Omnissa launches Elara beta for AI governance control

Omnissa has launched Elara, a new software layer for AI governance and enterprise change control. The product is entering beta as companies face growing scrutiny over how AI tools are used inside corporate systems.

The launch is aimed at IT and security teams trying to track how employees and automated systems use AI applications, models and agents across an organisation. Elara is designed to bring together data from people, identities, devices, apps and AI tools, then use that information to decide whether actions should be allowed before they take place.

The approach reflects a broader concern among large companies as AI moves from experimentation into routine operations. Omnissa cited its own research showing unsanctioned AI tools on 75% of enterprise-managed devices, while separate research from Turbotic found that more than a third of businesses saw governance and compliance as the main barrier to moving AI projects beyond the pilot stage.

Governance focus

Elara is positioned as a layer that sits above existing systems rather than replacing them. Many tools can authorise actions only within the platforms they directly manage, while Elara is intended to connect different systems and apply policy controls using broader business context.

The software works across Omnissa products and third-party environments. Its aim is to give organisations a single point from which to assess AI use, apply controls and manage changes that could affect multiple business systems.

Functions in the beta version include detection of unapproved AI tools, controls on access to AI models and centralised approval of high-impact changes. The system also records decision trails in real time, including what happened, who authorised it and which systems were affected.

Brian Link, Product Lead for Elara at Omnissa, said the product was developed in response to the fragmented way companies oversee AI and digital work tools.

"Enterprises are trying to enable AI without losing control of what is happening across their environments," Link said.

"That gets harder when every system has its own, limited view of the world. Elara builds on the foundation Omnissa has established helping customers manage digital work across traditionally siloed systems. By bringing those signals together, Elara gives IT the context to apply the right guardrails and make better-informed decisions about what should happen next."

Broader challenge

The product arrives as companies shift from AI tools that support staff to systems that can act on their behalf. That change has raised questions about who authorises those actions, how they are monitored and how firms maintain records for compliance and internal oversight.

Elara can be used to set guardrails for users, applications and agents accessing AI models through Omnissa's AI Gateway. The software can also meter token consumption across different model providers, which may appeal to companies seeking tighter control over AI usage without relying on a single supplier.

Another focus is change management. Omnissa said Elara evaluates proposed changes for conflicts, freezes and dependencies before execution, then either clears the action, limits it or sends it to an approver.

Industry analysts say the need for that kind of oversight is becoming more pressing as autonomous systems spread through business workflows.

"As the agentic AI era sees the technology move from just assisting employees to actually taking actions across business systems, governance becomes a necessity. Organisations need to understand not only which AI tools and agents are being used, but the identities, devices, applications and data the tools access, the actions they are permitted to take, and what happens when something goes wrong," said Adam Holtby, Principal Analyst at Omdia.

"This creates an important role for technologies such as Elara that can help organisations with visibility, policy and control across digital business ecosystems. The challenge now is connecting those existing control points into a governance model that can keep pace as AI becomes more autonomous and embedded across business workflows," Holtby said.

Omnissa, previously VMware's end-user computing business before becoming independent, said it serves 26,000 customers worldwide. Elara is now available in beta.