What Are the Main AI-Assisted Cyber-Attacks and Scams?

This report outlines how AI amplifies traditional cybercrime—phishing/BEC, deepfake-enabled impersonation, automated recon, and rapid malware scripting—across the entire attack pipeline, and prescribes practical defenses such as phishing-resistant MFA, SSO with conditional access, UEBA/behavioral analytics, macro/script restrictions, sandboxing, and rigorous verification workflows. It provides response guidance for individuals and organizations (containment, telemetry preservation, token/session revocation, OAuth/MFA audits), distinguishes AI from deepfakes with tailored controls, and forecasts increasing automation and realism in attacks, urging centralized telemetry, identity hardening, stress-tested human processes, and AI governance to sustain resilient, layered defenses.