Webinar Preview: Experts to Discuss the Limits of Human Oversight in the Era of AI Agents

mmitchell_ai · x · 2026-07-31

The Data & Society Research Institute will host a webinar on August 20th titled 'The Limits of Human Oversight in the Age of AI Agents.'

Based on their primer 'The Oversight Fallacy,' the event will explore why simply putting 'humans in the loop' (via approval prompts and pause buttons) is insufficient to control AI agents. Speakers include Margaret Mitchell (Chief Ethics Scientist at Hugging Face) and Madeleine Clare Elish (Head of Evaluations at Google DeepMind). The discussion will draw on fieldwork in a computational biology lab to analyze what effective agent oversight truly requires.

Original post →

More from AGI Musings

AGI Musings channel →