Campus AI workflow
PilotCourse assistants that answer from the material an instructor approved, inside the limits that instructor sets.
- 81%Helped understandingSaid it helped explain concepts
- 86%Ease of useFound it easy to use
- 67%Want moreWant it in more courses
Accountability
How this use case is governed
- Service owner
- Participating instructors and the TritonAI instructional AI team
- Human oversight
- Instructors define the learning purpose, content boundaries, and acceptable use.
- Measurement plan
- Student usefulness, learning evidence, answer quality, instructor workload, and escalations.
- Data boundary
- Public description; course data follows approved instructional controls
Product media
See the workflow in action
- Course-grounded AI Tutor
- Instructor-selected sources
- Canvas access
How it works
How the workflow fits
The problem it addresses, what it actually does, and how far along it is.
Problem
A general AI tool knows nothing about the course goals, the assigned reading, or what the instructor considers acceptable. Students need to know where the line is and how to check what they get back.
Solution
Participating courses can test assistants grounded in approved course material. Instructors determine how the assistant fits the learning design and communicate when and how students may use it.
Current status
Instructional AI is available only in participating pilot courses. We are evaluating learning quality and equity, along with how many students use it.