59 Episodes

  1. 35 - Peter Hase on LLM Beliefs and Easy-to-Hard Generalization

    Published: 24/08/2024

  2. 34 - AI Evaluations with Beth Barnes

    Published: 28/07/2024

  3. 33 - RLHF Problems with Scott Emmons

    Published: 12/06/2024

  4. 32 - Understanding Agency with Jan Kulveit

    Published: 30/05/2024

  5. 31 - Singular Learning Theory with Daniel Murfet

    Published: 07/05/2024

  6. 30 - AI Security with Jeffrey Ladish

    Published: 30/04/2024

  7. 29 - Science of Deep Learning with Vikrant Varma

    Published: 25/04/2024

  8. 28 - Suing Labs for AI Risk with Gabriel Weil

    Published: 17/04/2024

  9. 27 - AI Control with Buck Shlegeris and Ryan Greenblatt

    Published: 11/04/2024

  10. 26 - AI Governance with Elizabeth Seger

    Published: 26/11/2023

  11. 25 - Cooperative AI with Caspar Oesterheld

    Published: 03/10/2023

  12. 24 - Superalignment with Jan Leike

    Published: 27/07/2023

  13. 23 - Mechanistic Anomaly Detection with Mark Xu

    Published: 27/07/2023

  14. Survey, store closing, Patreon

    Published: 28/06/2023

  15. 22 - Shard Theory with Quintin Pope

    Published: 15/06/2023

  16. 21 - Interpretability for Engineers with Stephen Casper

    Published: 02/05/2023

  17. 20 - 'Reform' AI Alignment with Scott Aaronson

    Published: 12/04/2023

  18. Store, Patreon, Video

    Published: 07/02/2023

  19. 19 - Mechanistic Interpretability with Neel Nanda

    Published: 04/02/2023

  20. New podcast - The Filan Cabinet

    Published: 13/10/2022

2 / 3

AXRP (pronounced axe-urp) is the AI X-risk Research Podcast where I, Daniel Filan, have conversations with researchers about their papers. We discuss the paper, and hopefully get a sense of why it's been written and how it might reduce the risk of AI causing an existential catastrophe: that is, permanently and drastically curtailing humanity's future potential. You can visit the website and read transcripts at axrp.net.

Visit the podcast's native language site