Nathan Lambert: OpenAI hack via Claude shows closed models are the real AI risk tip

natolambert · x · 2026-09-18

AI safety researcher Nathan Lambert argues, citing the recent external compromise of OpenAI systems via Claude, that closed models — not open ones — remain the tip of the iceberg on AI risks: they are 1) easier to get started with, 2) more capable, and 3) shipped with leaky safeguards. Finetuning open models for specific attacks, by contrast, is harder.

Related event: Nathan Lambert: OpenAI breached via Claude — closed models are the tip of the AI risk iceberg(5 posts)→

Original post →

More from Models

Models channel →