AI Workflows (powered by n8n)
AI Workflows (powered by n8n) is a low-code platform that helps you automate complex tasks by connecting your data, systems, and AI into a single, end-to-end process. It is available to all active U-M faculty and staff on the Ann Arbor, Flint, Dearborn campuses, and Michigan Medicine.
It allows users to design, build, and operate automated workflows that connect campus and digital tools and university services, reduce manual effort, and support decision-based processes.
Now, instead of manually moving information between systems or repeating the same steps, you can build a “workflow” that does it for you. These workflows can pull in data, make decisions using AI, and take action, such as updating a spreadsheet, sending an email, or triggering another system.
Using a visual interface, AI Workflows supports both straightforward automations and more advanced AI-assisted workflows.
Visit our Video Use Cases and Walkthroughs section for detailed overviews and examples of what you can do with AI Workflows.
Getting Started
You do not need to code to get started with AI Workflows, but more advanced use requires some technical familiarity. Users should be comfortable thinking step-by-step and, over time, working with APIs, JSON, and data.
In practice, AI Workflows works best for users who are comfortable thinking through processes step-by-step and are open to learning some technical skills along the way.
To gain initial access to the service, all users must complete the AI Workflows Training Canvas course (approx. 30 min). Once you complete the course and your request is processed, you will gain access to the AI Workflows environment.
See more information about getting started with AI Workflows.
What You Can Do with AI Workflows
AI Workflows enable teams to replace repetitive, multi-step processes with reliable, automated workflows.
For example, a workflow can take a process that once required hours of manual effort—gathering data, analyzing it, and formatting results—and complete it automatically in minutes, while still providing transparency into how decisions were made.
AI Workflows is especially useful for processes that are:
- Multi-step
- Spread across multiple systems
- Dependent on human judgment or interpretation
By combining automation with AI, it helps units work faster, reduce manual effort, and deliver more consistent results
Common use cases include:
- Research: Monitor RSS feeds for new publications, summarize content, and deliver results to Slack.
- Administrative workflows: Trigger approvals, notifications, and updates when forms or documents are submitted.
- Operational monitoring: Send alerts or take action when systems, sensors, or thresholds reach defined conditions.
- Service processes: Route requests, enrich data, and support triage across ticketing and intake systems.
Data Protection and Privacy
The U-M Data Classification Levels define four classifications (sensitivity levels) for U-M institutional data. ITS AI Workflows is suitable only for institutional data in the Low and Moderate sensitivity categories. Check the ITS AI Workflows Service entry in the Sensitive Data Guide to IT Services for a list of the data types that are permitted to be used with ITS AI Workflows Service.
For example, FERPA data is classified as Moderate and may be passed to AI workflows. HIPAA data, however, is in the high-sensitivity category and may not be used in AI Workflows.