AI 101 for Higher Ed
An initial primer for a higher education institution that ISN'T about education...
On Tuesday, I had the opportunity to talk to the staff at North Shore Community College, which some folks know is a place near and dear to my heart. This was a slightly different talk than my previous two types of talks. Typically, I’ve been talking with faculty or folks in career counseling. Those continue to be areas where I’ll have conversations. This conversation though was more focused on the institution as a whole and helping staff to consider the role(s) generative AI will play in their different work—so not just teaching and learning but all the other things the institution might be doing. As I continue to do, I made an annotated slidedeck as part of the process and you’re welcome to it (and the slidedeck itself is included).
AI 101: Possibilities & Considerations of Generative Artificial Intelligence
Hi all,
It never ceases to bring me joy to be here in community with all of you. As an alumn, part-time faculty for 17 years, and someone who started his instructional designer career here with the folks in the Center for Teaching, Learning, and Innovation (CTLI), it’s always an honor and probably always a yes to be asked to be here.
So, I’ve got a big task–take the next 30- minutes to explain what generative AI is and its relevance to your work in higher ed–and do it in a way that sticks.
I know every educator in the room is chuckling about such a task.
But I’ll say, challenge accepted and let’s get to it!
First, I’ll mention the link at the bottom there (the one I’m throwing into the chat now) is to my annotated slide deck. This resource has everything that I’m about to say, the slides, and embedded resources for each slide.
I can be here for this hour but to really start learning about generative AI, there’s a lot to sift through and that resource is a great start. All is covered with a Creative Commons license, because that’s how I roll.
What are we talking about? As you can see, I have two agendas because whenever we talk there’s always what’s said and what’s not said. I’ll do my best to address both today but if not, I hope to lead you places where you can continue to understand both.
We’ll start with a meta level of what we’re talking about today and clarify what we won’t be talking about (and why).
Then, we’re gonna get a little bit interactive with the chat. We’re going to talk about what generative artificial intelligence actually is….and what it isn’t! Because there’s a lot of folks selling a lot of garbage out there and sifting through it isn’t easy.
Next, we’ll get into how generative AI may help us in our work in some interesting ways in the near future. Of course, we can’t talk about the ways it will help without talking about how it will complicate things as well.
I’ll frame ways that it may be used in the near future…and balance that out with what’s realistic in this current moment.
Finally, we’ll get into what you probably should and definitely not do at this time.
Sound good? Then, if we have our directions, let’s actually figure out what we’re talking about!
What are we talking about?
We could talk about generative AI endlessly but today, we’re going to be talking using generative AI as part of the staff at North Shore Community College. We’ll highlight different use cases in that particular context.
What aren’t we talking about? These are important topics. But this is an introductory discussion about these tools and those topics deserve their own deeper dives (well, except for the last one).
We won’t be talking about generative AI in the teaching and learning. That’s an essential topic and also the Center for Teaching, Learning, & Innovation has been leading conversations around this topic and that is where the conversation should continue–with those amazing and talented folks! Honestly, so much of my thinking and navigating about this has been influenced, shaped and improved from so many conversations & learning opportunities with Andrea Milligan among other all stars on that team. So, go there for help with your teaching on this, or really, everything!
We’re also going to avoid getting into the big questions about what these tools mean for the world at large.
These things we may touch upon tangentially but they’re not central to our talk today.
What is/isn’t Generative AI?
Here are some of the currently most popular AI tools. Which ones have you heard of?
Nearly everyone has heard ChatGPT, but what about the rest? Put into the chat which of these you HAVEN’T heard.
Write in all the ones you don’t know.
Now, that’s quite the variety of experiences here in the room.
If you are familiar with any on the list, feel free to provide others with a 1-2 sentence description of what it is.
And no cheating by going to ChatGPT to generate an answer! You can learn more about these on the annotated slide deck.
These tools are becoming increasingly ubiquitous. And it’s easy to dismiss this as another tech-hype cycle. I think in many ways it is. But I also think the AI Hype feels different from prior cycles.
Generative AI is different from the AI we’ve been using for the last decade.
We all have stories of how Siri or Alexa has misheard us (in fact, did I just set off anyone’s Alexa just by saying it?). These didn’t feel as impressive or easy to use right out of the box. But generative AI–AI that can generate text, audio, image, or video through the use of everyday language–that feels different because of its ease of use.
We’re several years since the launch of the Metaverse and can ANYONE tell me what it really is and if they have visited it?
The lift to figure out what the Metaverse is, how to access it, how to create in or with it, and why it would be better than other things–that’s a lot of figure out.
But the generative AI comes in the form of a chatbox–a chatbox many of you have already been using in this very zoom meeting…something that’s been around for decades. We all know: see a textbox on the computer, enter text.
Generative AI took something terribly complex and nuanced and made it usable in a textbox. That’s a devilishly easy invitation.
What is generative AI? Feel free to put your answers in the chat. I’m almost tempted to play a game of “wrong answers only” but I won't.
These are good! I used AI to tell me what it is. I asked ChatGPT, Claude, Bing, and Bard to define generative artificial intelligence. They all gave decent answers of varying and similar degrees.
But then I poured all of them into Claude and told it to: “Take these 4 definitions of generative AI and come up with the cleanest, clearest, and simplest explanation of generative AI that is under 50 words, can be understood by an 8th grader and maintains the full integrity of what generative AI is:”
It gave me that answer on the right: “Generative AI is a type of AI that can create new content like text, images, and videos after learning from large amounts of data.”
I wasn’t that excited with that answer so I asked it to elaborate. It came up with
“Generative AI refers to artificial intelligence systems that can generate new, original content such as text, images, audio, and video. These AIs are trained on large datasets which allows them to learn patterns and relationships in data. They can then use what they've learned to produce new examples that are similar to but not exactly the same as the training data.”
That’s a clearer answer. Others and I would nitpick and challenge some part of this definition, but that’s the case with many loaded terms.
I also want to demonstrate here a way that I often use it is to pull together different ideas or bits of information to see what it can create.
And it’s not that I can’t do this myself–but that I can have generative AI do in 30 seconds what might take me anywhere from 30 minutes to several hours.
Ok–let’s ask some more questions–remember to put your answers in the chat and I’ll call out what I can see.
What do we think–true or false:
Generative AI may be becoming sentient. That is, alive.
False. It is not a thinking being–no matter what all those sci-fi novels, movies, comics, TV shows, and Silicon Valley Tech-Bros tell us. We have a long history of telling ourselves that our objects are real–going back millennia.
It is a complex technology with lots of possibilities, but sentience is not one of them in the immediate future. What we hear and see in the media is often sound bites and misrepresentations of what generative AI is.
This is important to hold in our minds. This tendency to anthropomorphize or mythologize our technology as having life is common throughout history going back thousands of years. In more modern times, this has become known as the ELIZA effect after experiments in the 1960s with a computer program that people became attached to when it used language in ways that felt real to the users.
We want to remember this effect as we use it.
Ok, what do you think about this one?
“Generative artificial intelligence is similar to human thinking when it arrives at its answer.”
Does generative AI think like we do?
False. Generative AI doesn’t think like you or I do. First of all, we all think differently–that’s why we talk about neurodiversity. But our mind is composed of whole things be it words, images or sounds, tastes, etc. By contrast, generative AI works through mathing the hell out of things.
When working with text, a generative AI will rely on its “large language model”--that is a massive data set of text and the statistical models calculated related to that data set.
That large language model was created by an AI tool mathematically analyzing the relationships across that massive text. It doesn’t analyze the words themselves but rather strings of characters. Those strings of characters are referred to as “tokens” and are short, something around 4-5 character long.
The AI calculates the probability of relationships of those tokens across its data set and also with any prompt you give it. So when the AI gets a prompt by a user, it mathematically examines the relationship among tokens and then searches its large language model for similar relationships of tokens to help generate the most probable answer. It strings these tokens into words and sentences that mathematically make sense.
It’s not thinking, it’s mathing. It has no awareness or internal understanding of the relationships of the words its using any more than a calculator knows that its also spelling out a word with the number 376616.
Now you’re all wondering what word does that number spell? It spells giggle. Which, I’m pretty sure some of y’all just did!
Ok, what about this one? “Generative AI can lie.”
This one’s tricky right? We have a mixture of answers on this one as we should.
It is true that Generative AI doesn’t lie.
Lies come with intention and intention is human. See what I mean about it being tricky to remember this is a calculating machine and not a sentient being?
Can it give false information. You betcha! But when you hear language about it lying or hallucinating–it’s important to understand someone is assigning human characteristics to AI.
It will present false information because it’s not thinking and rationalizing or proving things in the ways that we humans think about it. It’s mathing relationships and that may lead it to come up with poorly calculated wordings.
So while it doesn’t lie, you still shouldn’t trust it.
Finally, how about this: Generative AI will take all our jobs. I’m sure we’re catching onto the theme here. So it is the last one, I promise!
As Public Enemy said, “Don’t believe the hype.”
It will disrupt and augment work and part of the challenge for all of us will be to stay abreast of it to understand its impact on our areas of work. Any knowledge worker is going to have to consider and contend with how these tools can amplify their work and not replace it.
It will help us do a lot of things we didn’t do before and that can create new opportunities.
And yes, there are jobs that will be lost and there will be new jobs that emerge. The history of the world is filled with technologies that would supposedly ruin all employment and yes, there’s still plenty of work to be had.
We’re an educational institution and we offer degrees and programming in many kinds of work that did not exist when the college it was formed nearly 50 years ago and we have terminated programs that no longer have relevance. Technology continues to drive a lot of those professional changes.
How will generative AI help?
The first two on this list are mostly aspirational or anticipated ways it will help. They may feel more realistic when and if we achieve general artificial intelligence. AI that can perform the full range of tasks embedded in a complex command and not just generate text or images in response.
From 4 downward, many folks are starting to use generative AI tools in these particular ways. Maybe still not wholistically but definitely in capacities that hint at what’s to come.
Some of my work is around leveraging these tools for professional development and career counseling.
Generative AI can tap into what I would call the hidden curriculum of the world to help folks quickly and effectively understand the nuances of different practices around the employment process that aren’t obvious or clear for people trying to move into better paying knowledge-industries that come with a variety of coded signals.
Institutions and organizations are starting to leverage it in grant processes to help them pull together things more quickly, while others are playing with how a self-evaluative chatbot might help one more effectively capture their previous work and develop goals for the next year.
There’s definitely more and you can find out more about these in the annotated slide deck.
But generative AI is definitely going make things more complicated in some capacities.
More than previous significant technologies, we know right from the get-go that generative AI has a lot of baggage. Like–a lot.
Generative AI has significant environmental impacts from massive energy consumption to contributing to significant and harmful extractive resource practices.
The datasets are hyper-focused on North American, English, white, and male voices and also, the fullness of the dataset and how it operates is largely hidden from users.
And finally, there’s the labor practices of subjecting vulnerable populations to the worst content–a practice made pretty by the euphemism “content moderator” which often comes with healthy doses of post-traumatic experiences by those underpaid workers.
Then, there are the things we can anticipate but don’t know the full extent of their impact.
Something that I’m thinking about particularly is the homogenizing of the English language as we use these tools more and more.
Because of the mathematical models, it will reproduce the style and approach most present in the dataset–and in doing so, pushes all of us to expect specific ways and structures of talking or writing in professional contexts.
Now, we already have norms but the scale of this feels like it will move from norms to constraints and limit the linguistic diversity that language is supposed to have. Maybe that won’t be a problem but it’s something we know can happen as we move from one technology to another. After all, national broadcast television for decades dictated a supposed “normal” way to talk that many folks were forced into reproducing.
And then there are the things that we just don’t know and aren’t prepared for…and yes, I realize I’m borrowing from Donald Rumsfeld…and apparently, that’s where generative AI has led me…clearly, we’ve got a lot of things to think about…
How might we use generative AI in the next decade?
We’re in the early stages of generative AI and there are many ifs to consider, but we can see the trajectories of how the tool may be used in many areas of higher education.
In particular, if we get to what is called, “general artificial intelligence”--we could see real systematic changes.
For instance, it might be something along the lines of telling a generative AI to run a report on student completion in the Fall semester courses. The general AI may run the report and disseminate it instantly to different departments with their relevant information highlighted.
Consider the flow of how that information happens now. And now, what would happen if the registrar gets a report identifying students that are struggling and already creating communications to the student, the advisor, and whoever else may need to be in the loop. Meanwhile, academic departments get a report of the completion rate of their courses for first semester students vs second semester students along with initial analysis and recommendations of next steps. Over in student support, student advisors get reports about trends among their particular students and what is and isn’t working.
I could go on but that kind of ripple flow of information and action could be so powerful and helpful in a way that currently is hard to experience because so much of this happens manually throughout the institution.
Again, future looking but more likely and possible as we move into the general AI space.
What can we do right now?
The next are just a couple prompts and answers that show you some of the interesting ways AI can be used. The annotated slide deck will have a lot of additional prompts and guidance on how to use it.
I use now at the beginning of each semester and anyone needing to get a list of dates will adapt it.
We all do this and here is an example of minimize how long a task will take with generative AI. I asked ChatGPT to give me all the Tuesdays between the start and end of the semester. This saves me toggling back and forth between my syllabus and a calendar to get these days. I also asked it to include holidays and such so I can keep that in mind as well.
Here are the results that I got in less than 30 seconds. Now I can copy & paste that list into my syllabus. It also included Islam and Hindu holidays but you can find those in the annotated slide deck.
Generative AI can help with tedious tasks like this.
We can often figure things out on our own and we also know that our thinking can be improved through dialogue or having someone provide examples. In this instance, I asked ChatGPT to provide some examples of ways folks might should be able to use generative AI.
I also framed the output in the form of a table with some added factors. I’m not just asking it to give answers but clarifying the response.
There’s more on the annotated slide deck but it provides a lot of ideas right out of the gate. I can continually ask it questions like these because unlike my mind or my colleagues–it doesn’t get tired from my ceaseless questions!
This tool can also be really helpful with 1st drafts. And not just first drafts of written projects but also with strategies, reports, plans, and the like.
In this case, I asked it to pretend it was an expert in communications and student support. I followed this with asking it to create a communications calendar to students about the different timely information and supports throughout the semester.
And now, I have the start of a plan. I could take it further an ask it to go deeper on one of the items and start to flesh out that part of the plan and in less than 30 minutes have a fully detailed first draft that the team can adjust and update as needed.
I find it help for looking at qualitative data and can use it to make sense of things quickly. I’ll still dive into the data for more insights but the high level analysis can help me move with more clarity.
Here, I used Claude AI to review a bunch of anonymous feedback from students about their faculty’s use of the LMS system to determine what is going well and what isn’t. This is qualitative data and often, we’re dealing with hundreds of responses to this question every 4 weeks during the semester.
From here, I can engage in further dialogue for recommendations, plans, what to do next and the like.
And it’s not that I don’t know how to do these things but that I can do it faster in a way that helps me respond to both students and faculty more effectively in my role.
Ok–so these are some examples of the ways we will be using them for now. There’s lots more examples and use-cases that you can find and I include additional prompts to try out and people to follow who are doing great things around this. And, of course, you have your colleagues–some of whom are doing really amazing things with generative AI in their work already!
What not to do next?
Now, let’s start with what not to do. Some may be self evident, but they need saying.
Don’t assume it is going away and not to bother. It’s ok to be uncomfortable and have all the feels about it–there’s lots of feels to have–I have a different feel about every 10 minutes around generative AI. But ignoring it isn’t an option.
You do not have to figure it all out on your own. You do need to figure out your own relationship with generative AI as it becomes more ubiquitous but learning about it, understanding the myriad elements about it–that’s not on you. We’re all doing this and many of us are sharing what we’re finding.
Do not put anything data of other people into it–especially if it has to do with your work here at NSCC. Wait for unequivocal guidance from the IT department.
Always review and reconsider its outputs. It’s far from being entirely reliable.
Last but not least and circles back to the first. While you don’t want to ignore it, I wouldn’t recommend going all in on it. We’re still in a formation stage with this technology and how we might manifest.
What to do next?
Ok–then what should you do?
Play with it and its variations. Don’t wait–go and kick the tires. You’re not going to break it and there’s lots of support materials to get you started (again, see the annotated slide deck).
Always think about what you are putting into it and if you have the right to do so or are violating someone’s privacy.
Does your usage align with your mission? For instance, any institution that upholds supporting their community and antiracist practices in their mission opens up interesting quandaries. Does marketing save time and money by using generative AI image-creations of a “diverse student body” for brochures, social media, the website and such. OR do they spend the extra time hiring community photographers or getting permissions of individuals to build out their stock photos for visual materials?
Find your outlets to learn about it. You’re part of an institution of higher education–you’re part of an entity that whether we like it or not, are required to be lifelong learners if we want to continue to adapt to an every-changing and complex world. Put this on your personal agenda for learning and development.
Hone in on how the tools are being used in your field. Start to learn what that conversation is like and how folks are considering it. Along those lines, –find your people! People in your fields and areas of work are already doing things with this. And they are writing, making videos, doing podcasts–all the things! Find them–and if you need help, let me lift up our amazing librarians–the OG’s of knowledge search and creation–the folks who could teach Google a thing or two about effective searching!
Keep an eye out for the guardrails that your industry or field are communicating or establishing. Standards for generative AI use are going to be different for market than it will be for recruitment than it will be for alumni outreach.
As you start to learn about it and use it, have conversations within your departments and teams to figure out the norms and acceptable use. Make sure you’re all on the same page and learn from your industry about what should be on those pages!
Create your communities of practice–at NSCC, among your professional colleagues, among your friends–but start learning and sharing–you do not have to be the holder of all the knowledge.
But as you learn more about these tools and find new ways of using generative AI–share them with colleagues. If generative AI is a collective of what already exists, then amplify its benefits by sharing with others.
Within that, determine among your community of practice how you are going to share. Some folks are collectively creating prompt books, others are creating Youtube playlists of things they are doing or of what they find others doing. But find a way to capture and share that works for your group.
I hope this talk has provided folks with a stronger grounding and consideration for how these tools might come into play in your work. And I hope these recommendations are useful ways to get started.
I know any new thing can be overwhelming. I know we can feel challenged by things that come at us this fast as generative AI has. And I also know, you all work at an open-access institution whose fundamental core belief is that everyone can and should have the opportunities to learn and grow. In order for them to do that, we too much uphold, believe, and enact this practice of learning new things as well.
Thank you!
AI+Edu=Simplified by Lance Eaton is licensed under Attribution-ShareAlike 4.0 International
