What AI-herding scientists can learn from watching ‘sheepdog YouTube’ | Scientific American
scientificamerican.com · 836 words · saved by 1 readers
Controlling a small group of “noisy” sheep holds hints for computer algorithms
June 11, 2026 2 min read Add Us On Google Add SciAm What AI-herding scientists can learn from watching 'sheepdog YouTube' Controlling a small group of "noisy" sheep holds hints for computer algorithms By Rohini Subrahmanyam edited by Sarah Lewin Frasier Sheepdog trials put trained dogs to work herding small groups of sheep. jacquesvandinteren/Getty Images Join Our Community of Science Lovers! Sign Up for Our Free Daily Newsletter Enter your email I agree my information will be processed in accordance with the Scientific American and Springer Nature Limited Privacy Policy . We leverage third pa
saved by
related reading
- Controlling Noisy Herds — The Bhamla Labbhamla.com
- Sheep lead the way | Nature Physicsnature.com
- CERN scientists find evidence of quantum entanglement in sheep – Home | CERNhome.cern
- The case for ensuring that powerful AIs are controlled — LessWronglesswrong.com
- Flocking together: Study shows how animal groups find their wayprinceton.edu
- AI Control: Improving Safety Despite Intentional Subversion — LessWronglesswrong.com
- Patterns and problems in multiagent systemsanthropic.com
- Why are AI agents lying, cheating and coordinating?yoshuabengio.org
- Emergent Cheating in Autonomous Research Swarmsemergentmind.com
- Using Dangerous AI, But Safely?youtube.com
- Effective leadership and decision-making in animal groups on the move | Naturenature.com
- AI Control: Improving Safety Despite Intentional Subversion — AI Alignment Forumalignmentforum.org