An AI agent holding a long-lived, overprivileged credential can put systems and data at risk if it is compromised, IBM says. At its TechXchange conference in Atlanta on Tuesday, the company said HashiCorp Vault will issue AI agents secured, short-lived credentials scoped to what each one needs, with access denied and the session revoked in real time if an agent tries something outside its approved permissions.

The Vault work sits on top of IBM Agent Identity , which became generally available Monday. Agent Identity keeps the record for each agent: who owns it, its lifecycle status and the authority it has been granted. It then passes that identity and authorization context to Vault’s credential controls, IBM said in its TechXchange announcement .

Vault 2.2, Secure Hub and the identity layer

Vault 2.2, now generally available, gives agents their own secured accounts, renews their credentials automatically and adds post-quantum encryption.…