Ex-AI2 post-training lead: Chinese labs lack compute, not talent; distillation edge is months
lishali88 · x · 2026-10-15
Nathan Lambert, former post-training co-lead at AI2 and founder of nonprofit open AI lab Trillium, joined a16z's podcast with Lisha Li to discuss whether open models can keep up with the frontier.
Key points:
- When he tells researchers in China that an average OpenAI researcher reportedly has 5,000 H100s to work with, "their jaw just falls to the floor"; he says Chinese labs have the talent but are mostly short on compute.
- The distillation story is overblown: based on research, its edge is only a few months, and "there's a bit of cope" in thinking only American labs can build good models.
- Much frontier progress outside math and code isn't easily replicated via open models.
- Also covered: how open models play into RSI dynamics, who captures open-model value, and why he's building Trillium Labs.
Related event: Nathan Lambert launches nonprofit open-source lab Trillium(3 posts)→
More from AGI Musings
- Lonely men turned to AI companions after years of being told to leave women alone — and got judged for that too — VraserX · 2026-10-15
- MIT study: best AI models went from solving 2% to 35% of theory problems in two years — nordicinst · 2026-10-15
- danielrock: the real AI divide is staying informed and chill while acting — danielrock · 2026-10-15
- Researcher calls out influential AI voices for feeding doom and pessimism narratives — _onionesque · 2026-10-15
- AI came for math. Now it's chasing the rest of science, MIT researchers say — Wired AI · 2026-10-15
- Researcher pushes back on "robotics will soon be solved" claim from top lab insiders — adam_karvonen · 2026-10-15