AI agent security

Present empirical findings from your research on how autonomous agents fail under adversarial conditions - the Summit's 'real research only' focus means they want data-backed insights, not vendor solutions, so frame your submission around specific vulnerabilities you've documented

Position your research as addressing the gap between agent capability growth and security infrastructure - this global expansion suggests the League is tracking how AI agent risks scale across different regulatory environments and deployment contexts

Highlight any cross-functional security frameworks your research has tested - the NYC deadline urgency and global scope indicate they're building a practical knowledge base, so research that bridges AI safety, policy, and technical defense will stand out