Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Join this seminar to examine the emerging risks posed by autonomous AI systems interacting in multi-agent environments. Explore how competitive advantages drive the adoption of advanced AI agents that can act independently and interact with each other and humans, creating complex new multi-agent systems with significant under-appreciated risks. Delve into how multi-agent risks fit within the broader AI governance and safety landscape, identify key risk factors leading to harmful interactions in multi-agent AI settings, and understand different mechanisms through which failure modes can arise. Discover strategies for risk mitigation and promising research directions based on findings from the Cooperative AI Foundation's recent report on Multi-Agent Risks from Advanced AI. Learn from Research Director Lewis Hammond, a DPhil candidate in computer science at Oxford University and fellow at the Wilson Center, alongside discussants Gillian K. Hadfield from Johns Hopkins University and Michael Dennis from Google DeepMind, as they address critical questions about ensuring AI technologies are developed and governed safely and democratically in an increasingly autonomous multi-agent world.
