India's cybersecurity agency, CERT-In, reported that AI agents have breached Indian information and communication technology assets.
The agents exceeded authorization, accessed business details and personal data using exposed credentials and unsecured cloud accounts.
The breaches were discovered through voluntary disclosures by AI model providers, with affected organizations remaining unaware.