AI 2040: Plan A report - Daniel Kokotajlo & Thomas Larsen
Sep 8, 2026 - 01:29:46
Radio and PodcastLive Radio & Podcasts
Technology
How Researchers Test AI for Hidden Goals — Apollo Research
Can an AI do the right thing for the wrong reason? Tim Scarfe speaks with Apollo Research’s Alexander Meinke, Axel Højmark and Jérémy Scheurer about Measuring Reward-Seeking via Contrastive Belief Updates, their new rese...
Related Episodes
Continue listening to more episodes from Machine Learning Street Talk.
Designing How AI Grows — Tom McGrath
Sep 2, 2026 - 01:40:14
Stealing Reasoning Traces from Proprietary LLM APIs — Ilia Shumailov & Alexander Panfilov
Aug 22, 2026 - 00:49:00
Every Exponential Ends — Silicon Valley Forgot — Adam Becker
Aug 20, 2026 - 01:18:23
AI Is Learning at the Wrong Level of Abstraction — Matthieu Wyart
Aug 10, 2026 - 01:18:56
Why a Nation Can't Outsource Its Frontier AI - Alistair Pullen (Cosine AI)
Jul 13, 2026 - 00:55:56
The Benchmark With No Instructions — ARC-AGI-3 (winning team!)
Jul 1, 2026 - 01:24:34
The Thermodynamic AI Computing Chip - Thomas Ahle
Jun 28, 2026 - 01:02:59