Skip to Main Content

Artificial Intelligence in Teaching & Learning

Information for Students and Faculty about ChatGPT and Other AI Tools

Use With Care!

When usingGenAI, it's important to verify everything. AI output can sound confident, but these tools can make up (“hallucinate”) or misrepresent information, draw false conclusions, make major mistakes, and generate fake sources. 

Keep these things in mind:

  • AI doesn’t “understand” the way humans do; these models lack real-world experience and context, so they don’t easily handle irony, humor and complex metaphors.
  • Don’t just read AI-generated summaries; take time to read original articles and understand detailed points and context. 
  • Some AI tools are trained on information only up to a certain date and may not have access to recent events or new discoveries.
  • Challenge AI responses and require the AI to justify its output by citing sources and data.
  • Guard against over-reliance on AI; challenge yourself to learn and exercise your mental muscles.

Checklist for evaluating AI-generated content

Accuracy and source check

❏ Verify all facts, statistics, and data points with multiple reputable sources (academic journals, government publications, respected news organizations); don’t rely on a single source.

❏ If the AI cites sources, check those sources directly; if the AI cannot cite a source, disregard the information.

❏ Watch for outdated information; confirm the publication date of source materials.

❏ Try the same prompts in multiple AI tools and traditional search engines and compare the results.

Bias check

❏ Ensure the content presents multiple perspectives and does not omit certain viewpoints or promote stereotypes.

❏ Check cited sources to see if they favor a specific viewpoint, ideology or group.

❏ Use critical thinking to evaluate information that seems slanted to serve business, government or advocacy interests, or influence buying decisions.

Logical consistency check

❏ Ensure the arguments being presented flow logically and make sense.

❏ Watch for broad generalizations from limited evidence.

❏ Look carefully for contradictions or misleading jumps in reasoning.

❏ Assess the depth of reasoning and avoid simplistic analysis.

Emotional and manipulative language check

❏ Look for content that uses neutral, fact-based language; be wary of language that is overly dramatic or inflammatory.

❏ Watch for loaded words designed to provoke emotions (fear, anger, excitement) or influence opinions or actions.