Latest
Google DeepMind Backs $10M Research Fund to Address Safety Risks in Large-Scale Multi-Agent AI Systems
Google DeepMind has joined Schmidt Sciences, the UK's ARIA, the Cooperative AI Foundation, and Google.org to establish a $10 million research fund focused on identifying and mitigating safety risks that emerge when millions of AI agents interact autonomously. Concerns include prompt injection attacks, fraud, and broader cybersecurity threats that could destabilize digital infrastructure before large-scale agent deployment.
AI自動駕駛飛行安全
10 days agoSource: MIT Technology Review
