The purpose of this site is to provide curated information, resources, and examples of AI and how it might be applied in teaching and learning. It does not address important foundational questions about the role of AI in teaching and learning at the institution, its specific role in your course or program (if any), policy questions regarding the use of AI at Waterloo, or broader ethical questions – which are being discussed both within Faculties and centrally.
AI Basics
What is Artificial Intelligence (AI)?
- AI is an umbrella term that describes the ability of computer systems or algorithms to mimic human intelligence in a variety of ways, such as recognizing speech, solving problems, and identifying patterns in and learning from data.
What is Generative AI (GenAI)?
- GenAI is a type of artificial intelligence that can create new content, including text, images, videos, or code, by identifying and extrapolating patterns from its training data. GenAI generates this output through user inputs called “prompts."
Important note: The quality of user prompts directly impacts the quality of GenAI’s outputs.
How is AI Impacting Students and Their Learning?
- Ongoing research is investigating the benefits and risks of AI usage by students, including impacts on their learning. AI usage can assist students with various tasks (e.g., provide examples for self-testing, give immediate feedback, enable more complex tasks). However, when students offload cognitive processes to AI, they are not learning those processes or building persistence. Such results suggest carefully considering when and how AI usage can support (or detract from) the intended learning outcomes of your courses.
- To learn more: See the Digital Education Council Global Survey 2026 and explore research on topics such as cognitive offloading, including Artificial Intelligence, Cognitive Offloading, and Implications for Education (2026)
Starting Points for Instructors
An easy entry point for learning about how to use AI in teaching is to use it for developing teaching materials or supporting administrative processes. You can also take steps to mitigate or integrate AI use in teaching activities and learning assessments. While this resource shares a lot of information, explore, adopt, and build on what resonates in your context. Regardless of how you plan to use AI, apply the UW Principles of AI Enablement, developed in consultation with the Teaching Fellows.
The table below presents some starting points for responding to AI in ways that align with the UW Principles.
Principle
Purpose first: AI that matters
Effective teaching and learning involve human interactions and cognitive engagement with appropriately challenging material.
What to do
- Use AI in ways that supports but doesn’t replace learning (e.g., brainstorming, checking grammar, generating practice questions), (known as “beneficial cognitive offloading”).
- Review course learning outcomes to identify where students should demonstrate skills or knowledge without AI, and where AI could strengthen learning.
- Provide opportunities for students to develop and demonstrate foundational knowledge, critical thinking, problem-solving, communication, and other core disciplinary skills without the ability to offload these skills to AI.
What to avoid
-
Overloading students as this may lead them to outsource the cognitive effort necessary for learning (known as “detrimental cognitive offloading”).
-
Using only unsupervised assessments without verifying that students have done the effortful work of learning themselves.
Principle
Safe & Fair Use: Responsible, ethical, inclusive AI
Consider the security of student data, course materials, and other intellectual property when selecting an AI tool.
What to do
-
Use only UW-approved AI tools for restricted information according to Policy 46, such as student academic information (e.g., assigned grades, course enrolment info, completed coursework, student IDs).
-
Ensure that unlicensed AI tools are reviewed for security, privacy and potential risks; to do this, complete an Information Risk Assessment and Privacy Impact Assessment (IRA/PIA) form.
What to avoid
- Requiring students to use third party/ unlicensed AI tools (you must provide an opt-out for third party tools).
- Prohibiting all technology in the classroom: some students require technology for accommodations.
Principle
Human agency: People in control
Responsibility for teaching and learning remains with instructors and students.
What to do
- Evaluate and verify all AI-generated content, including learning outcomes, course slides, learning assessments, and grading schemes.
- Instructors may use UW-approved AI tools to assist with feedback and the assessment of student work. A human grader must review and verify the output and exercise independent academic judgement. The instructor remains responsible for the final feedback and grade and should inform students when AI is used in the assessment process.
- Communicate expectations for student use of AI clearly on course outlines and in assignment instructions.
- Educate students about when and how to use AI effectively
- E.g., Remind students that they should not upload course materials into unlicensed AI tools or even into licensed AI tools without permission from their instructor, owing to intellectual property and copyright concerns.
What to avoid
- Using AI without disclosing your use in course design and the creation of teaching and assessment materials. See the following wording samples:
Principle
Shared learning: Empowering our community
It is important that both students and instructors develop skills to confidently and critically engage with AI.
What to do
- Engage with colleagues, read various resources, participate in webinars and workshops.
- Inquire about your program’s curricular approach to AI.
- If permitting or integrating AI use in an activity or assessment, provide opportunities for students to critically reflect on and evaluate AI-generated outputs.
What to avoid
- Assuming that students know what appropriate and inappropriate uses of AI are for learning within your discipline. (Make these explicit.)
Principle
Exploration: Opportunity to experiment
Instructors are encouraged to investigate, evaluate, and innovate, and to share their findings and new approaches.
What to do
- Share your examples [form] of how you’ve used or are responding to AI in your teaching.
- Explore sample prompts for
- Create more effective prompts by using a UW-tested prompt formula: Prompt like a Pro.
What to avoid
Learn More
AI Tools for Teaching
UW-Approved AI Tools
- Confirm whether an AI tool you’d like to use is UW-approved for the type of university data you intend to use. A UW-approved AI tool is licensed by the university and has been through an Information Risk Assessment and Privacy Impact Assessment (IRA/PIA) process. Currently, Waterloo has enterprise licenses for Microsoft Copilot and Anthropic Claude, which include privacy and data protection.
Important note: To use UW-licensed versions of Microsoft Copilot and Anthropic Claude, you must request a license from UW to ensure that these services are used in the most secure manner.
- To learn more:
- Visit AI Tools and Guidance for a list of UW-approved tools (and to request a license to use these), as well as a list of problematic (high risk/unsafe) tools to avoid.
- Read Getting Started with Microsoft Copilot chat
Unlicensed AI Tools
- Ensure that unlicensed AI tools are reviewed for security, privacy, and potential risks. To do this, complete an Information Risk Assessment (IRA) and a Privacy Impact Assessment (PIA) application form.
- To learn more:
- Read about the IRA/PIA Approval and Support process
- Submit an Information Risk Assessment and Privacy Impact Assessment (IRA/PIA) application form
- Read about AI System Deployment requirements when developing or deploying a system that uses AI as part of its processes
- Learn about Safeguards for working with AI tools
- To learn more:
- Ensure that students can opt out of using unlicensed AI tools. You cannot require students to accept a third party’s terms or to relinquish their personal data or intellectual property. If the tool is part of a required activity or assessment, you should provide a viable alternative for students who object to using it.
- To learn more: Read Waterloo’s Guidelines for Cloud-Based Educational Technologies, which includes the following principle:
“Most cloud-based tools require users to accept their Terms of Use (usually by clicking a box). If a student is not comfortable accepting these, the instructor should provide an alternative assignment that does not require using the cloud-based tool.” (General Principle #3)
- To learn more: Read Waterloo’s Guidelines for Cloud-Based Educational Technologies, which includes the following principle:
Important note: If you’re unsure whether an AI tool is safe to use with students (e.g., whether it has completed the IRA/PIA process), or want to learn more about an AI tool, or want to request licensing, submit a ticket to IST’s AI Discovery/Intake Help Portal.
Waterloo Principles, Policies, and Guidelines
UW AI Principles
- Become familiar with principles guiding the use of AI at Waterloo. These are not rigid rules but reflect a commitment to use AI responsibly and in alignment with the university’s mission.
UW Policies
- Become familiar with relevant Waterloo policies: