Skip to Main Content

Ethical AI

Learn about how to ethically use AI tools in your schoolwork.

Artificial Intelligence (AI)

Artificial Intelligence (AI) is the use of computers to mimic or model human thinking, reasoning, learning, and related behaviors. Generative AI (Gen-AI) is a subset of artificial intelligence that generates content based on the tool's knowledge base.

AI tools can be very useful, but AI-generated content also comes with challenges related to copyright and plagiarism.

What "counts" as AI?

AI encompasses a wide range of tools including (but not limited to):

  • ChatGPT
  • Bard
  • Gemini
  • Grammarly
  • DALL-E
  • Quillbot
  • Character.AI
  • CapCut
  • Google Translate
  • Artemis Creator

The Bad

When is AI plagiarism (or unethical)?

  • Coping/pasting AI-generated written content and submitting it (with little or no changes)
  • Getting citations (without actually finding, reading, and evaluating the sources)
  • Using AI-generated summaries of a source in place of actually reading the source

AI can be bad at...

  • Context - understanding who your audience is and what you can expect them to already know (background)
  • Emotions & Culture - understanding nuances of human-to-human communication
  • Bias - may inherit any biases present in the data/content used to train it

The Good

What are some ethical uses for AI in education?

  • Using AI to help with brainstorming topics, subtopics, and keywords
  • Checking your understand of dense academic material*
  • Brainstorming a title for your paper, speech, presentation, etc.
  • Checking for typos or grammatical errors

*Be careful! AI may misinterpret things since it can't always recognize nuance.

The In Between

Even when ethically using AI, these tools can still come with unintended problems.

For example, asking a gen-AI tool to summarize an article can be helpful to make sure that you understood the main points from the article you just read. However, this isn't foolproof! Here's an example where ChatGPT completely missed the point when summarizing this article:

Khoddami S, Kiser MC and Moody CM (2023). Why can’t we be friends? Exploring factors associated with cat owners’ perceptions of the cat-cat relationship in two-cat households. Front. Vet. Sci.

ChatGPT's Summary

This article examines the potential applications of artificial intelligence (AI) in veterinary medicine, particularly in diagnosing diseases and predicting treatment outcomes for animals. It highlights how AI tools, such as machine learning algorithms, can analyze large datasets to identify patterns and provide valuable insights for veterinarians. The authors discuss various AI techniques, including deep learning and natural language processing, and their potential to revolutionize veterinary practice. They also address challenges such as data quality and ethical considerations, emphasizing the need for collaboration between veterinarians and AI experts to maximize the benefits of these technologies in animal healthcare.

My Summary (quoting a succinct statement from the article)

“Our survey results show that the cat-cat relationship in two cat households is complex and impacted by many factors such as, cat sex, age, relatedness, outdoor access, resource provision in the home, and aggression directed toward other people and animals in the home.” (Khoddami et al. 8)