Ben Werdmuller · 2026-07-20 · notable
Ben Werdmuller: American AI is locked down and losing to open Chinese models
Werdmuller argues US labs' closed-and-proprietary strategy is costing them the AI platform war. He cites estimates that about 80% of new startups now build on Chinese open-weight models like Qwen and Kimi.
Werdmuller argues America's proprietary AI stance is a losing strategy — open Chinese models are already the developer default.
What is it?
'American AI is locked down and proprietary. It's losing.' is a July 20 essay on Ben Werdmuller's personal site, werd.io. Werdmuller is an independent technologist and open-web advocate. The piece reached the Hacker News front page with 284 points the day it was posted.
How does it work?
The essay's central claim is that AI models have no durable moat beyond brand and distribution, so restricting access to a US frontier model does not lock in an advantage — it simply hands the developer ecosystem to whoever ships open weights. Werdmuller cites an estimate that roughly 80% of new startups today are building on Chinese open-weight models, and points at Alibaba's Qwen line and Moonshot's Kimi K3 as concrete recent evidence of that pull.
Why does it matter?
Werdmuller's argument matches what public data now shows — Kimi K3, Qwen 3.8 Max, GLM-5.2, and DeepSeek V4 have taken over the top of the open-weight leaderboards while US labs mostly ship closed APIs. If Werdmuller is right, the US risks the same ecosystem loss that Windows had against Linux servers or that iOS is starting to have against Android globally. He calls for American labs to compete via open weights, plus support for public and federated AI aligned with democratic values.
Who is it for?
policy watchers, open-source advocates, AI infrastructure decision-makers
Try it
Read the essay at werd.io