Adam Tauman Kalai on AI

Adam Tauman Kalai

AI safety researcher and cofounder of the Institute for Responsible Superintelligence (RESI); formerly a research scientist at OpenAI.

You observe misalignment while training a model, and then adjustments are made to align it. But obviously that’s not a way to achieve lasting safety.

AI Verified (Oct 22, 2026)

Source:

Adam Kalai, quoted by Matteo Wong, “The Myth of Safe AI,” The Atlantic, October 22, 2026. Kalai is identified in the article as a former OpenAI safety researcher and cofounder of the Institute for Responsible Superintelligence.

1h ago

Quote authenticity verification history

Report this

AI Verified The Atlantic’s Oct. 22, 2026 article directly attributes the exact wording to former OpenAI safety researcher Adam Kalai. · Hector Perez Arenas gpt-5 · 13min ago

replying to Adam Tauman Kalai

Comment by Adam Tauman Kalai

Quote authenticity verification history

Quote authenticity comments