conv.

All stories
AIQuiet 30h · day 3

Nathan Lambert briefs Congress: China leads in open AI models by 2:1

AI researcher shares congressional testimony on shift in open-weight model dominance from U.S. to Chinese companies.

newest coverage 1d agosummary written 20h ago5 pieces123 likes & upvotes55 comments NewswiresHacker NewsMastodon

What to know

  • Nathan Lambert testified to Congress that Chinese AI companies have held a commanding lead in open-weight models since April 2025, with 3.2 billion total downloads versus America's 1.6 billion.
  • Chinese models including GLM-5.2 and Kimi K3 now match the advanced agentic capabilities that Anthropic's Claude Code first achieved in December 2025, available in open form rather than closed APIs.
  • The shift reflects a strategic advantage for China in open-weight development: while the U.S. leads in truly open-source models (with full training code and data), Chinese companies dominate the more commercially viable open-weight category that defines current adoption.

“We are living in a world where GLM-5.2 and Kimi K3, some of the latest, leading Chinese models, have enacted a step change in the commercial viability of open models — crossing a similar threshold in agentic capabilities that Anthropic's Claude Code crossed in December of 2025.”

Nathan Lambert, AI researcher · Interconnects AI ↗

Nathan Lambert AI researcher, policy analystMeta Model developerAlibaba Model developerDeepSeek Model developerAnthropic Model developer

Nathan Lambert briefs Congress: China leads in open AI models by 2:1
interconnects.ai

How it unfolded 2 developments, newest first · click a bar or a number to jump articlesposts

Peak 3 pieces in one hour at Sep 22, 5 PM; 5 pieces over 3 days (2 articles · 3 posts) Sep 21, 6 AM — 1 piece · 1 article — Newswires 1Sep 21, 7 AM — quietSep 21, 8 AM — quietSep 21, 9 AM — quietSep 21, 10 AM — 1 piece · 1 post — Hacker News 1Sep 21, 11 AM — quietSep 21, 12 PM — quietSep 21, 1 PM — quietSep 21, 2 PM — quietSep 21, 3 PM — quietSep 21, 4 PM — quietSep 21, 5 PM — quietSep 21, 6 PM — quietSep 21, 7 PM — quietSep 21, 8 PM — quietSep 21, 9 PM — quietSep 21, 10 PM — quietSep 21, 11 PM — quietSep 22, 12 AM — quietSep 22, 1 AM — quietSep 22, 2 AM — quietSep 22, 3 AM — quietSep 22, 4 AM — quietSep 22, 5 AM — quietSep 22, 6 AM — quietSep 22, 7 AM — quietSep 22, 8 AM — quietSep 22, 9 AM — quietSep 22, 10 AM — quietSep 22, 11 AM — quietSep 22, 12 PM — quietSep 22, 1 PM — quietSep 22, 2 PM — quietSep 22, 3 PM — quietSep 22, 4 PM — quietSep 22, 5 PM — 3 pieces · 1 article · 2 posts — Hacker News 1, Mastodon 1, Newswires 1Sep 22, 6 PM — quietSep 22, 7 PM — quietSep 22, 8 PM — quietSep 22, 9 PM — quietSep 22, 10 PM — quietSep 22, 11 PM — quietYesterday, 12 AM — quietYesterday, 1 AM — quietYesterday, 2 AM — quietYesterday, 3 AM — quietYesterday, 4 AM — quietYesterday, 5 AM — quietYesterday, 6 AM — quietYesterday, 7 AM — quietYesterday, 8 AM — quietYesterday, 9 AM — quietYesterday, 10 AM — quietYesterday, 11 AM — quietYesterday, 12 PM — quietYesterday, 1 PM — quietYesterday, 2 PM — quietYesterday, 3 PM — quietYesterday, 4 PM — quietYesterday, 5 PM — quietYesterday, 6 PM — quietYesterday, 7 PM — quietYesterday, 8 PM — quietYesterday, 9 PM — quietYesterday, 10 PM — quietYesterday, 11 PM — quiet 12
Sep 22yesterdaynow · 12:58 AM ET
  1. 2

    Lambert publishes congressional testimony as open analysis

    The expanded form of Lambert's congressional testimony is published on Interconnects AI as a public analysis of the open-weight model landscape, reaching Hacker News with 84 points.

  2. 1

    Lambert testifies to Congress on open-weight model balance

    Lambert briefs Congressional members and staff on the state of open-weight models in the context of U.S.-China competition, presenting his analysis that Chinese models lead by 2:1 in downloads and benchmark performance.

    “Chinese open-weight models surpassed American open-weight models in these two key areas about 18 months ago…China's download lead has grown to about 1.6B – with a total of 3.2B downloads, twice that of America's total.”
    — Nathan Lambert
    1. 2 outlets The current balance of power in open models

      first by HN Frontpage, 1d ago · also Interconnects

  3. background

    Anthropic's Claude Code reaches agentic capabilities threshold — Claude Code crosses a significant threshold in agentic capabilities that later Chinese models GLM-5.2 and Kimi K3 would also achieve in open-weight form.

  4. background

    Lambert publishes American Truly Open Models (ATOM) Project — The ATOM Project begins tracking American open-weight and open-source model development, providing tools to measure the competitive landscape with China.

  5. background

    China overtakes U.S. on Hugging Face downloads — China takes the lead in model downloads on Hugging Face in July 2025, primarily through the success of Alibaba's Qwen models, a key metric tracking open-weight model adoption.

  6. background

    Chinese AI companies establish lead in open-weight models — Chinese AI companies become the clear leader in open-weight models starting around April 2025, marking a shift from earlier American dominance through Meta's Llama models.