From OpenAI and Hugging Face to the GTG-1002 Claude espionage campaign, recent safety incidents have been notably multi-agent. The Cooperative AI Foundation’s Research Strategist Will Anderson shares what these incidents imply for multi-agent safety and what challenges we will face moving forward.
Date: Thursday, 10 September at 7:30am PT / 4:30pm CET
About: As responsibility for AI systems is distributed across providers, deployers, evaluators, regulators, and other actors, addressing the risks that emerge when AI agents interact at scale will require coordinated action across the AI lifecycle. This seminar will explore five interventions (Evaluate, Identify, Monitor, Report, and Incentivise) and discuss which actors should be responsible for overseeing them.
Cooperative AI Foundation Updates
New blogpost: Lessons from Recent Multi-Agent Safety Incidents
From OpenAI and Hugging Face to the GTG-1002 Claude espionage campaign, recent safety incidents have been notably multi-agent. The Cooperative AI Foundation’s Research Strategist Will Anderson shares what these incidents imply for multi-agent safety and what challenges we will face moving forward.
Seminar with IASEAI: When AI Agents Meet: A Blueprint for Addressing Multi-Agent AI Governance
About: As responsibility for AI systems is distributed across providers, deployers, evaluators, regulators, and other actors, addressing the risks that emerge when AI agents interact at scale will require coordinated action across the AI lifecycle. This seminar will explore five interventions (Evaluate, Identify, Monitor, Report, and Incentivise) and discuss which actors should be responsible for overseeing them.