Modern IT operations teams don’t lack data. They lack time and clarity once the data tells them something is wrong. Observability platforms like Dynatrace give deep visibility into applications, infrastructure, dependencies, logs, and traces. But when an issue surfaces, operators still have to work out what’s affected, what’s the likely root cause, who owns it, and what to do next.
That gap between observability insight and operational action is where ServiceNow AIOps is evolving.
Meet the AIOps AI Specialist
Traditional AIOps reduces noise, correlates events, and identifies probable root cause. The AIOps AI Specialist builds on that foundation with AI agents that perform much of the investigation an operator would otherwise do by hand. It assists with alert triage, incident investigation, service health analysis, and service reliability workflows.
For a deeper look at the AIOps AI Specialist, check out my colleague Victoria Lo's (@vlo ) blog.
Bringing Dynatrace into the investigation
[Figure 1: Express List showing automated grouping of 7 alerts]
The Dynatrace integration is where things get interesting. ServiceNow provides a Dynatrace MCP server AI agent that investigates an alert using a Dynatrace problem ID or entity name. It retrieves problem and entity details, gathers relevant spans and logs, and returns a problem overview, key findings, and actionable next steps.
The integration uses Model Context Protocol (MCP) to give the ServiceNow AI agent access to Dynatrace capabilities and context. The investigation doesn’t stop at what’s already in the alert. The agent can pull additional evidence directly from the observability platform.
Think of the division of labor this way:
- Dynatrace detects and explains what’s happening across your environment.
- ServiceNow connects that intelligence to service context, operational processes, and the workflows that decide what happens next.
The value isn’t just connecting two platforms. It’s giving AI two complementary kinds of context:
- Dynatrace provides observability context: telemetry and technical detail about applications, infrastructure, and services.
- ServiceNow provides operational context: alerts, incidents, services, configuration items, ownership, historical data, and workflows.
- AIOps AI Specialist provides reasoning and orchestration, using agents and tools to investigate and surface findings.
Together, these create a richer investigation than alerts or raw telemetry could offer alone.
What this looks like for an operator
[Figure 2: Observability dashboard in Service Operations Workspace]
Say Dynatrace flags a problem affecting an important application, and that signal reaches ServiceNow AIOps. Without AI, the operator moves between platforms, finds the matching entities, reviews telemetry, interprets the problem, and documents what they found. With the AIOps AI Specialist, the Dynatrace MCP agent gathers problem details, entity information, spans, logs, and root-cause information, then returns key findings and recommended next steps. That context becomes part of a broader ServiceNow investigation, so the operator sees the issue in relation to the operational environment.
The experience shifts from:
Alert → swivel-chair investigation → manual correlation → decision
to:
Alert → AI investigation → enriched Dynatrace context → impact and root-cause insights → recommended next steps
That’s the real promise of agentic AIOps.
Keeping humans in the loop
[Figure 3: AI Activity tab showing steps in investigative analysis]
Autonomy doesn’t mean removing operators. AI handles repetitive investigative work while people keep ownership and visibility. Recent AIOps enhancements reflect this. The AIOps AI Specialist won’t override an alert already assigned to a human operator. And an alert handled by AI can return to its original operator after analysis or remediation, with the generated context and notes preserved.
The model is simple: AI gathers evidence, investigates, correlates, and recommends. People apply judgment where it matters. Over time, organizations decide which workflows to automate further.
Moving toward autonomous operations
The bigger story goes beyond a single integration. For years, the industry has talked about moving from reactive operations toward proactive and, eventually, autonomous ones. Getting there takes more than another algorithm analyzing telemetry. It takes connecting observability, service context, AI reasoning, and automation.
The expanded ServiceNow and Dynatrace collaboration aims at exactly that. ServiceNow’s announcement describes a vision where real-time observability and AI-powered IT service and operations management work together to create systems that learn, adapt, and self-heal at scale.
The AIOps AI Specialist and the Dynatrace MCP integration show what that future looks like today. Operators don’t need another screen full of data. They need an investigation that brings the right data, context, and expertise together and helps them decide what to do next. That’s where agentic AIOps gets exciting.
See it in action
The demo below shows how the AIOps AI Specialist works with Dynatrace in an AI-assisted operational investigation. It includes some customizations built with ServiceNow Otto that aren’t available out-of-the-box. However, they show what's achievable with the appropriate licenses and subject matter expertise.
