With the acquisition of AuthMind, IBM is expanding HashiCorp Vault’s approach to securing the agent lifecycle, helping organizations protect credentials and secrets not only when they are issued, but as agents use them across applications, tools and data.
AI agents are transforming how organizations operate. They increasingly act on behalf of people and other agents, accessing applications, tools, and data to get work done across complex environments.
As an agent’s autonomy increases, organizations face new security challenges. They need to know which agents are operating, who or what authorized them, what they can access, what actions they are taking, when to intervene and have accountability.
Securing the agent runtime from identity to action
Traditional identity security focuses on authenticating identities and controlling access. But autonomous agents introduce a more dynamic challenge. For an agent, knowing “Who are you?” is no longer enough. Organizations increasingly need to ask: Who authorized you? What are you allowed to do? For how long? And what are you actually doing with that authority? Agents can operate continuously, interact with multiple systems and delegate tasks to other agents, often without direct human involvement in every action.
Organizations need security controls that extend across the entire agent runtime, from establishing an agent’s identity and authority to monitoring its actions and responding when risk emerges.
IBM is building these capabilities into its security and automation platform, creating a continuous approach to securing agents across five critical areas:
- Register agents: Establish a trusted identity for each agent, with clear ownership and accountability for its actions.
- Delegate authority: Define and govern the authority agents receive to act on behalf of users, applications or other agents.
- Authorize access: Enforce least-privilege access to applications, tools and data using identity-based policies and short-lived credentials.
- Audit actions: Maintain visibility into agent activity, including which identities access resources, how permissions are used and whether behavior aligns with intended policies.
- Revoke risky behavior: Identify emerging identity risks and enable security teams to restrict or revoke access in real time when unsafe or unauthorized activity is detected.
HashiCorp Vault provides a foundation for securing access through identity-based controls, secrets management and short-lived credentials. As part of HashiCorp Vault, IBM Agent Identity can extend the approach to establishing and governing the identities and authority of AI agents. The addition of AuthMind, an identity security company focused on identity observability and threat detection, strengthens this approach by extending visibility beyond the point of access, helping organizations understand how credentials are used once they leave Vault.
Turning runtime identity observability into actionable intelligence
A critical challenge in securing autonomous agents is understanding what happens after access is granted.
An agent may have a legitimate identity and authorized credentials, but its behavior can change as it interacts with new tools, delegates tasks or accesses additional resources. Security teams need visibility into these interactions to distinguish expected activity from potential risks.
AuthMind brings identity observability capabilities designed to provide continuous visibility across human, machine and agent identities. At the center of its technology is an identity activity graph that maps relationships and access paths between identities and the resources they interact with. By connecting identity activity across environments, the technology helps security teams:
- Discover shadow agents and identities operating outside established governance processes by analyzing network traffic.
- Understand runtime behavior by observing how identities interact with applications, systems and data.
- Discover identity risks including unusual access patterns, excessive permissions and potential policy violations.
- Investigate and respond with greater context about which identities, resources and actions are involved.
This runtime intelligence complements Vault’s existing security controls by providing greater visibility into how authorized access is actually being used.
Rather than relying solely on static permissions or isolated security events, organizations can gain a more continuous understanding of identity activity and emerging risks across their hybrid environments.
Building the future for agent runtime security
As agents become more deeply embedded in business operations, securing them cannot be a one-time decision made when credentials are issued. Security must extend across the runtime and continuously connect what an agent is authorized to do with what it is actually doing. This is where IBM’s investments in agent runtime security come together to move organizations toward a continuous security model in which agent identities can be registered, authority delegated, access authorized, actions audited and risky access revoked.
AuthMind has previously collaborated with IBM through IBM Identity Protection, to extend governance, risk detection and real-time remediation to AI agents and other non-human identities operating alongside IBM Vault. IBM’s relationship with AuthMind began in 2023, when IBM Ventures invested in the company’s seed round, recognizing the strategic potential of its identity security portfolio. The acquisition enables us to build on that foundation and integrate these capabilities more deeply into Vault.
We’re excited to welcome the AuthMind team to IBM and continue helping organizations securely scale AI agents across their environments.
https://newsroom.ibm.com/blog-ibm-advances-agent-identity-security-with-new-vault-capabilities?shem=aimgspe,i3ppal,
