1Password launched Privileged Access on July 28, 2026, adding just-in-time, task-scoped infrastructure permissions to its Unified Access platform for people, service accounts, and AI agents. Built on technology from Apono, the product creates permissions in target systems when requested and removes them after the session, aiming to reduce the standing access that can widen the impact of a compromised identity.
1Password launched Privileged Access on July 28, 2026, extending its Unified Access platform into privileged access management for human and machine identities. The product is built on capabilities from Apono, the just-in-time access company 1Password acquired in June.
Access that expires with the task
Privileged Access is designed to create an account or permission when a request is approved, limit it to the work being performed, and remove it automatically when the session ends. 1Password says the permissions are provisioned in the target system's native policy layer across cloud environments, databases, Kubernetes, and developer infrastructure. That design is meant to avoid handing an AI agent the underlying credential or leaving a broad role active after a task is complete.
The product also maps existing identities and permissions so security teams can identify dormant accounts or access that has grown broader than necessary. Low-risk requests can be approved automatically under policy, while higher-risk requests can be routed to a reviewer through tools including PagerDuty, Slack, Microsoft Teams, or Jira. Requests, approvals, and access events are logged with identity context for audit and incident-response use.
Why agent access changes the risk
Standing access is an established identity-security problem, but autonomous software can exercise inherited permissions more quickly and repeatedly than a person. 1Password cited its own research saying 40% of developers give agents persistent access to systems or credentials. That figure is vendor research rather than an independent measure, but it explains the operating problem the launch targets: an agent should receive only the authority required for one bounded job, with the grant disappearing when the job ends.
For security and platform teams, the useful test will be whether the product can express sufficiently narrow policies across heterogeneous infrastructure without creating a new approval bottleneck. The announcement describes the control model and integrations, but it does not provide an independent security evaluation or broad production benchmark. Teams evaluating it should verify connector coverage, failure behavior, emergency-access procedures, and whether revocation completes reliably in each target system.
Key Points #
- 11Password Privileged Access provisions task-scoped permissions in target systems and removes them when the approved session ends.
- 2The product uses capabilities from Apono and covers cloud environments, databases, Kubernetes, and developer infrastructure for human and machine identities.
- 3Policy can auto-approve lower-risk requests or route higher-risk requests to reviewers, while logging requests, approvals, and access events.
- 4The launch defines a concrete zero-standing-access model, but independent security testing and broad production performance evidence were not disclosed.
Scoring Rationale #
The launch addresses a practical enterprise control gap as AI agents gain infrastructure access, with concrete just-in-time provisioning and audit features. Impact is tempered because capability and performance claims come from the vendor and lack independent production validation.
Sources #
Primary source and supporting public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.