According to Cybersecurity Insiders, nearly half of AI agents in use are operating without meaningful security controls in place. The report highlights a significant security gap in AI agent deployment across organizations, raising concerns about potential vulnerabilities and data protection risks.