AI in the shadows: From hallucinations to blackmail

T2Covered by 1 source · first reported by Podcasts: Practical AI / TWIML / No Priors / Dwarkesh at 07 Jul 2025, 19:04 UTC

In the first episode of an "AI in the shadows" theme, Chris and Daniel explore the increasing concerning world of agentic misalignment. Starting out with a reminder about hallucinations and reasoning models, they break down how today’s models only mimic reasoning, which can lead to serious ethical…