A convincing video call from the CEO triggers an urgent credential reset. An autonomous AI agent conducts reconnaissance, crafts personalized messages, accesses internal tools, and moves stolen assets through cryptocurrency wallets. Every individual control appears to function—but the attack succeeds because the organization trusts synthetic identities, automated decisions, and fragmented evidence.

This technical panel reconstructs that attack from beginning to end. Experts in synthetic-media detection, agentic AI security, digital forensics, and blockchain intelligence will examine the artifacts left at every stage: media-generation signals, identity and authentication logs, prompts, retrieval context, agent tool calls, endpoint evidence, cloud telemetry, wallet activity, smart-contract interactions, and laundering paths.

Rather than debating whether AI attacks are coming, the panel will show how defenders can detect, contain, investigate, and prove them. Attendees will receive a practical framework for preserving AI evidence, correlating artifacts across systems, distinguishing human actions from autonomous execution, and reaching defensible conclusions when attribution is uncertain.