Agentic AI capabilities in the NetBrain 12.3 release can autonomously diagnose network issues, suggest fixes, and support IT teams in the remediation process.
Network automation vendor NetBrain is getting an intelligence boost via human and artificial intelligence updates.
The company had a busy 2025, delivering multiple releases that add AI capabilities to its platform and finalizing its acquisition by Blackstone. The momentum is continuing into 2026 with the NetBrain 12.3 release. The new version adds AI agents that can investigate network issues, identify root causes and suggest fixes.
The release marks NetBrain’s first major update under new CEO Bernadette Nixon, who joined the company in January. Nixon previously led AI search company Algolia, and she sees the 12.3 release as part of a shift toward what she calls “agentic NetOps.” The term describes AI acting alongside human engineers rather than just assisting them.
“We’re a leader in network automation, and we believe that we can help our customers most in the coming 12 months, and indeed coming years, by being the leader in agentic NetOps,” Nixon said.
AI Deep Diagnosis uses ReAct framework
One of the new highlighted features is a capability the company calls AI Deep Diagnosis. It uses a ReAct (Reasoning and Acting) framework that combines reasoning steps with automated actions. The AI agent queries NetBrain’s automation library iteratively based on what it finds at each step.
The system works with NetBrain’s digital twin representation and live network data. It shows its reasoning process on a network map so engineers can see how it reached its conclusions. If initial checks don’t find the problem, the agent tries different diagnostic approaches.
In testing, the system handled the majority of real-world network issues. “90% of the real-world network issues that they had when they threw them at it, it handled it,” Nixon said. “[People] couldn’t quite believe that it was at the 90% mark. People went in thinking, ‘Well, if this gives me 50% then great.'”
Cloud automation and rapid assessments
Beyond AI Deep Diagnosis, NetBrain 12.3 expands cloud automation capabilities across AWS, Azure and Google Cloud Platform. “We keep adding to our cloud automation capabilities,” Nixon said. “So, understanding cloud health, application assurance, security, all of this sort of stuff. “
The release also improves Quick Assessment capabilities for teams that might not be ready for AI Deep Diagnosis. “If you don’t want to go as far as, you know, maybe leveraging Deep Diagnosis, because you’re a little bit more conservative in how you want to approach things, then we’ve improved our Quick Assessment capabilities as well, and they can be added to run books and that sort of thing,” Nixon said.
The Quick Assessment features automate network validation and troubleshooting checks across multiple devices. Release 12.3 also includes an AI Runbook Companion that analyzes runbook results during troubleshooting and an AI Ticket Analysis feature that classifies historical ticket data. NetBrain added pre-approved auto-remediation for defined change management processes.
Tackling network operations challenges
Nixon met with multiple customers during her first weeks as CEO. The conversations revealed common pain points. Changes cause 50% to 80% of network incidents, according to Nixon. Resolution times for complex issues stretched from days to weeks at some organizations. Critical security vulnerabilities faced tight patch windows that teams struggled to meet.
A major airline has cut mean time to response from days to 30 minutes using NetBrain and is targeting five minutes, Nixon shared. One company reduced critical security vulnerability patch rollout from four months to two weeks, well within their four-week compliance window. And a managed service provider said it eliminated incidents caused by changes entirely after implementing NetBrain.
Nixon said that she doesn’t expect that organizations are going to want the linear progression that they’ve been used to over the last 10 years. Rather, they’re going to want to go a lot faster.
“They’re going to use AI as a catalyst to transform how their networks operate, have them think and act and respond alongside the humans that have managed them,” Nixon said.
