1Password introduces privileged‑access limits for AI agents

1Password announced a new security feature called 1Password Privileged Access, designed to restrict the permissions granted to autonomous AI agents. The solution replaces permanent privileges with on‑demand access that is scoped to a specific task and automatically revoked when the task completes. Built on technology from Apono, which 1Password acquired in June, the system can enforce limits across cloud, hybrid, database, Kubernetes and other enterprise environments. Access requests are evaluated against the requester’s identity, the resource involved, the intended action and the sensitivity level; low‑risk requests are approved automatically, while higher‑risk ones trigger human approval via tools such as Slack, Microsoft Teams or Jira. All actions are logged for audit and compliance purposes. A key focus is on AI agents, which often inherit broader permissions from the configuring user. Under the new model, each AI agent receives its own identity and a narrowly‑defined privilege set, and the platform can compare declared intent with actual actions to revoke access if an operation deviates from its purpose.