Virtana Agentic Observability Lets AI Agent Teams Run IT Response

Virtana has introduced a new way for enterprise technology teams to handle outages, and it puts teams of AI agents at the center of the response. The Palo Alto company announced Agentic Observability Enterprise Management on October 28, 2026, its new Virtana agentic observability offering, a capability that lets its AI agents assemble specialist agents from across a business to manage an IT incident from the first warning signal through to a verified fix.

According to the company, the launch turns the traditional IT war room into a coordinated workforce of agents. When something goes wrong, Virtana agents use the Model Context Protocol and the open Agent2Agent protocol to bring in other agents with the right expertise, whether they sit in IT service management, cloud infrastructure, security, or software development.

The agents share more than 25,000 signals as they investigate a problem together, according to Virtana. They coordinate decisions and carry out governed remediation, work that once required engineers working across several separate systems. The company says the result is incident handling that moves at machine speed, with engineers staying in the loop and enterprise policy deciding which actions still need human approval.

Why agent teams change incident response

The announcement matters because it treats agent teamwork as the product, not a side feature. Most observability tools watch systems and raise alerts for people to act on. Virtana is betting that the next step is agents that can call on other agents, compare what each one sees, and agree on a fix before a human has to stitch the picture together.

That shift echoes a wider pattern across enterprise software this week. Vendors are racing to show that their agents can do real operational work, not just answer questions. Jedify, for example, announced Data Agents on the same day, autonomous analysts that monitor business metrics and investigate changes in plain language. The common thread is agents that act inside real workflows, with guardrails around what they are allowed to change.

For IT leaders, the practical question is trust. An agent team that can remediate incidents on its own needs clear limits, an audit trail, and a fast way for a person to step in. Virtana says enterprise policy governs which actions require approval, which keeps humans as the backstop while agents handle the repetitive coordination that slows incident response down.

Open protocols and the road to autonomous operations

The use of open protocols is the detail worth watching in this Virtana agentic observability launch. The Model Context Protocol gives agents a shared way to reach tools and data, while Agent2Agent lets agents from different vendors talk to each other. If those standards hold, an enterprise will not be locked into a single supplier's agents. It will be able to mix specialists from across its stack, which is exactly the kind of interoperability that agent ecosystems need to grow.

That interoperability is also where the idea of an Agentic IQ comes in. Virtana chief executive Paul Appleby described it as the collective intelligence of agents that understand the whole business, reason across complex systems, and solve problems no single agent could solve alone. Reported by the company, the framing points to a future in which the strength of an enterprise is measured by how well its agents collaborate, not by how capable any one model is.

There are still open questions. Multi-agent incident response will need to prove itself on messy, real-world outages, where signals conflict and a wrong remediation can make things worse. Governance, cost control, and clear accountability for agent actions will decide whether IT teams embrace this model or keep agents on a short leash. The launch is a strong statement of intent, but the proof will come from the first wave of deployments.

For readers following the agent ecosystem, the Virtana launch is a useful marker. The agentic web is moving from single assistants toward teams of agents that divide work, share context, and finish jobs together. Related coverage on this site includes the recent Meta Muse Singapore launch, which showed task-doing agents expanding into everyday apps, and the broader genznewz front page for ongoing agent news.

Sources: Virtana announcement and Jedify Data Agents announcement.