1 of 23

The Future’s Already Here:

AI Generative Tools and Teaching

Lance Eaton

Director of Digital Pedagogy

College Unbound

lance.eaton@collegeunbound.edu

@leaton01

http://www.ByAnyOtherNerd.com

This work is licensed by Lance Eaton under a Creative Commons Attribution-ShareAlike 4.0 International License.

2 of 23

Acknowledgements

Land Acknowledgment: I start by acknowledging that the land in which I am presenting from in Cranston, Rhode Island, is the ancestral and unceded territory of the Pokanoket, Wampanoag, and Narragansett People and pay my respect to elders of the past, present, and future. One thread of this workshop is dedicated to them and others whose lives and cultures have been marginalized or threatened through the privileging of certain kinds of knowledge, productivity, and property. �

Equity Acknowledgment: This presentation was prepared using ChatGPT, an “AI Chatbot”. I acknowledge that ChatGPT and many AI Generative Tools do not respect the individual rights of authors and artists, and ignore concerns over copyright and intellectual property in the training of the system; additionally, I acknowledge that the system was trained in part through the exploitation of precarious workers in the global south. In this work I specifically used ChatGPT to test out some ideas about its usage, better understand the tool, and may also demonstrate some of the ways it generates answers. (Inspired by Lawrie Phipps and Donna Lanclos's An Offering)

3 of 23

Outline

AI in Context
The Conversations
College Unbound’s Case Study
A Smidge of Prophecy
Questions

Resources

Slides: https://bit.ly/AI-HigherEd-Webinar

Resources: https://bit.ly/AI-HigherEd-Resources

So today, we’re going to cover some ground; some of it will be familiar, some may be new. I strongly encourage folks to make active use of the chat for conversation, resource sharing, connecting, but most importantly, violently disagreeing with me–particularly with emojis!

First, I’m going to place AI in the context of higher ed and educational technology.

Then, we’re going to spend some time talking about the terrain–the different conversations that are evolving within higher education in the last 3 months since ChatGPT first showed upon on the scene.

I’m move into explaining the situation at my institution (College Unbound) and what we have done and why.

And then, we’ll do questions & answers.

Those 2 links on the right of the slide are to this slide deck and a hand out of resources to consider–some created by me and others that I’ve grabbed and collected and think are solid ones to consider.

4 of 23

Things I Can Promise About This Webinar Will Deliver On

Lots of links and additional materials
A minimum amount of staring at the screen while I enter prompts into ChatGPT
A clearer sense of where you fit within the conversations about AI and education
More questions than when you started
Community with other folks that are as concerned, worried, angst, confused, excited, alarmed, curious, [insert your emotion here] as you are!
Few answers than you had hoped to find
A least 1 photo of each of my pets

Let’s begin!

5 of 23

Things I Can Promise About This Webinar Will Deliver On

Lots of links and additional materials
A minimum amount of staring at the screen while I enter prompts into ChatGPT
A clearer sense of where you fit within the conversations about AI and education
More questions and fewer answers than you had hoped to find
A least 1 photo of each of my pets

Let’s begin!

6 of 23

7 of 23

8 of 23

Are you $@&@%#! kidding me?

9 of 23

AI Generative Tools, ChatGPT, & “Creative”

ChatGPT vs AI-Generative
Generative vs. Creative
AI=Mathing the hell outta things

For some folks, this conversation is entirely new and for others, it’s been something they have been aware of directly or indirectly for a while. We kept hearing about AI–particularly in pop culture and many of us grew up with the ever-persistent story that AI was going to kill us all–whether it was from The Terminator, Cylons, Daleks, Hal9000, Ultron, or that most insidious of them all, Clippy.

And this is where folks in the know have really been trying to center the conversation in terms of the language that we use. ChatGPT is the catalyst for a whole slew of tools that have come our or will be coming out in the near future and it’s not just about text. Therefore, “AI-generative tools” is the bigger conversation to be having. And this is where I do my best to focus on language to help us understand the tool.

I named most of the work that I’m doing around this as “AI generative” for 2 reasons. The first is that it is the larger bucket than just ChatGPT. The second is that I want an emphasis on generative as opposed to creative. I think we

10 of 23

The process can be like

ChatGPT or the like
DALL-E or the like
VALL-E or the like
Slides AI or the like
And so much more…

Image Source: Bill Simser

I hope this scenario can better visualize why it’s the bigger conversation than ChatGPT.

This scenario falls somewhere along the continuum from it’s super-amazing and “we live in the future” to it’s super-scary and “we live in the future.”

Right now, it is possible to have ChatGPT write a speech. Plug that text into an AI-slides generator like Slides.AI and then use an AI voice cloner to read the speech. You could have that recorded and submitted for many online assignments for sure or build out your YouTube channel. But these are the range of tools–text-generators like ChatGPT, image generators like DALL-E, voice-generators like VALL-E.

It feels very much like the vast majority of mental work that gets turned into tangible deliverables for evaluation in higher education is very quickly becoming possible to being generated by AI.

11 of 23

The Arrival of ChatGPT is like…

Image source: Mark Mathosian

Image source: John Fader

12 of 23

How We EdTech’d Our Way to Now

Internet

LMSs

Online Video

Web 2.0

Wikipedia

Mobile Devices

MOOCs

Hybrid

Zoom Learning

Hybrid Flexible

I also want to situate ChatGPT and AI Generative tools in a larger context of higher education and technology that often goes unacknowledged within this discussion. That is, there are all these things–that have become part and parcel of of the higher education landscape–many of which we have to think about, prepare for, utilize, and integrate in our classes–and while the claims are that all this technology saves us time, I have trouble believing anyone feels like when it comes to teaching they are spending less time now than 30 years ago. I’ve only been teaching for 17 years and it doesn’t seem like I’m spending any less time on it and definitely, often, more time.

So let’s take a quick rundown of all the revolutionary technology and tech-practices that have become essential to a modern classroom or was, is, or is soon to be disrupting higher education.

The internet–the grand-parent of them all.

With that came the bane and workhorse of every institution: the LMS

Then we started to see and realize, oh, online videos are possible!?!?!?

That was just about the time that we were getting ridiculously ecstatic about Web 2.0! How many folks started blogs? Anyone still doing their blog? My blog recently became a teenager and it is such a pain in the butt–wants its own smartphone, is angry that I won’t let it go on TikTok…the struggle is real. I read recently–with a whole lot of excitement–that blogs were making a come back…of course, they’re all going to be filled with ChatGPT content.

Web 2.0 gave us the modern version of Wikipedia which also was the scariest thing for many in higher ed since chalkboards were slowly being replaced with white boards–the pure anarchy of it all!

Then, came the mobile devices! First it was the smartphone and then—gasp–the tablet! Schools scrambled to figure out if they should ban them or give them to every student or just tell everyone, bring’em if you got’em!

On the coattails of the mobile devices, we saw the rise of the Godzilla of higher education: the MOOC! It was going to change everything. EVERYTHING! Until it didn’t–but lotta people got rich, lotta learning happened, and an even larger amount of people started 1 or MOOCs and never looked back…at the MOOCs.

Hybrid courses also took off during this period and while we had more flavors of hybrid than ketchup, it largely settled on the idea that we would meet face to face less often and do more learning in the LMS or elsewhere.

And that seemed to be the way things were going but then we had a pandemic and everyone had to learn to how to teach via zoom in the middle of a semester while everything was shutting down and toilet paper temporarily became a currency.

The pandemic was still raging and killing people, but administrators insisted that being back in the classroom–at least for instructors–was essential and so we had to start do this thing called hybrid-flexible teaching where you were designing for in-person students and virtual students AND possible, creating a parallel asynchronous course–all for the same pay, of course.

And just when everyone had given up caring about masks or quarantining and we thought we might have a normal semester or two, ChatGPT came along.

And if you’re not exhausted from thinking about all that and how you and your colleagues survived and adapted through all that, then I’m impressed and also would love for you take it from here!

13 of 23

The Conversation about AI Generative Tools in Higher Ed Thus Far

Ban & Punish
Simplify & Avoid
A Classroom Tool
An Instructor’s Tool
Critical Takes
Absent Voices

So here we are! Over the past few months, the conversation has been swirling and happening in department and division meetings, across meals and classrooms, and very much across the internet in blogs, podcasts, Twitter, Mastodon, Facebook, TikTok, Reddit and beyond. What I hope to do in this section is give you the contours of these conversations.

A few caveats to start:

There are likely others and more emerging–but these are the strongest that I’m currently seeing.
Many are not mutually exclusive–as you think about your own approach–consider what makes sense and draw from whichever ones make sense.
I would not say that there are necessarily right or wrong conversations to be a part of or to take up, though I think it’s really important to be grounded in a firm pedagogical understanding of of learning science before taking too strong a stance in any of them.
I have biases and will try not to show them, but I make no promises.

14 of 23

Ban & Punish

Taking a firm and strongly-languaged stance about using ChatGPT or other AI Generative Tools
Simple put: You use it; you’re in trouble.

Benefits

Simply
Clear
More time spent on other things

Drawbacks

Abstinence doesn’t work
Students lose opportunity to understand the tool in course-context
Lot of time spent mouse-trapping

15 of 23

Simplify & Avoid

Connect it to Academic Honesty (e.g. must cite when in use)
Less need to discuss it any further

Benefits

Connects with something students are more familiar with
Mostly clear
More time spent on other things

Drawbacks

May not be as clear as needed
Lose opportunity to understand the tool in course-context
Some time spent mouse-trapping

16 of 23

A Classroom Tool

A learning tool in creation or writing process to generate ideas or examples to work with
Consider how it will be used in their respective industrations/careers and how to use it in those contexts
A tool for critical exploration

Benefits

Its benefits and limitations
Contextual familiarity
Discover new uses

Drawbacks

May mandate usage of tools we don’t fully understand
May violate/push privacy boundaries for students
Give up time on other course elements

17 of 23

An Instructor’s Tool

Create curriculum
Create examples
Develop assessments
Provide targeted feedback

Benefits

Saves time
Creates content that might not have been within reach of the instructor
Insights that can help in teaching
More effective/insightful feedback

Drawbacks

Reinforces the de-skilling of faculty
Lacks transparency/equity if students can’t use it
Questions of privacy, student labor, and copyright

18 of 23

Critical Takes

Examine the context of AI-generative tools, what they do, and how that intersects with the course discipline in disruptive and problematic ways
Identify the underlying structures of inequity that empower AI-generative tools such as algorithmic bias, digital redlining, digital colonialism, de-skilling and gig-economical practices, etc
Identify, determine and exploit flaws in the tool to provide a better understanding of its limitations

Benefits

Clarify the limitations of techno-utopian discourses
Reiterate the importance of liberal arts approaches to study
Raise concerns about populations vulnerable to technological exploitation

Drawbacks

Detracts from the course’s focus
Creates an adverse view that limits learning and use of the tool when necessary
Generates questions but not necessarily answers

19 of 23

Absent Voices

STUDENTS

Benefits

They should not have a say

NONE

Drawbacks

This
Never
Ends
Well

20 of 23

College Unbound’s Approach

Discovery
Consideration
AI & Education
Temporary Policy
Student & Faculty Voices

21 of 23

What does this all mean for education?

So this is where some of my predictions come in for consideration and which, we should always take with several grains of salt

Places like SNHU, WGU, ASU –the production behemoths of courses will likely be the first entities to do large scale implementation of AI Generative Tools to replace most of their army of instructional/course designers and find ways of making instructional faculty to “teach” more students while not paying them any different–they’ll do this by requiring those instructors to work with AI for feedback and the like as well as AI systems monitoring and giving them feedback to faculty and students on their performance.

And a lot of higher ed is going to miss the conversation around AI and education because they are going to be looking at it straight on. They’re going to be thinking about how AI will replace them in the classroom and frame it in the current model of how teaching and learning exists. If AI does take hold, it’s going to look different from how traditional learning in higher education looks because traditional higher ed is a model of convenience and doesn’t necessarily center the students and how they learn; rather they center on a critical mass of students, a centralized location, a frequency of visiting that location for arbitrary sets of time set by the Carnegie Foundation over a 100 years ago. And that’s not going to be how learning with AI is going to look like. It’s going to be dynamic in when and how long it happens, it’s going to be a mixture of generalized and contextualized to the individual, and probably–unfortunately–less social.

But importantly for the rest of us, we will have to anticipate changing (or continuing to change) our teaching and learning practices; we can no longer be production lines and focused on the banking model of education. AI Generative tools will beat us every time in terms of speed, timeliness, context, scale, and efficiency. So we’ll need to reposition teaching and learning increasingly to relationship-building through which trust and risk can feel more possible by students–it’s a way of offering something that will be different from what AI-learning can do.

I have more thoughts on that but that’s forthcoming blog post.

22 of 23

Questions?

23 of 23

Thank you!

Lance Eaton

Director of Digital Pedagogy

College Unbound

lance.eaton@collegeunbound.edu

@leaton01

http://www.ByAnyOtherNerd.com

This work is licensed by Lance Eaton under a Creative Commons Attribution-ShareAlike 4.0 International License.

Slides: https://bit.ly/AI-HigherEd-Webinar

Resources: https://bit.ly/AI-HigherEd-Resources