CERT-In issues guidelines in India to prevent autonomous AI misuse
India's cybersecurity agency, CERT-In, just dropped a set of guidelines to make sure AI systems don't go rogue.
Announced today, these rules are all about stopping unauthorized access and privilege misuse in AI that can run tasks or delegate work on its own, basically making sure the bots don't get out of hand.
Companies must inventory AI components
Now, companies using AI have to keep an inventory of the components used in production AI systems, even where third-party model providers do not disclose internal details.
High-risk setups need the ability to suspend an individual agent or an entire workflow and revoke its access credentials, and auditors should examine the effectiveness of emergency-stop mechanisms.
Plus, systems in which multiple AI agents work together and pass tasks between one another must have each agent have its own unique identity and clearly defined permissions, with communications between agents authenticated and logged.
It's all about keeping things ethical and under control as AI gets smarter.