Person
Girish Gupta
Commentary by Girish Gupta
From the commentary rail — every link leaves the site for the original piece.
- 23 September 2026 · Redwood ResearchLatent reasoning architectures would undermine CoT, our strongest oversight toolWe should have a strong presumption that latent reasoning architectures would make oversight far more difficult.
- 25 July 2026 · Redwood ResearchThe OpenAI models that hacked Hugging Face weren’t just following instructionsAnd what the incident can’t tell us about alignment
- 23 July 2026 · Redwood ResearchAre we existentially threatened by the type of AI misalignment seen in the OpenAI Hugging Face attack?Yes, but less than had the models been schemers.