IBM is pulling together a package of tools to help enterprises discover, identify, manage, and control the growing number of AI agents in use.
The package makes use of IBM’s newly available Agent Identity software and its HashiCorp Vault and watsonx Orchestrate packages, which together can identify and manage AI agents built across multiple platforms, according to the company.
The agentic challenges IBM wants to address are myriad, said Tyler Lynch, outbound CTO for IBM Automation, in a press briefing about the news.
“Today, anybody can launch an AI agent, but how do we, as an enterprise, have visibility of every agent and what it’s doing?” Lynch said. “We also have to really come to grips with the fact that when business users and knowledge workers go create or use agents, they’re not security experts. They don’t understand what least privilege is. They often will operate under the principle of that user and just impersonate them.”
That creates an audit problem, as it can be unclear if an agent performed a specific action or operation, he said. “And then we have this governance problem and revocation problem. If an agent is going rogue or spending too much money, how do we identify that? How do we stop it in real time?”
The need to more effectively manage AI agents is growing and becoming impossible to ignore. Research firm Gartner predicts that by 2028, an average global Fortune 500 enterprise will have more than 150,000 agents in use, up from less than 15 in 2025, generating significant agent sprawl, IT complexity and management challenges.
“Many organizations resort to blocking or restricting the use of AI agents, but this is not a long-term solution. If employees are unable to work in the sanctioned tools, they will likely go around the organization’s controls and start using shadow AI which presents far greater risks,” Max Goss, senior director analyst at Gartner, stated in a recent report. “Organizations need to find a balance where they can govern agents and manage sprawl, but also safely empower employees to innovate with these tools.”
With today’s announcement, IBM is extending its HashiCorp Vault security management system to assign AI agents short-lived credentials that will let them perform whatever actions they were assigned. In addition, IBM is adding Vault Agent Identity to keep tabs on who owns each agent, what authorization it has, and its lifecycle status.
Currently, Vault manages the secrets and credentials that applications and services use to access business systems. It provides that control through centrally managed secrets, automated rotation for supported credentials, and dynamically generated credentials that expire or can be revoked.
HashiCorp Vault already secures non-human identities for organizations including ABN AMRO, Adobe, athenahealth, BBVA, BlackRock, Canva, Duke Energy, Expedia, London Stock Exchange, Starbucks, and Swiss Federal Railways. Now, HashiCorp will extend those capabilities to AI agents, issuing secured, short-lived credentials that limit each agent to what it needs, IBM says.
When used with HashiCorp Vault, Agent Identity registers agents in a central inventory, captures ownership and metadata, classifies risk, documents approved tool connections, and becomes the authoritative record for each agent, such as who owns it, its lifecycle status, and the authority it has been granted, according to IBM.
Built to extend existing identity and security investments, Agent Identity supports Okta, Entre, PING, and others to act as an agent registry. It then can share relevant identity and authorization context with Vault’s credential controls—enforcing least-privilege access and enabling real-time revocation when an agent attempts an unauthorized action, according to IBM. Once an agent is known, IBM Agent Identity helps control how it acts, IBM stated.
IBM has also integrated HashiCorp Vault with Nvidia’s OpenShell secure runtime software to help enterprises scale agents.
In addition, IBM watsonx Orchestrate works with Vault’s credential controls and Agent Identity to ascertain what each agent’s job is, govern its authority, and enforce least-privilege credential access. Together, these controls help organizations authorize and audit agent actions and rapidly revoke trust and credentials when risk changes, according to IBM.
Together, these capabilities provide a coordinated agentic identity security control plane spanning three domains where agent security is enforced: agentic orchestration runtime, an identity and access management layer, and a secrets and credentials layer, IBM stated.