Episode

Emergent Deception in LLMs

Podcast
Data Skeptic
Published
Oct 9, 2023
Duration seconds
1636
Processing state
failed
Canonical source
https://dataskeptic.com/blog/episodes/2023/emergent-deception-in-llms
Audio
https://pscrb.fm/rss/p/mgln.ai/e/35/traffic.libsyn.com/secure/dataskeptic/emergent-deception-in-llms.mp3?dest-id=201630
JSON
/v1/public/podcasts/data-skeptic/episodes/emergent-deception-in-llms
Markdown
/podcast/data-skeptic/emergent-deception-in-llms.md

Actions

  • POST https://stenobird.com/v1/public/podcasts/data-skeptic/episodes/emergent-deception-in-llms/transcription-requests
    Idempotently request low-priority transcript generation for this episode.
  • GET https://stenobird.com/podcast/data-skeptic/emergent-deception-in-llms.md
    Read the agent-friendly Markdown representation of this episode resource.

Summary

On today's show, we are joined by Thilo Hagendorff, a Research Group Leader of Ethics of Generative AI at the University of Stuttgart. He joins us to discuss his research, Deception Abilities Emerged in Large Language Models . Thilo discussed how machine psychology is useful in machine learning tasks. He shared examples of cognitive tasks that LLMs have improved at solving. He shared his thoughts on whether there's a ceiling to the tasks ML can solve.