Overview
U-M Maizey is a tool that enables U-M faculty, staff, and students to use custom datasets to enhance their GenAI experience, helping them extract insights, discover patterns, and gain deeper knowledge from the data.
What’s New - May 2026
Maizey now enables users to create multiple AI experiences from a single dataset through a new capability called “Apps.” At launch, Apps will include a chatbot experience similar to what exists today, but with the ability to tailor multiple chatbots for specific audiences or purposes, such as separate versions for students, faculty, or staff.
This means:
- One project = your data and configuration
- Multiple apps = different chatbot experiences built on that data
Understanding how to use U-M Maizey effectively will give you the most insight into your data. Current and New Maizey Users: Explore the use cases and walkthroughs below to help you get the most out of Maizey's new capabilities.
How Maizey Works
1. Create a Project (Your Data Layer)
A project is where you:
- Connect and manage your data sources
- Configure access and permissions
- Maintain your dataset over time
2. Create One or More Apps (Your Chatbot Experiences)
Apps are the user-facing chat experiences built on your project.
Each app can have:
- Its own tone, personality, and custom instructions (system prompt)
- A specific audience or purpose
- A unique shareable URL
Why This Matters
This new structure allows you to:
- Use one dataset for multiple purposes
- Create tailored experiences for different audiences
- Avoid duplicating projects
- Maintain consistency while customizing behavior
Managing Your Project
You can:
- Add or update data sources at any time
- Reindex data when content changes
- Schedule indexing on a monthly or weekly basis
- Modify permissions and access
- Create, edit, or remove Apps without affecting the dataset
Videos: Use Cases and Walk-Throughs
What are Maizey Apps and How Do They Change U-M Maizey?
Walk-Through, Part One: How to Make a Maizey Project
Walk-Through, Part Two: How to Make a Maizey App Interface for Your Project
Sensitive Data
Check the ITS AI 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 Services.
Before using institutional and research data with third-party AI services, carefully review the ITS Safe Computing guidelines.
Get Started with U-M Maizey
Instructional Content
- Getting Started with Maizey (for Dropbox, Google Drive, public website)
- Getting Started with Maizey Canvas Connector
- Edit Maizey Project Settings
- Project Name, Billing Shortcode, Private/Non-Private project
- Edit Maizey App Settings
- App Title, App Path, App Description, Viewer Group, Publish/Unpublish App
- Edit Maizey App AI Settings
- Conversational, Return Data Sources, Temperature, System Prompt, Chunks, Debugging Log
- Guidance on AI Settings in Maizey
- Supported File Types
- Change Data Chunks Setting
- Delete a Maizey Project
- Export a Chat Conversation
- Reindex Data in Maizey
- Google Drive as a Data Source in U-M Maizey
- Enable/Disable Debugging Log
- Using Maizey with Canvas and CAEN Lecture Capture
- Maizey System Prompt Library
- Resolve a 500 Error when Adding Google Drive as Data Source
