AI-driven threats are accelerating as attackers exploit hallucinated domains, remote code execution flaws, and agent manipulation to bypass defenses. Today’s coverage highlights urgent patching needs, risks with AI automation, and the policy whiplash impacting global model deployment.