AI and Your Teaching Practice
Join us, from 10:00am until noon, 11 January, 2024, for a semester-start discussion on generative artificial intelligence. This session will offer an overview of current efforts at UAF to integrate AI with teaching and learning, as well as share resources available to faculty and your students. We will look at example assignments that help introduce you to working with generative AI, and hear from the Student Success Center about available tools for students. Along the way, we will continually examine how generative AI relates to assessment and to academic honesty.
Agenda
| Time | Title | Facilitator | Description |
| 10:00AM – 10:15AM | Welcome & Overview | Dan LaSota | Welcome to the session and Sign in |
| 10:15AM – 10:30AM | Intro to Generative AI with Images | Christen Booth | We’ll look at AI tools for text-to-image creation, prompting strategies, ethical use and bias considerations. We’ll dig into one of the tools with an icebreaker activity. |
| 10:30AM – 10:45AM | Resource Verbiage | Nathan Feemster | Is AI use by students a breach of Academic Conduct? Well, it depends. In this session we will look closely at UAF’s Academic Conduct guide and outline how it pertains to AI use in a course. |
| 10:45AM – 11:00AM | Syllabus Example | Dr. Karen Nielsen knielsen4@gsu.edu | Review of example AI permissible Syllabus policy from Dr. Karen Nielsen, Assistant Professor of Biostatistics, Georgia State University. |
| 11:00AM – 11:15AM | Assignment Example | Dr. Lane Schwartz (Sugri) | UAF Associate Professor of Computer Science Dr.Lane Schwartz will walk us through an assignment he created and ran with his Computer Ethics course Fall 2023. |
| 11:15AM – 11:30AM | Student Success Center | Jennifer Tilbury | UAF Associate Vice Provost for Student Success Jennifer Tilbury will overview the Student Writing Center’s resources for using AI with your students, and tips for incorporating it into your class. |
| 11:30AM – 11:50AM | Hands-On Q&A Forum | Dan LaSota | Participants are challenged with three scenarios and they group-think some approaches. |
| 11:50AM – 12:00PM | Closing & Semester Programming | Nathan Feemster | Thank you for joining! This semester, we have a once a month meeting called “What’s New with AI” where we discuss the most pressing AI news. Come join us! |
Session Recording
Presentations slides and resources
An Introduction and Prompting Visual Generative AI
Turning Ideas into Images with Visual Generative AI
- Myriad tools are available for creating images. The most common include Firefly (Adobe), DALL-E (OpenAI), Imagine (Meta), and Bing Image Creator (Microsoft).
- Strategies for writing prompts for visual generative AI varies from platform to platform. Using the same prompt an different platforms results in wildly different results. In general,
- avoid prompting AI for work in the style of a living artist.
- craft an effective text-to-image prompt by stating purpose, lighting, broad stylistic direction, composition, and content. You are also able to prompt using your own imagery.
- revise your prompt to refine the image further. Incremental prompt revision allows you uncover effective strategies for the AI you’re using. It often results in a higher-quality image.
Examples (same prompt, different platforms)
DALL-E
PROMPT: I want a spirit animal that is an intersection of many animals. It has the head of an owl with one eye. Its body is that of a bear, and it has a feathery ridge that goes down the middle of its back.
Firefly
Compared to DALL-E, Firefly rendered an animal in a very different style even when given the same prompt.
PROMPT: I want a spirit animal that is an intersection of many animals. It has the head of an owl with one eye. Its body is that of a bear, and it has a feathery ridge that goes down the middle of its back.
PROMPT: I want a spirit animal in the style of a Dutch renaissance painting. It has the head of an owl with one eye, and the body of a bear. It has a feathery ridge that goes down the middle of its back.
Imagine
PROMPT: I want a spirit animal that is an intersection of many animals. It has the head of an owl with one eye. Its body is that of a bear, and it has a feathery ridge that goes down the middle of its back.
Bias and Ethical Use
- Who owns artwork generated by AI?
- Which LLMs work to get consent from artists when training their models?
- Adobe Firefly
- Loudly
- How inclusive are the images generated by AI?
- What is AI’s role in spreading disinformation?
Icebreaker Activity
Consider the three activities below and select one. Use an AI you can easily access (Firefly, DALL-E, Imagine , or Bing Image Creator, for instance) to generate your creature/character.
Spirit Animal
Think of an animal that most embodies your personality. Consider the visual aspects of this animal and describe it to a visual AI so it can render an image. If it doesn’t render the animal you imagine in your mind, revise your prompt to be more specific.
Pet Personality
Think of a pet you have or have had in the past. What parts of that animal’s personality really stood out? Describe your pet so that an AI can create a visual caricature. Revise your prompt if the AI isn’t rendering the image you want.
Create a Character
Create a totally new character. It can be an animal, a person, a robot, or combinations of any of these. Think of the world this character lives in, what they wear, how they communicate, and aspects of their personality. Ask the AI to create an image of this character by describing aspects of its appearance and personality.
Resource Verbiage (Syllabus and Policy)
How is AI usage in a course connected to Academic Misconduct?
Based on UAF’s Academic Misconduct Policy, there are a few ways that AI usage could be considered Academic Misconduct. The UAF Office of Rights Compliance and Accountability states the following clauses as pertinent to a case of AI usage:
UAF ORCA – Academic Misconduct
- utilizing devices not authorized by the faculty member;
- using sources (including but not limited to text, images, computer code, and audio/video files) not authorized by the faculty member;
- acting as a substitute or utilizing a substitute;
- deceiving faculty members or other representatives of the university to affect a grade or to gain admission to a program or course;
- violating the ethical guidelines or professional standards of a given program.
This boils down to
- Unauthorized usage
- Deception
- Program specific violations
What happens if I suspect a student of using AI in a way that violates the UAF Code of Conduct?
Before reaching out to a student, clearly identify the work in question, why you believe it to be a violation, and prepare examples of evidence as outlined in the quote below.
“In order to determine a violation occurred additional evidence of unauthorized use of AI must be available. Examples of evidence include: other writing examples, unexplained advanced techniques, prior work completed by the alleged individual, admissions of use, etc.”
UAF Academic Misconduct Policy
The UAF Office of Rights, Compliance, and Accountability (ORCA) can help facilitate this conversation and their CARE team can provide support for students. Depending on this conversation, the instructor may decide to file a case of Academic Misconduct. Both parties will then work with the UAF Office of Rights, Compliance, and Accountability (ORCA) to follow the misconduct process.
What are some tips for my online students, when it comes to AI usage and my education?
Review the following Blog post by UAF Instructional Designer Dan LaSota – Using AI as a Student
- Written assignments are potentially going to come under closer scrutiny. Working in a word processor that tracks your version history gives you a way to show instructors your writing process. Try using a tool like Google Docs.
- If you are unsure if AI is allowable in a way you are considering using it, check with your instructor via their preferred method of communication.
- Try using AI tools outside of your course work to better understand what they are capable of. There is a lot of information out on how AI is impacting our present day, and it isn’t all accurate. Getting first-hand experience is valuable.
- Have patience with your instructors. They are doing the best they can to adapt to a technology that upends conventional teaching methods.
Syllabus Example
Review of example AI permissible Syllabus policy from Dr. Karyn Nielsen, Assistant Professor of Biostatistics, Georgia State University.
Assignment Example
UAF Associate Professor of Computer Science Dr.Lane Schwartz will walk us through an assignment he created and ran with his Computer Ethics course Fall 2023.
Student Writing Center
UAF Associate Vice Provost for Student Success Jennifer Tilbury will overview the Student Writing Center’s resources for using AI with your students, and tips for incorporating it into your class.
Closing & Semester Programming
Thank you for joining! This semester, we have a once a month meeting called “What’s New with AI” where we discuss the most pressing AI news. Come join us!
Related content and resources
- What do we do now?
- What are the longer form opportunities?
- Faculty Alliance symposium (if this gets off the ground)
- 1/18 What’s New in AI Session (these will be every month)
In addition, the following longer form programs were suggested:
- Book club – Cheating Lessons (I learned about this book after listening to the Teaching in Higher Ed podcast episode 481, “Assignment Makeovers in the Age of AI”
- FLC or something similar looking into Cheating
- Transdisciplinary iTeach looking into cheating and AI
