Be Aware
What to Watch Out For
AI is a powerful tool, but it is not perfect. Understanding its limitations helps you use it more thoughtfully and avoid common pitfalls. Think of this section as your "heads up" before diving in.
Where AI Can Help
AI tends to work well when it supports your learning process:
- Brainstorming and generating ideas to get started
- Organizing and structuring your thoughts
- Practicing and revising drafts
- Creating study questions and practice problems
- Getting a concept explained in a different way
- Improving the clarity and flow of your writing
- Carrying out repetitive manual tasks, aka busy-work
Where AI Falls Short
AI is not the right tool when it comes to:
- Doing the core intellectual work an assignment is designed to teach you
- Acting as a definitive source of truth or authority
- Making high-stakes decisions that require human judgment
- Providing ethical guidance or moral reasoning
- Replacing the care and accountability that come from human relationships
- Substituting for your own critical thinking and original analysis
Truth & Accuracy
Bias in Training Data
AI learns from the data it was trained on, and that data is not always balanced or representative. This means outputs can sometimes reflect biases you should watch for.
Hallucinations
AI can sometimes generate information that sounds convincing but is completely made up. Always double-check facts, especially citations and statistics, before using them in your work.
Unverified Sources
AI tools often pull from many sources without checking their reliability. If you rely on AI-generated information, take the extra step to verify it yourself.
Ethics & Rights
Privacy Matters
When you type something into an U of U approved AI tool, that data will not be used for training purposes. However, be careful not to share personal information, student records, or anything confidential. University data guidelines, including FERPA protections for student records and Rule 4-004C: Data Classification and Encryption, apply to AI use just as they do to any other tool.
Autonomy & Oversight
AI should support your decisions, not make them for you. Staying in the driver's seat means reviewing, questioning, and taking responsibility for the final output.
Accountability
If AI produces something inaccurate or harmful, the responsibility falls on the person who used it. That means you are accountable for anything you submit, even if AI helped create it.
Technical Limitations
Data Quality
AI is only as good as the data behind it. If the training data is incomplete, outdated, or biased, the outputs will reflect those shortcomings.
Context Gaps
A model trained in one setting may not perform well in another. AI might miss nuances specific to your field, assignment, or cultural context.