The Five Levels of IT Autonomy
A New Framework for Evolving ITSM and ITOM with AI Agents
The evolution of automation in the automotive industry is famously captured by the Six Levels of Vehicle Autonomy from Level 0, where the human driver retains full control, to Level 5, where the system takes complete responsibility for navigation and safety.
Inspired by this structured framework, we believe that agent-powered automation in IT Service Management (ITSM) and IT Operations Management (ITOM) deserves a similar taxonomy. Such a model helps organizations assess where they are today and chart a roadmap toward a more intelligent, proactive, and business-aligned IT function.
Levels of ITSM/ITOM Automation
Level 0: System of Record
The foundation of modern IT. At this stage, platforms act purely as repositories capturing and storing business-critical data such as asset inventories, user access rights, and incident logs. There is no automation beyond basic data entry and retrieval.
Level 1: Help Desk Automation
Automation begins at the employee-facing layer. The system provides:
Self-service for routine inquiries
Knowledge base for how-to guides (e.g., "Why isn't my Microsoft Team microphone working?")
Ticket deflection using chatbots or guided troubleshooting
This reduces the burden on human agents by handling simple, repetitive questions.
Level 2: Ticket Workflow Automation
Automation extends to task execution across tools and systems. IT workflows are codified and delegated to agents or scripts that:
Provision/de-provision access
Initiate device setup via tools like Jamf or Microsoft Intune
Integrate with SaaS and cloud systems
Examples:
"Remove John's access to email and Slack."
"Provision a laptop with standard apps for Jane and ship it to her home."
Level 3: Business-Aware Automation
At this level, automation becomes context-aware and business-aligned. It doesn't just perform tasks it understands their broader impact. Examples:
"When deactivating John's email, reassign upcoming sales meetings to a colleague."
"Verify shipping address against a government ID before dispatching company hardware."
Here, the system shifts from automation of tasks to guardianship of business continuity, security, and compliance.
Level 4: Proactive Business Optimization
Imagine a world where IT bandwidth is unlimited. Level 4 agents:
Continuously monitor systems and user behavior
Predict and preempt issues before they are reported
Auto-resolve configuration drift, expired access, or underused services
This level transforms IT from a reactive support function into a strategic enabler identifying and resolving friction before it disrupts productivity.
Key Observations from the Model
System of Record vs. System of Action: The transition from Level 0 to Levels 1–4 mirrors the shift from passive record-keeping to active execution. This echoes a broader trend in enterprise software moving from data storage to intelligent action.
Automation of Today vs. Reinvention for Tomorrow: Levels 1–2 represent automation of existing workflows. In contrast, Levels 3–4 reimagine what IT operations could be leveraging AI agents to deliver outcomes that are infeasible with human analysts alone.
Human-in-the-Loop Collaboration: Unlike autonomous vehicles that aim to eliminate the driver, agentic IT systems augment human analysts. Collaboration, escalation, and approval workflows remain central throughout all levels ensuring trust, control, and compliance.
Final Thoughts
While inspired by the automotive model, our framework highlights a critical distinction: IT automation isn't just about replacing human effort it's about amplifying business value. As organizations mature along this spectrum, they unlock faster, safer, and smarter operations while empowering IT teams to focus on strategic initiatives, not busywork.