SolarWinds launches Avior AI system with human approval for IT operations

SolarWinds has announced a new product called Avior for managing IT systems. The standalone artificial intelligence system is designed for mission-critical environments where uptime matters. It launched on October 14, 2026, according to reports from the company and SiliconANGLE.

The system aims to solve a common problem with AI tools that can produce different results each time they run. SolarWinds says this inconsistency creates risk for businesses where security is on the line. To fix this, Avior uses a fixed framework of predefined goals and approval controls to govern how tasks get done.

Safety features keep human operators in control during critical operations. Potentially dangerous actions always require explicit human approval before they proceed. Every workflow is logged with granular audit trails so teams can track progress and review decisions later.

IT teams can use Avior to handle routine tasks like reducing alert noise or fixing configuration errors. The system correlates related events into a single actionable incident to help engineers work the problem. It also detects changes made outside normal channels and restores devices to their last-known-good state.

This application works across cloud and self-hosted environments without requiring data migration. Customer data is masked so it does not get used to train the underlying models. Avior reaches third-party systems through a connection protocol called Model Context Protocol.

Cullen Childress serves as interim chief executive after Sudhakar Ramakrishna’s tenure ended in late September. He says reaction time alone is no longer enough as environments grow more complex. The leadership team emphasizes moving proactively to reduce operational risk and build resilience.

Avior is available starting today through a guided onboarding program for customers. SolarWinds did not disclose pricing details for the standalone application. More capabilities are planned to be added over time as the system evolves.