<ul><li>The Indian Computer Emergency Response Team (CERT-In) has warned against vulnerabilities in AI design, training, and interaction mechanisms.</li><li>Users are advised to consider using an anonymous account not linked to personal or professional identity when signing up for AI apps.</li><li>The vulnerabilities include data poisoning, adversarial attacks, model inversion, prompt injection, and hallucination exploitation.</li><li>Threat actors can create fake AI apps to trick users into downloading malware and stealing data.</li></ul>

CERT-In Warns Users Against ‘Vulnerabilities’ In AI Apps

Discover more