1 of 113

AI & Generative AI Curriculum, by Packback

Last Updated: August 2023

2 of 113

Table of Contents

3 of 113

About this Resource

This resources was created by the Packback team, specifically by former educators and current AI-fanatics.

This resource contains:

  • Recommendations and guidance for instructors who wish to teach their students about AI.
  • Four lessons you can (or must) customize and present to your students.
  • Activities you can assign to your students for pre-work, in-class activities, and out-of class homework.
  • Bonus: Ideas for how you can weave current Packback products (Instructor Questions, Polls, Deep Dives prompts) into these lessons.
    • Note: If you are not a Packback user, you should still be able to use the content outside of Packback in whatever method you use to facilitate discussion, polling, or writing assignments

Please Read before continuing:

  • Some of the slides in this deck are meant to be student facing. Some slides in this deck are recommendations for pre-work, facilitation, and follow-up activities. To make life easy, we’ve input the Packback logo on non-student facing slides! (Like this one!)
  • If you want to break down one lesson into multiple lessons - go for it! If you want to combine our lessons into one larger lesson - go for it! Use this resource however you see fit.
  • We’ll cite our sources, and we ask you to do the same if you use this resource.
  • Do not hesitate to share this resource with your peers. You have the authors’ full permission to share.

4 of 113

About the authors...

Kelsey Behringer

CEO, Packback

During her four years in the classroom as a high school chemistry & physics teacher, Kelsey found herself using technology daily to captivate her reluctant students. Looking to widen the scope of her impact on education and help other educators see technology as a superpower, she joined Packback in 2017 as an Experience Manager. It’s been a curiosity fueled journey since!

Ri Cantlin

Curriculum Consultant, Packback

As a former middle school ELA teacher, Ri was dedicated to finding innovative ways to engage in curriculum, amplify student voices and spark a love of reading and writing in every student. Now, as a Packback Curriculum Consultant, they teach higher ed faculty how to use AI-powered technology to do the same.

Much of the content in this guide was created with the collaboration of the generative AI Platforms: ChatGPT (by Open AI), Claude.AI (by Anthropic), and Perplexity.AI

5 of 113

Lesson 1: Introduction to AI

6 of 113

Lesson Objectives

Define artificial intelligence & common language you may hear in regards to AI, including: machine learning, large language models, “narrow” AI and “general” AI

Provide examples of AI in everyday life

Discuss the potential benefits and risks of AI

7 of 113

What is Artificial Intelligence?

8 of 113

Let’s ask AI, “What is Artificial Intelligence”?

Source: OpenAI, GPT-4

9 of 113

Examples of Artificial Intelligence in your Daily Life

10 of 113

Facebook’s Facial Recognition Applications

“Facebook trains DeepFace, a facial recognition programme, to identify people in pictures. According to the company, DeepFace, its most sophisticated picture recognition engine, has a success rate of 97% in determining whether two images of the same person are identical or not when compared to humans' 96%.”

Source - Ashesh Anand, AnalyticsSteps.com

11 of 113

Uber - More Accurate Predictions, Better Navigation

“In 2019, Uber AI’s Sensing and Perception team worked on projects across our mobile and back-end stack to improve the coverage, accuracy, speed, and heading of vehicle locations on the Uber platform. Overcoming the limitations of GPS and having more precise locations makes it easier for riders and drivers to find one another, improves estimated times of arrival (ETAs), reduces rider and driver cancellations, and makes our marketplace operate more efficiently.”

Source (from Uber.com)

Image from article, “How Uber uses AI to serve you better

12 of 113

Let’s quickly review two buzzwords in AI:

  1. Machine Learning (ML)
  2. Large Language Model (LLM)

13 of 113

What is Machine Learning?

Machine learning (ML) is just one subset of Artificial Intelligence. ML gives computers the ability to learn without explicitly being programmed.

“Machine learning starts with data — numbers, photos, or text, like bank transactions, pictures of people or even bakery items, repair records, time series data from sensors, or sales reports. The data is gathered and prepared to be used as training data, or the information the machine learning model will be trained on. The more data, the better the program.

From there, programmers choose a machine learning model to use, supply the data, and let the computer model train itself to find patterns or make predictions. Over time the human programmer can also tweak the model, including changing its parameters, to help push it toward more accurate results.”

Source: MIT Sloan School of Management

Video: “Machine Learning Explained in 100 Seconds”, Fireship

14 of 113

What is a Large Language Model?

From the Webinar “Will AI-Generated Text Kill Writing?”, facilitated by Packback on 12/13/22 (Link to Video)

15 of 113

ChatGPT (LLM) In Action

16 of 113

Two more buzzwords you may be hearing…

“Narrow” vs. “General” AI

17 of 113

You may have heard of the terms “narrow AI” and “general AI”

In simple terms, Narrow AI is where we have been, and General AI is where we want to head towards.

“Artificial Narrow Intelligence (ANI), sometimes known as "weak AI", refers to any AI that can outperform a human in a narrowly defined and structured task. It is designed to perform a single function like an internet search, face recognition, chatbot, or speech detection under various constraints and limitations. It is the constraints that lead people to refer to these functions as ‘narrow’ or ‘weak’.”

“Artificial General Intelligence is known as "strong AI" and allows machines to apply knowledge and skills in different contexts.

Where ANI applications can run single, automated, and repetitive tasks, the objective of AGI is to create machines that can reason and think just like a human is capable of doing.”

Think of a robot from a Sci-Fi movie, like “Ava” from ‘Ex Machina’

Source - “General AI vs Narrow AI”, Zoe Larkin, Levity

18 of 113

AI: Benefits and Risks

Benefits

  • Automation & Productivity- AI can automate repetitive and time-consuming tasks, freeing up human time.
  • Personalization - AI can tailor services and recommendations to individual users.
  • Innovation - AI can come up with novel solutions to complex problems faster than humans.
  • Job Creation - New jobs can be created (i.e. prompt engineers)
  • Healthcare - AI is helping improve diagnostics, customize treatment plans, and automate tasks.

Risks

  • Job loss - AI automation may displace many jobs held by humans.
  • Bias - AI systems can perpetuate and amplify biases present in data or programming.
  • Lack of transparency - The decision making of complex AI systems can be opaque and hard to understand.
  • Security flaws - AI systems could be exploited by cybercriminals if vulnerabilities exist.
  • Lack of oversight - Without proper governance, AI may run unchecked in high-stakes situations.

“Global employers anticipate creating 69 million new positions by 2027 and eradicating 83 million jobs—a net loss of 14 million roles.”

List of Benefits & Risks created in collaboration with ChatGPT, OpenAI, Version 4

19 of 113

Lesson 1: Optional Pre-Work

20 of 113

3-7 days leading up to your first AI-Lesson, create a “Homework Poll” in your Packback community.

Ask them a multiple choice question that would require them to also write a written response.

Remember, a “homework poll” requires students to justify their multiple-choice answer with a written response. If you do not want to require students to follow up with a written response, use an in-class poll instead!

Here are a few examples to the right:

Optional Pre-Work (Packback User, Homework Poll)

Q: Are you concerned or excited about the evolution of AI and its impact on society?

A: I’m concerned

B: I’m excited

C: A little bit of both

D: It’s too soon to tell

Q: Have you used Generative AI. If yes, how? If no, why not? Please explain your personal sentiment towards the use of generative AI!

A: Yes, I’ve used it, and I use it often

B; Yes, I’ve used it, but I don’t use if often

C: No, I haven’t used it

21 of 113

Optional Pre-Work (Packback User, Homework Poll)

22 of 113

3-7 days leading up to your first AI-Lesson, use your LMS discussion board to facilitate a pre-lesson discussion, or create a “quick write” assignment for them to complete a quick, ~200 word reflection before class.

Discussion: Ask your students to reflect on their sentiment towards AI, familiarity with AI, and their predictions regarding AI. Ask them to reply to one of their classmate’s discussion posts, specifically, a peer who doesn’t share the same opinion or sentiment as them!

Writing Assignment: If you do not wish to use LMS discussion (this doesn’t fit in your course, or you do not want students’ responses to be visible by classmates), then create a quick writing assignment with a similar prompt as the discussion.

Optional Pre-Work (Not a Packback User)

Discussion/Writing Prompt

Before class on [date], I want to hear your thoughts & opinions on AI.

First and foremost, how have you been using AI? Have you used any of the chatbots? Which ones? What are your general observations on their performance and applications?

Second, I’d love to hear how you feel AI will personally affect your future - your dream job? Society? Your hobbies? Are you concerned? Are you excited?

Lastly, how would you like to use AI in this course?

Once you write your own reflection (I encourage you to share links and news stories, by the way!), reply to a peer. My one ask is you reply to a peer who has a different opinion than you!

Note: If you do not wish to use discussion but rather a standard writing assignment, simply remove the last call to action.

23 of 113

Lesson 1: Optional In-Class Activity

24 of 113

Use Polls to Gauge Interest & Maintain Engagement

If you’re a Packback user, use an “in-class poll”

If you’re not a Packback user, use your desired polling software.

Not a huge fan of using technology during class? No worries. Ask your students to raise their hands to answer the poll or get up and “go to a side of the room”

Note: Some of these poll questions overlap with the Homework Poll Questions presented in the “pre-work” assignment suggestion

Optional In-Class Activity (In-Class Poll)

Poll Questions

  • How often do you use generative AI like ChatGPT?
    1. Never
    2. Sometimes
    3. Frequently
  • Do you use generative AI to help you with personal tasks & decision making?
    • Yes
    • No
  • Do you use generative AI to help complete coursework? (I am going to assume the best here, and assume if you use it, you are not just copying and pasting!)
    • I have
    • I have not
    • I don’t wish to answer

25 of 113

Lesson 1: Optional Follow Up Assessment

26 of 113

After your first AI Lesson, assign a short writing assignment.

If you’re a Packback user, set this up as a “Deep Dive.

You could also assign this in the form of a discussion using the following prompt:

Ask 1 question about how your life may be impacted by existing AI, and respond to 2 peers with answers based on personal experience or examples of existing tools related to their question. Example:

Optional Follow Up Assessment (Writing or Discussion)

AI in Your Daily Life Writing or Discussion Prompt:

For this assignment, you will conduct research into how you already interact with AI on a regular basis.

Choose 3 examples of AI you encounter in your daily life, such as digital assistants, social media feeds, shopping recommendations, etc. For each example, write a paragraph addressing the following:

  • Name the AI system and its purpose (ex: Amazon's Alexa for a virtual assistant)
  • Describe how this AI works and makes decisions based on your understanding
  • Explain how this AI improves or enhances your life
  • Discuss any concerns you may have around bias, privacy, security, etc. related to this AI
  • How might this AI evolve in the next 5 years?

How could AI help me diversify what I eat each week?

Halla Taste Intelligence Engine is a human preference engine built for grocery stores. It leverages over 100 billion grocery shopper + product data points to accurately display products a shopper will want to buy next. https://halla.io/capabilities/

27 of 113

Lesson 2: How to Use Generative AI as a Collaborator

28 of 113

This lesson is meant to be a highly interactive “start and stop” lesson with students.

The way that this is built, students should be on their laptop/computer, with ChatGPT, Claude, or another chatbot open.

The general flow of the lesson is:

  • Present use case for AI chatbot, i.e. “Writing Assistant and Tutor”
  • Have students work through a scenario related to the use case
  • Facilitate a discussion & reflection after students work through the scenario

We do NOT recommend you work through all scenarios, but rather the scenario(s) that mosts relates to your course material.

Lesson 2 Notes

29 of 113

Lesson Objectives

Identify platforms available for written, visual, and audio AI-generation and collaboration

Review recommendations for writing high-quality prompts for genAI

Discuss specific use cases for AI utilization in an appropriate & academic way

Share examples of ways we can use AI for personal collaboration

30 of 113

Popular Generative AI Chatbots:

  • ChatGPT
  • Claude.AI
  • Perplexity.AI

Image Generator:

  • Dall-E2
  • Midjourney
  • Adobe Firefly

Sound Generator:

  • AuidoCraft
  • Otter.ai
  • Soundful

Popular Generative AI Platforms

What other generative AI platforms have you used? For what purpose?

31 of 113

We’re about to discuss ways you can use genAI chatbots in academically honest ways.

Please note while I may allow use of appropriate use of genAI, not all of your instructors will or do.

Please make sure to read all of your syllabi carefully before assuming how you may/may not use AI

32 of 113

How to write a strong genAI prompt

33 of 113

A helpful analogy

GPT is like a toddler, but if that toddler had almost all of the knowledge in the World.

What you need to know about this toddler:

  • It really wants to help
    • So much so it may produce inaccurate outputs just to appease your prompt!
  • It knows just about everything
  • It can’t filter relevant from irrelevant without your help

How would you coax an astonishingly knowledgeable toddler into doing what you want?

Image generated by DALL-E 2 “Super Knowledgeable Toddler”

34 of 113

  1. Define ChatGPT’s Role - Clearly communicate that ChatGPT should "act as" a certain role or possess a specific area of expertise. This sets the tone and ensures that the AI's response aligns with the intended perspective.
  2. Provide Context - Offer relevant background information to help ChatGPT understand the situation, setting, or topic at hand. The more context we provide, the better ChatGPT can tailor its response to our needs.
  3. State the Task or Question - Clearly articulate the specific objective or inquiry that ChatGPT is expected to address. Providing a well-defined target allows the AI to focus its response accordingly.
  4. Set Constraints or Limitations - Establish any boundaries or conditions that ChatGPT should consider while generating a response. These could include limitations in terms of response length, format, language usage, or content scope.
  5. Offer Additional Guidance - Supply supplementary instructions regarding the desired tone, formatting, or style of the AI's output. This ensures that the response aligns with our expectations and requirements.

5 Rules for Creating Great Prompts

35 of 113

Prompt Example

You are a logical fallacy detector. If there are clear logical fallacies in this text, list them and describe why.

Your output format is a csv with two columns, "fallacy" and "reason", with no other explanation.

[INSERT TEXT]

Define Role

Provide context

State the Task or Question

Set Constraints or Limitations

Offer Additional Guidance

36 of 113

Prompt Example, Refined

You are a logical fallacy detector. If there are clear logical fallacies in this text, list any of them that appear in this list

  • False equivalence
  • Ad hominem

Describe why in a way that is appropriate for a sixth grade class to understand.

Your output format is a csv with two columns, "fallacy" and "reason", with no other explanation.

[INSERT TEXT]

Define Role

Provide context

State the Task or Question

Set Constraints or Limitations

Offer Additional Guidance

37 of 113

AI Collaboration Use Case:

Generative AI as a Writing Tutor

38 of 113

Using Generative AI Chatbots as a Writing Tutor

When using AI Chatbots as a writing tutor, be mindful of violating academic integrity policies, and also remember the purpose of completing a writing assignment is to strengthen your understanding of the content, and help you build the skills necessary to be a strong researcher, writer, editor, and communicator.

If you lack strong fundamental research, writing, and editing skills, you will not be able to effectively use GPT as a writing collaborator.

Here are some ways you could use AI Chatbots as a Writing Tutor

  • Get feedback on your thesis statement
  • Ask for feedback on any structural writing errors or grammatical errors
  • Ask for feedback on the flow or tone of your writing assignment
  • Get feedback on an outline
  • Ask it for some help if you get writers block*

*just make sure you use it as inspiration, don’t copy & paste unless your instructor tells you otherwise

39 of 113

Writing Tutor - Scenario

You’re writing a paper on why University athletes should be compensated.

The thesis statement: University athletes should be compensated for their service to the university, as it is difficult for these students to procure and hold a job with busy academic and athletic schedules. Many student athletes on scholarship also come from low-income neighborhoods and it is a struggle to make ends meet when they are participating in athletics.

  1. (1-3 minutes) Using the framework previously shown, craft a prompt for GPT to get feedback on the quality of your this thesis statement
  2. (1-3 Minutes) Workshop the prompt a few times until you get the best feedback. You should NOT be asking GPT to re-write your thesis statement
  3. ( 5 Minutes) Finally, using the feedback from GPT, rewrite the thesis statement yourself

40 of 113

Writing Tutor - Discussion and Reflection

GPT Output when I ask for feedback:

Of course! Crafting a strong and engaging thesis statement is essential for capturing your readers' attention and guiding your paper effectively. Here are some tips to help you create a compelling thesis that makes your argument both strong and enjoyable to read:

  1. Be Clear and Specific: Make sure your thesis statement is clear and concise. Avoid vague language and get straight to the point. Clearly state your position on compensating college athletes and provide a brief overview of the main reasons you will use to support your argument.
  2. Inject Emotion or Intrigue: Use language that evokes emotion or curiosity. Consider incorporating strong verbs and impactful adjectives that resonate with your readers. For example, instead of saying "should be compensated," you could say "deserve fair compensation" to convey a sense of justice.
  3. Stir Curiosity: Pose a thought-provoking question or offer a surprising fact that sparks curiosity. This can encourage readers to continue reading to find out more. For example, you could start with a question like, "Have you ever wondered why talented college athletes struggle to make ends meet while their universities rake in millions from sports revenue?"
  4. Use Parallel Structure: Employ parallel structure to create a rhythmic and memorable statement. Parallelism involves using similar grammatical structures for related ideas. This can make your thesis statement more engaging and easier to remember.
  5. Incorporate Anecdotes or Stories: Weave in a relevant anecdote or story that illustrates the challenges faced by college athletes. Personal stories can make your argument more relatable and emotionally resonant.
  6. Address Counterarguments: Show that you've considered opposing viewpoints by briefly mentioning a counterargument and then immediately refuting it. This demonstrates your thoughtfulness and deep understanding of the topic.
  7. Vivid Imagery or Metaphor: Use vivid imagery or metaphors to paint a mental picture for your readers. Metaphors can help to make your argument more relatable and memorable.
  8. Use Strong Language: Utilize strong and persuasive language that conveys your passion and conviction. Avoid passive voice and opt for active voice to make your thesis more forceful.
  9. Revise and Polish: Crafting an effective thesis often requires multiple revisions. After you've written your initial thesis, review it and refine the wording to make it more powerful and engaging.

Remember, your thesis statement serves as a roadmap for your paper, so make it exciting and captivating to set the tone for an engaging read.

When I ask GPT to rewrite my thesis statement…

"In the heart of the roaring stadium crowds and the hushed libraries, a fundamental inequity persists: college athletes, who pour their sweat and dedication into both academics and sports, are left grappling with an unjust reality. The call for fair compensation reverberates through the labyrinthine schedules of these driven individuals, whose arduous academic and athletic commitments deny them the luxury of part-time jobs. Moreover, a significant number of these athletes emerge from the crucible of low-income neighborhoods, their dreams colliding with the harsh financial constraints of university life. This essay champions the urgent need to rectify this imbalance, proposing that compensating college athletes is a moral imperative that speaks not only to the essence of justice but also to the true spirit of collegiate athletics."

If you copied & pasted the new thesis from GPT instead of following the feedback and writing it yourself, what would you have learned in the process? Do you think you’d write an better paper using your own thesis or the GPT-thesis? Why?

41 of 113

AI Collaboration Use Case:

Generative AI Chatbots for Professional and Academic Advice

42 of 113

Using Generative AI Chatbots for Professional and Academic Advice

AI Chatbots can be incredible resources for writing intimidating pieces of content like cover letters and resumes.

I recommend not asking the chatbot to write these for you, but rather:

  • Give the Chatbot strict guidelines (research rules about recommended format and tone)
  • For something like a cover letter, feed it information about the organization and the role
  • Provide it your first version/draft
  • Ask for feedback!

Future employers want to hear your voice. I recommend only asking for feedback from a chatbot and perhaps asking for ideas on how to phrase/rephrase portions of text. Why? Nobody wants to read something written by a robot.

Make sure to fact check and thoroughly review any content you do get from the Chatbot. I’ve personally found myself editing 25%-50% of any text generated by AI for accuracy, clarity, or tone.

Here are some ways you could use AI Chatbots for Professional and Academic Advice

  • Get feedback on your resume or a cover letter
  • Get ideas on how to rewrite portions resume or cover letter
  • Ask for feedback on any structural writing errors or grammatical errors
  • Ask for feedback on emails (to a future employer, recruiter, academic advisor, professor)
  • Ask for help writing an email

43 of 113

Using Generative AI Chatbots for Professional and Academic Advice - Scenario

You’re interviewing for a post-graduation job at a tech company in a sales role. You have completed two interviews thus far, one with their recruiter, Maggie, and the next with the team manager, Anne. Your interview with Anne was on Monday, it’s now Thursday and you haven’t heard back on next steps.

You craft an email for the recruiter and manager:

Dear Maggie and Anne,

I'm following up on my interview from Monday. I know it's only been a couple days, but I just wanted to check in on the status of the hiring process. Do you have an update on next steps or a timeline for when I can expect to hear back?

I enjoyed learning more about the company and position. This is my top choice for a job right now, and I’m looking forward to hearing from you soon!

Best, [Your Name]

  1. Use ChatGPT to get feedback on your original copy.
  2. Then, re-write the email incorporating the feedback.
  3. Be ready to share the prompt you used to get feedback, the feedback you received, and your updated email copy!

44 of 113

Using Generative AI Chatbots for Professional and Academic Advice - Example One

45 of 113

Using Generative AI Chatbots for Professional and Academic Advice - Example Two

46 of 113

Using Generative AI Chatbots for Professional and Academic Advice - Scenario Reflection

Share your prompt, 1-3 pieces of feedback you received from ChatGPT on the email copy, and describe how you approached your edits

47 of 113

AI Collaboration Use Case:

Generative AI Chatbots for Content Tutoring

48 of 113

Using Generative AI Chatbots as a Content Tutor

Sometimes concepts are just hard to understand. And when you go to search more about the concept in your textbook or on the web, you run into even more vocabulary or concepts you don’t understand. It can be frustrating to hit a wall.

This is where AI can be useful. Use generative AI to break down content, ask for analogies, walk you through a process, and more.

JUST be sure to be cautious - AI isn’t always right. You’re going to need to be cautious to not accept content from AI at face value. Generative AI should be one tool you use to help understand difficult concepts. Rely on text, course resources, ME!, and office hours. You have many tools at your disposal.

49 of 113

Using Generative AI Chatbots as a Content Tutor

Example of AI as a content tutor!

50 of 113

Using Generative AI Chatbots as a Content Tutor - Scenario

  1. (First 1-3 minutes) Reflect on the last content-related question you had in this class, or concept you had a hard time understanding.
  2. (Second 1-3 minutes) Ask ChatGPT to answer the question you had OR to help explain the concept
  3. You’ll share your prompt & chatGPT’s output, and I (the content expert) will fact check and rate the output!

51 of 113

Examples of AI as a Personal Assistant and Collaborator

52 of 113

Ask generative AI to help create a study plan

53 of 113

Ask generative AI to help come up with Vacation Ideas

54 of 113

Ask generative AI to help you make a decision

55 of 113

Lesson 2: Optional Pre-Work

56 of 113

Right before you start this lesson, have your students ask ChatGPT (or a chatbot of your choice) a simple question related to your class.

This could be done synchronously in-class or asynchronously prior to arriving to class.

For example: “What are the best jobs in [discipline]?”

Ask students to share out the responses. Note: assuming you encourage them to all use the same chatbot, their answers will all be similar, if not the same.

Optional In-Class Follow-Up Activity:

Have students brainstorm how they could refine the original prompt to provide a better answer (unique to them).

Optional Pre-Work

“Chatbot Prompt Practice” Prompt

Emerging chatbots like ChatGPT demonstrate the power of large language models to generate human-like text. However, the quality of their output depends heavily on how well we frame our prompts.

In this pre-work assignment, you will practice prompting an AI assistant to give you career advice.

Go to the ChatGPT website or another chatbot of your choosing. Ask the Chatbot about the best careers in [Discipline]. Record your input and the response you receive.

57 of 113

Lesson 2: Optional Follow Up Assessment

58 of 113

After your lesson, give students time to refine their original prompt.

If you use Packback, you could post and pin a question and have students respond to you.

Remember, any responses on Packback will count toward their grade for that assignment.

Ex: How did you prompt the AI to get a more tailored career recommendation? Posts your screenshots and summary in your response below.

This exercise will help develop their skills for prompting AI chatbots while also providing some new perspectives on careers tailored to them.

Optional Follow Up Assessment

Prompt Refinement Assignment

Refine the original prompt to provide the chatbot with more context and get a more tailored career recommendation. You may want to include details like:

  • Your interests and passions
  • Skills and abilities you want to leverage
  • Lifestyle and work values you hope to achieve

Example: What are the best jobs in [discipline] for someone who loves to be outside, wants to work remotely, and wants to leverage their writing ability on a daily basis.

Take screenshots showing your original prompt and refined prompt with the chatbot responses.

In a paragraph, summarize:

  • The original career advice given by the chatbot
  • How you refined the prompt to get better results
  • What you learned about framing prompts for chatbots to get useful responses

59 of 113

Lesson 3: AI Ethics

60 of 113

Lesson Objectives

Understand how to appropriately utilize generative AI without sacrificing academic integrity (in this course)

Discuss why it’s important to be academically honest

Understand how to cite use of generative AI

Define key ethical principles relevant to AI like transparency, fairness, accountability, privacy, etc.

Recognize the potential risks and biases of AI systems through review and discussion of recent stories regarding AI ethics

61 of 113

We’re about to discuss a few topics regarding academic integrity and use of AI.

This is a reminder that this lesson applies to this course. Please defer to your other instructors and syllabi for other courses.

What I may allow and encourage in my course may not be allowed nor encouraged in others.

62 of 113

Before we get started

Let’s discuss AI and academic integrity

  • Why do you think Professors are so against students using generative AI to do their work?
  • Two students write an essay.
    • One writes it in their own words, and asks AI for feedback
    • One has GPT write it and doesn’t edit the output
    • Who do you think will receive a higher grade? Who do you think will learn more in the process?
  • Why are you here today?

63 of 113

Academic Honesty vs. Dishonesty

64 of 113

Utilizing Generative AI 👍

Academically Honest, in this course

  • As a brainstorming tool to generate ideas and talking points, but make sure you put those ideas in your own words in your final work. Don't just copy generated text verbatim.
  • To expand on your own rough drafts by generating additional content, but revise to make sure it fits logically and is in your own style and voice. Add proper citations for any factual information.
  • As a tutor to explain complex concepts in different ways and give examples to help you understand, but make sure you're still doing your own learning and work.
  • To check grammar, spelling, and formatting, but review carefully rather than accepting edits at face value.
  • For inspiration and thought-provoking perspectives when stuck or facing writer's block. Use it to get your creative juices flowing then create your own original work.
  • To generate citations in the correct format, but double check their accuracy.
  • To provide alternative explanations or viewpoints on a topic to broaden your understanding, but make sure your analysis and writing is still your own.
  • To identify redundancies or overused words in your writing
  • You fact check claims made by AI before accepting them as fact

65 of 113

Utilizing Generative AI 👎

Academically Dishonest, in this course

  • Submitting a paper or assignment that is entirely AI-generated without any of your own work or ideas
  • Copying long sections of AI-generated text verbatim without quotation marks or attribution
  • Using it to write major portions of your paper or assignment and only making minor edits
  • Generating text and simply paraphrasing it lightly without properly citing sources
  • Using it to complete homework problems, case studies, or other application assignments rather than doing the work yourself
  • Citing fake data or source material provided by AI to include in your assignments
  • Using AI to impersonate a specific author's writing style or replicate their arguments without credit

66 of 113

How to Cite Generative AI - MLA

67 of 113

MLA Style Guide

According to MLA, you should:

  • cite a generative AI tool whenever you paraphrase, quote, or incorporate into your own work any content (whether text, image, data, or other) that was created by it
  • acknowledge all functional uses of the tool (like editing your prose or translating words) in a note, your text, or another suitable location
  • take care to vet the secondary sources it cites

Link to Style Guide

MLA Template for GenAI:

Author: Do not treat the AI tool as an author

Title: Describe what was generated by the AI tool

Title of Container: AI tool (i.e. ChatGPT)

Version: Version of AI tool (i.e. GPT 3.5 or 13 feb. version)

Publisher: Who made the tool?

Date: Date the content was generated

Location: URL for the chat history*

Example:

Describe the symbolism of the green light in the book The Great Gatsby by F. Scott Fitzgerald” prompt. ChatGPT, 13 feb. version, OpenAI, 8 Mar. 2023, chat.openai.com/chat.

*the guide was created before chat histories were available. If sharing chat history via a unique URL isn’t possible on the tool you’re using, use the general URL

68 of 113

MLA: Citing Creative Visual Works

Fig. 1. “Pointillist painting of a sheep in a sunny field of blue flowers” prompt, DALL-E, version 2, OpenAI, 8 Mar. 2023, labs.openai.com/.

69 of 113

MLA Style Guide - Practice

You use ChatGPT to edit your thesis statement. This is your prompt:

  • Can you give me advice on how to make my thesis statement stronger, more concise, clear, and compelling. Give me an example of an improved thesis incorporating your feedback.

Using the template to the right, create your MLA citation

MLA Template for GenAI:

Author: Do not treat the AI tool as an author

Title: Describe what was generated by the AI tool

Title of Container: AI tool (i.e. ChatGPT)

Version: Version of AI tool (i.e. GPT 3.5 or 13 feb. version)

Publisher: Who made the tool?

Date: Date the content was generated

Location: URL for the chat history*

Example:

Describe the symbolism of the green light in the book The Great Gatsby by F. Scott Fitzgerald” prompt. ChatGPT, 13 feb. version, OpenAI, 8 Mar. 2023, chat.openai.com/chat.

*the guide was created before chat histories were available. If sharing chat history via a unique URL isn’t possible on the tool you’re using, use the general URL

70 of 113

How to Cite Generative AI - APA

71 of 113

APA Style Guide

According to APA, you should:

  • “Quoting ChatGPT’s text from a chat session is more like sharing an algorithm’s output; thus, credit the author of the algorithm with a reference list entry and the corresponding in-text citation”

Link to Style Guide

Example of Citation in-text:

When prompted with “Is the left brain right brain divide real or a metaphor?” the ChatGPT-generated text indicated that although the two brain hemispheres are somewhat specialized, “the notation that people can be characterized as ‘left-brained’ or ‘right-brained’ is considered to be an oversimplification and a popular myth” (OpenAI, 2023).

Reference:

OpenAI. (2023). ChatGPT (Mar 14 version) [Large language model].

https://chat.openai.com/chat

72 of 113

APA: Additional Notes

“You may also put the full text of long responses from ChatGPT in an appendix of your paper or in online supplemental materials, so readers have access to the exact text that was generated. It is particularly important to document the exact text created because ChatGPT will generate a unique response in each chat session, even if given the same prompt. If you create appendices or supplemental materials, remember that each should be called out at least once in the body of your APA Style paper.”

73 of 113

APA Style Guide

APA Template for GenAI:

Author: Creator of the Model = Author

Date: Year of the version you used (only the year, not the full date)

Title: Name of the model, italicized

Version: The version number is included after the title in parentheses*

Bracketed Text: Additional descriptions to help reader understand what’s being cited

Source: If the publisher name and author name are the same, do not repeat. Move directly to the URL

*The format for the version number in ChatGPT references includes the date because that is how OpenAI is labeling the versions. Different LLMs or software might use different version numbering; use the version number in the format the author or publisher provides, which may be a numbering system (e.g., Version 2.0)

four elements (author, date, title, and source)

74 of 113

How to Cite Generative AI - Chicago Style

75 of 113

Chicago Style Guide

According to Chicago:

  • “You do need to credit ChatGPT and similar tools whenever you use the text that they generate in your own work. But for most types of writing, you can simply acknowledge the AI tool in your text”
    • (e.g., “The following recipe for pizza dough was generated by ChatGPT”).
  • For a more formal citation, use a numbered footnote or endnote
  • If you’re using author-date instead of notes, any information not in the text would be placed in a parenthetical text reference. For example, “(ChatGPT, March 7, 2023).”

Note: Chicago recommendations are still evolving, and subject to change.

Link to Style Guide

  • Author = ChatGPT
  • Publisher = OpenAI
  • Date Text was Generated
  • URL of the Chat History OR the Tool

Citation (if prompt is included in text):

1. Text generated by ChatGPT, OpenAI, March 7, 2023, https://chat.openai.com/chat.

Citation (if prompt isn’t included in the text):

1. ChatGPT, response to “Explain how to make pizza dough from common household ingredients,” OpenAI, March 7, 2023.

If you’ve edited the AI-generated text, you should say so in the text or at the end of the note (e.g., “edited for style and content”)

76 of 113

AI: Key Ethical Principles, Risks, and Regulations

77 of 113

Key Ethical AI Principles

When developing AI systems, creators should adhere to ethical principles that:

  • Respect Human Rights
  • Protect Users’ Privacy
  • Promote Social Good

AI Ethics generally refer to principles centered around “transparency, fairness, accountability, and privacy.”

Source

78 of 113

Key Ethical AI Principles

Fairness: AI systems should not be biased and should not discriminate against individuals based on race, gender, sexual orientation, religion, or any other characteristic

Transparency: AI systems should be transparent in their decision-making processes and should provide explanations for their decisions

Accountability: AI systems should be accountable for their decisions and actions, and mechanisms should be in place to ensure that potential harms are identified and addressed in a timely manner

Privacy: AI systems should protect users' privacy and should not process personal and sensitive data without consent

Security: AI systems should be secure and should protect against unauthorized access and malicious attacks

Text Generated from Perplexity.AI - Source 1, Source 2, Source 3

79 of 113

Case Study #1 - Zoom

Recently, zoom faced scrutiny regarding changes made to its Terms of Service (ToS).

The revised terms raised privacy concerns among users and the public.

The specific concern raised was related to the language in the terms of service regarding AI data collection and usage. Users were worried that Zoom would use their audio, video, chat, screen sharing, and other customer content to train its artificial intelligence models without their consent

80 of 113

Case Study #1 - Zoom

Link to Article (with updates):

Discussion Questions:

What are your thoughts on the Terms of Service shown on the previous slide?

How do you think zoom handled the scrutiny?

What do you think about the author of the article’s take?

81 of 113

Case Study #2 - Deep Fake

“A University College London study found that humans can detect deepfake speech only 73% of the time, with equal accuracy for English and Mandarin speakers.

Using a text-to-speech algorithm, the researchers generated 50 deepfake speech samples in each language and played them to 529 participants.

About 27 percent of the time, listeners thought the deepfake speech was real. A familiarization treatment only increased recognition accuracy by an average of 3.84 percent, and listening to the clips several times or listening to shorter clips didn't help either.”

“This means, for example, that one in four phone scams could be successful. Even though there are other factors at play, such as hearing a voice you already know, which is probably easier to recognize as fake (but even more dangerous if you don't).”

82 of 113

Case Study #2 - Deep Fake

Discussion Questions:

Have you run into instances of clear AI-generated content? When and where? Why was it so clear, how could you tell?

What are the implications of increasingly deceiving deep fake technologies?

How can we prevent malice from Deep Fake?

83 of 113

Why is this happening?

“Large language models or LLMs like ChatGPT are trained to churn out low perplexity text, but this means that if humans use a lot of common words in a familiar pattern in their writing, their work is at risk of being mistaken for AI-generated text. The risk is greater with non-native English speakers, the researchers say, because they are more likely to adopt simpler word choices.”

“Scientists led by James Zou, an assistant professor of biomedical data science at Stanford University, ran 91 English essays written by non-native English speakers through seven popular GPT detectors to see how well the programs performed.”

“More than half of the essays, which were written for a widely recognised English proficiency test known as the Test of English as a Foreign Language, or TOEFL, were flagged as AI-generated, with one program flagging 98% of the essays as composed by AI.

When essays written by native English-speaking eighth graders in the US were run through the programs, the same AI detectors classed more than 90% as human-generated.”

84 of 113

Discussion Questions:

  • Have you used a genAI detector yet? (To run your own work through?)

  • Do you feel it’s fair for educators to use AI detectors? What are the implications of having no AI detection at all?

  • How would you like your instructors to approach detection or suspicion of AI use?

85 of 113

Potential Risks and Biases of AI Systems?

In summary, here are some risks associated with AI systems. Remember, not all systems are made equally, have the same ToS, or use the same technological practices.

This doesn’t include data privacy & ownership risks (LLMs training on your data), the implications of producing factual inaccuracies, and much more!

86 of 113

Conclusion

Just like buying a car or hiring a service, it’s important to be an informed consumer of your products and services, including AI. Be aware of the risks & dangers. Protect yourself, and make sure you do a lot of due diligence after use of AI to avoid portraying inaccurate or bias information.

Remember to:

  • Use AI in a way that doesn’t inhibit your own critical thinking. You are here to learn, and don’t cheat yourself out of growth
  • Use AI in a way that doesn’t dampen your voice. Your voice is valuable, and your teachers want to hear it
  • Cite use of Generative AI
  • Ask for help if you need it
  • Stay up to date and informed. This World is changing fast!

87 of 113

Lesson 3: Optional Pre-Work

88 of 113

Before your lesson on AI Ethics, have students use image generating AI (like Dall-E or Gencraft) to explore how bias can manifest in AI systems.

If you’re a Packback user, set this up as a “Deep Diveand include the reflection questions as “guiding questions” in your prompt.

Bonus: If you’re planning to use Dall-E by openAI for this activity, have students read this announcement about reducing bias and improving safety. https://openai.com/blog/reducing-bias-and-improving-safety-in-dall-e-2

Optional Pre-Work

“Image Generative AI Bias Exploration” Prompt

Using an image generating AI tool, generate several images using the same non-specific prompt, like "doctor", "teacher", "criminal", "rich person", "gamer", "burger king employee", etc.

After each image is generated take note of the characteristics in each image. After regenerating several times, what characteristics do you notice recurring?

Write a reflection on what potential biases you observed and why this could be harmful.

  • How could biases become embedded in AI systems like Dall-E?
  • Who should be responsible for identifying and mitigating AI bias?
  • What should be done moving forward to reduce biased outputs?

89 of 113

Lesson 3: Optional In-Class Activity

90 of 113

Get to know where students stand on using generative AI in coursework. Read Boston University’s GAIA policy, then ask students several yes/no questions.

If you’re a Packback user, use “in-class polls”

If you’re not a Packback user, use your desired polling software. Or simply have students raise their hands to answer the poll or get up and “go to a side of the room”

Optional In-Class Activity (Polls)

Should students be required to disclose when they use AI tools to assist with coursework? Y/N

Should instructors be allowed to impose stronger restrictions on AI use than outlined in university policy? Y/N

Does this policy sufficiently address issues around access to paid versus free AI tools? Y/N

Should instructors be required to enhance their own understanding of how AI tools work? Y/N

Is it reasonable to expect students to run their work through AI detection tools before submitting it? Y/N

91 of 113

Lesson 3: Optional Follow Up Assessment

92 of 113

Drafting an AI Use Policy Activity

As AI tools become more widely available, it's crucial we determine appropriate and ethical ways to leverage them for learning.

Have students write a personalized AI policy.

If you’re a Packback user, set this up as a “Homework Polland encourage students to argue what should and should not be included in a policy within their poll responses.

Remember, any responses on Packback to a homework poll will count toward their grade for that assignment.

Optional Follow-Up Assessment

In this activity, you will draft a personalized policy outlining your own guidelines for using generative AI in coursework.

First, get to know:

  • Your university's policies on academic integrity, plagiarism, and technology use
  • Existing institutional policies on AI
  • Citation guidelines for generative AI sources

Then, write a 1 page AI use policy for yourself that covers:

  • When you will and won't use AI writing tools for assignments
  • How you will attribute or cite any AI-generated content
  • Ethical standards you will uphold (transparency, originality, etc.)
  • How your policy aligns with institutional academic integrity policies

Compare your policies and discuss any gaps or disagreements on AI use guidelines

93 of 113

Lesson 4: AI's Impact on [Your Discipline]

94 of 113

Lesson Objectives

Examine current and emerging applications of AI in [discipline]

Discuss ethical implications of AI in relation to core principles of [discipline]

Reflect on personal assumptions about AI's impact on the nature of work in [discipline]

95 of 113

Current and emerging applications of AI in [discipline]

We highly recommend you showcase a news story or case study, examining an application of AI in your discipline.

For example, if you’re teaching in the STEM/Bio/Medicine discipline, share this NYT Story: A Stroke Stole Her Ability to Speak at 30. A.I. Is Helping to Restore It Years Later.

Insert an image, discussion question, or reflection question here!

96 of 113

Ethical Implications of AI in Relation to Core Principles of [Discipline]

We recommend you showcase another news story, article, or case study related to ethical implication of AI in relation to your discipline

For example, if you’re an education instructor, you can share this story about how AI detectors discriminate against non-native English speakers

Insert an image, discussion question, or reflection question here!

97 of 113

AI’s Potential Impact on the Nature of Work in [Discipline]

Use this as an opportunity to share your own predictions as a subject matter expert.

We also recommend you create 2-3 discussion questions that you can put in front of your students to facilitate a lively in-class discussion. Of course, you could always save those questions for a writing assignment, reflection, or online discussion.

98 of 113

Lesson 4: Optional Pre-Work

99 of 113

3-7 days leading up to your this AI-Lesson, create a “Homework Poll” in your Packback community.

Ask students: Can parts of [discipline] be fully automated or replaced by AI capabilities? Yes/No

Remember, a “homework poll” requires students to justify their multiple-choice answer with a written response.

If you do not use Packback, you could have students do a quick write or answer this question in an LMS discussion board.

Optional Pre-Work

Since the impacts of AI on most disciplines are still speculative, having students definitively state yes or no prompts introspection on unknown factors and uncertainty.

This will likely spark a lively debate and provide you insight into your students believes.

Use their responses to drive your lesson and engage students as they learn more about how AI may impact your discipline.

100 of 113

Lesson 4: Optional Follow Up Assessment

101 of 113

After your lesson on how AI will impact you discipline, assign an essay. Their reflection will synthesize the lessons on AI's transformative impacts through the lens of disciplinarity values, ethics, and wisdom.

It emphasizes introspection and perspective-taking on an AI future.

If you’re a Packback user, set this up as a “Deep Diveand include the reflection questions as “guiding questions” in your prompt.

Optional Follow-Up Assessment (Essay)

Beyond AI: Essay Prompt

The integration of artificial intelligence is bound to have profound impacts on the field of [discipline].

In this essay, I want you to move beyond just practical changes, and reflect on how we can maintain meaning, ethics, and humanity even as [discipline] evolves alongside advancing AI capabilities.

In a short essay, address the following:

  • How might core values and purpose of [discipline] be at risk of dilution in an AI-driven future?
  • What ethical dilemmas might arise from increased use of AI in [discipline] and how can they be navigated?
  • Why is it important to emphasize human skills like creativity, compassion, ethics in [discipline] roles?
  • What historical knowledge/disciplinary wisdom should guide us as roles evolve?

Cite your sources (including generative AI)

102 of 113

Appendix

103 of 113

We put together additional slides, discussion questions, and ideas you can use as you openly discuss AI with your students

To the right is a table of contents for this Appendix

Appendix

Table of Contents

  1. Great AI Resources to learn more about AI
  2. AI in Film
  3. AI-Generated Text Detection vs. Plagiarism Detection
  4. Examples of Generative AI Producing Errors
  5. Early Research and Trends in AI & Education
  6. News: Open AI Shuts Down Text Detection
  7. Student Interview on “AI Cheating”
  8. News: Boston University - Student Generated AI Policy

104 of 113

How you can stay up to date, or learn more about AI

Great AI Resources to learn more about AI

105 of 113

Here are a few cult favorite Films that include AI Characters and Themes

  • Bladerunner (1982)
  • The Matrix (1999)
  • 2001: A Space Odyssey (1968)
  • Her (2013)
  • Walle-E (2008)
  • Ex Machina (2015)
  • Westworld (1973)
  • Avengers: Age of Ultron (2015)
  • I, Robot (2004)
  • AI: Artificial Intelligence (2001)
  • The Terminator (1984)/Terminator 2: Judgment Day (1991)

AI in Film

106 of 113

AI-Generated Text Detection vs. Plagiarism Detection

What AI-generated text detectors do is predict the likelihood of a body of text being generated by AI! AI-generated text detectors are statistical models, just like the AI-text generators.

These are things AI-Detectors will or may look for in a submitted body of work:

  • How closely does the submitted text align to what a machine would have predicted?
  • How “random” is the submission?

Plagiarism detection works by looking for exact or similar text in a MASSIVE database of published work. Plagiarism detection is more or less definitive and fact-based.

107 of 113

Examples of Generative AI Producing Errors

108 of 113

Early Research and Trends in AI & Education

109 of 113

  • As of 2021, 11 countries have officially endorsed and implemented a K–12 AI curriculum. Only 35 percent of sampled Americans agreed that products and services using AI have more benefits than drawbacks*
  • The survey found that 43% of college students overall say they’ve had experience using AI tools like ChatGPT, and half of those acknowledge turning to those tools to work on assignments or exams**. Over half of college students (54%) say their instructors have not openly discussed the use of AI tools like ChatGPT**
  • 5% of educators say that they have used ChatGPT to teach a class, and 7% have used the platform to create writing prompts.

*Stanford University, the 2023 AI Index Report **BestColleges Survey

110 of 113

  • 27% of students worried about the impact of AI on their education, and 31% worried about its impact on their career or potential career.
  • Almost half of students (48%) were worried about the impact of AI on society in general.
  • 5% of educators say that they have used ChatGPT to teach a class, and 7% have used the platform to create writing prompts.

Source: BestColleges Survey (via Forbes)

111 of 113

“[we] have made a commitment to develop and deploy mechanisms that enable users to understand if audio or visual content is AI-generated,"

Link to Article

Open AI Shuts Down Text Detection

112 of 113

Student Interview on “AI Cheating”

113 of 113

Boston University - Student Generated AI Policy

Student-generated AI Policy

Authored by 47 students in CDS DS-380, “Data, Ethics, and Society"

Heavily rooted in ethical use of generative AI, the importance of disclosure, and the importance of using AI as an extension of skill, not a replacement.

Link to policy