Infosecurity Magazine

Indirect Prompt Injection in Web Content Targets AI Agents

Attackers have begun embedding hidden instructions in websites to target AI agents, according to new research. Zscaler's ...