says a growing share of U.S. business AI traffic is being processed by Chinese-developed models, often without the customer’s knowledge. - The firm says that creates exposure for sensitive data, especially in healthcare, legal and finance. - Layer3Labs says firms with high shadow AI exposure face an estimated $670,000 breach premium per incident.

What happened: - Layer3Labs said as much as 46% of token traffic on the largest U.S.-based AI model router is now routed to Chinese models. - The company said U.S. models handled roughly 70% of that traffic a year ago and now handle closer to 30%. - The Miami-based AI implementation firm released the warning on July 22, 2026.

The details: - Layer3Labs said business owners often buy an AI feature without knowing which model processes the data underneath. - Cost-optimizing routers can send prompts to the cheapest available provider, which Layer3Labs said is increasingly a Chinese endpoint because inference can run 60% to 90% cheaper. - Andreessen Horowitz has estimated that 80% of U.S. startups build on Chinese base models, with Airbnb, Uber and Cursor named as adopters. - Research cited by Layer3Labs says 70% of enterprise AI activity sits outside IT oversight, including models small enough to run on a laptop. - Layer3Labs said any prompt processed by a Chinese-provider endpoint can fall under China’s National Intelligence Law, which can compel data disclosure to the state. - In July 2026, the House Select Committee on the CCP opened a probe into U.S. firms’ use of these models. - The NDAA already bars DeepSeek from Defense Department systems. - Layer3Labs said its review of dozens of small- and mid-sized business AI stacks found 70% contained at least one Chinese model the company had not knowingly approved.

Between the lines: - The issue is not just model choice. It is also routing, resale and white-labeling, which can hide the model behind a business-facing product. - That makes AI governance harder because companies may not know where customer data is actually processed or which jurisdiction applies. - The warning also suggests a widening gap between what businesses think they bought and what their vendors are actually using.

What's next: - Layer3Labs is urging businesses to run a provenance check on every AI tool that touches customer data. - The firm says regulated businesses should treat model visibility as a compliance step, not just an IT task. - Layer3Labs is offering a free AI Stack Sovereignty Check to identify which models are processing company data and to flag data-sovereignty and compliance risk. - Jonathan Teplitsky said business owners should ask whose jurisdiction their customer data lives under. - Mark Boyhous said companies cannot govern what they cannot see and should first find out what is in the stack.

The bottom line: - Layer3Labs says U.S. businesses may be relying on Chinese AI models far more than they realize, and the biggest risk is hidden exposure.