YouCongress
About Sign up

Comment by Nathan Lambert

Avatar
Machine learning researcher; post-training lead at the Allen Institute for AI (Ai2); founder of the Interconnects newsletter and open-models advocate
Tools like Mythos will make the best attackers have more powerful tools of the trade, but it won't be handing a nuke to every teenager connected to the internet.
AI Verified source (2026)
Like Share on X 20d ago

Quote authenticity verification history

AI Verified Verified. The source URL resolves to the article "Claude Mythos and misguided open-weight fearmongering," credited to Nathan Lambert and dated April 9, 2026; the exact quoted sentence appears in the body of the article. ([interconnects.ai](https://www.interconnects.ai/p/claude-mythos-and-misguided-open)) · YouCongress gpt-5.4-2026-03-05 · 6h ago
AI Verified Quote verified via web search (interconnects.ai returned HTTP 403 to WebFetch, but search results quote the exact text verbatim from Nathan Lambert's article "Claude Mythos and misguided open-weight fearmongering" on interconnects.ai, his own newsletter — the primary source). Year 2026 is correct. The article's thesis is that anti-open-weight fearmongering is misguided, so Lambert's "against" vote on the statement "Open-source AI is more dangerous than closed-source AI" aligns correctly with the quote, which downplays the danger of open models. Author attribution confirmed. · Hector Perez Arenas claude-opus-4-7 · 20d ago
replying to Nathan Lambert