OpenAI Operator Agent Used in Proof-of-Concept Phishing Attack

Symantec demonstrated a proof-of-concept using OpenAI's Operator agent to automate a phishing attack: the agent identified a target, deduced their email, generated a PowerShell script after browsing web resources, and sent a convincing lure email. The exercise shows how generative-AI agents can extend attacker capabilities and reduce the barrier to entry for automated phishing, representing a future operational risk rather than reporting active exploitation today.