Title: When AI Becomes the Crisis Commander: Cyber-Resilient Decision Systems for Global Logistics

Abstract: Cyber incidents affecting transportation platforms, logistics providers, cloud services, carrier integrations, or operational data sources can quickly lead to physical supply chain disruptions. Responding effectively requires more than restoring technology; organizations must also determine which shipments, routes, facilities, partners, and customer commitments may be affected and identify appropriate recovery actions.

This presentation introduces the concept of an AI Supply Chain Crisis Commander, a decision-support framework that connects cyber-incident awareness with operational logistics recovery. The framework explores how AI can help identify potentially affected shipments, map operational dependencies, assess the impact of disruptions, evaluate alternative routes or recovery strategies, and generate evidence-based response plans for human decision-makers.

A central focus of the session is the security of the AI system performing this coordination. Compromised operational data, manipulated external information, prompt injection, excessive agent privileges, unreliable third-party integrations, or unauthorized tool execution could cause an AI-assisted recovery process to amplify rather than contain an incident.

The proposed approach combines data provenance, cross-source validation, agent identity, least-privilege access, human approval thresholds, runtime monitoring, decision auditability, and containment mechanisms to establish trust throughout the response process.

Drawing on real-world logistics scenarios and publicly available evidence, the session demonstrates how cyber resilience and supply-chain resilience increasingly intersect, and why organizations should design AI-enabled crisis-management systems to support both rapid recovery and secure, accountable decision-making.