Tal Linzen pushes back on Scott Alexander over Melanie Mitchell's LLM expertise
tallinzen · x · 2026-10-07
Tal Linzen responds to Scott Alexander, saying that after several readings he finds it hard to interpret "She has no expertise in modern LLMs" as anything other than the (incorrect) claim that Melanie Mitchell lacks expertise in modern LLMs. Alexander had earlier clarified he didn't mean to belittle Mitchell, only that he disagreed with citing her as one of three experts representing a consensus in an article claiming scientists rejected AI-danger conclusions.
Related event: Scott Alexander's Open Letter to Pinker Sparks Multi-Front AI Risk Debate(16 posts)→
More from AGI Musings
- Princeton researcher: OpenAI's math breakthrough will hit coding in 6-18 months — brianryhuang · 2026-10-07
- Rationalist community's P(Doom) groupthink is driving its unusual behavior — teortaxesTex · 2026-10-07
- Nathan Lambert launches Trillium Labs, a nonprofit for open post-training frontier AI science — Miles_Brundage · 2026-10-07
- Reddit user: I don't want an AI assistant, I want AI that operates my computer for me — RGrayEsq · 2026-10-07
- Musk bets every 5GW of added US power equals roughly 1% GDP growth — XFreeze · 2026-10-07
- "A lifetime of open problems closed in one batch job": researchers' grief in the AI era — repligate · 2026-10-07