WEBVTT
00:00:34.839 --> 00:00:43.000
Welcome to Intentional Teaching, a podcast aimed at educators to help them develop foundational teaching skills and explore new ideas in teaching.
00:00:43.000 --> 00:00:45.079
I'm your host, Derek Bruff.
00:00:45.079 --> 00:00:51.000
I hope this podcast helps you be more intentional in how you teach and in how you develop as a teacher over time.
00:00:51.320 --> 00:00:54.119
I'm back with another Take It or Leave It panel.
00:00:54.119 --> 00:00:55.799
This one is a little bit different.
00:00:55.799 --> 00:01:00.759
Y'all have told me many times that you like the take it or leave it episodes of this podcast.
00:01:00.759 --> 00:01:15.719
So, the summer, when my University of Virginia colleagues and I were planning a series of professional development workshops for the fall aimed at faculty who teach online, I suggested that we use the take it or leave it format for one of the events.
00:01:15.719 --> 00:01:31.479
My colleagues liked the idea, and on October 2nd, in my role as Associate Director of the UVA Center for Teaching Excellence, I hosted a virtual panel titled Take It or Leave It, AI's Role in Online Learning, featuring three fantastic UVA colleagues.
00:01:31.479 --> 00:01:39.319
The conversation went very well, and the panelists and the CTE gave me permission to share the audio from the panel here on the Intentional Teaching Podcast.
00:01:39.480 --> 00:01:44.519
The panelists for this edition of Take It or Leave It are all at the University of Virginia.
00:01:44.519 --> 00:01:53.560
Michelle Beavers is an associate professor and coordinator of the Administration and Supervision Program in UVA's School of Education and Human Development.
00:01:53.560 --> 00:02:02.120
Leo Lo is Dean of Libraries and University Librarian, also advisor to the provost on AI literacy and a professor of education.
00:02:02.120 --> 00:02:07.799
Special thanks to Leo for agreeing to do this panel just two weeks after he started his job at UVA.
00:02:07.799 --> 00:02:20.280
Our third panelist is Sara McClellan, Assistant Professor of Professional Studies, and Program Coordinator of the Public Administration Certificate Program at UVA's School of Continuing and Professional Studies.
00:02:20.759 --> 00:02:29.079
For those of you new to the Take It or Leave It format, you'll hear me briefly describe five recent op-eds on teaching and learning in higher ed.
00:02:29.079 --> 00:02:37.960
For each op-ed, I'll ask each of our panelists if they take it, that is, generally agree with the main thesis of the essay or leave it.
00:02:37.960 --> 00:02:49.159
This is an artificial binary that I've found to generate rich discussion of the issues at hand, and that was definitely true for this panel, focused on hot takes about generative AI and online education.
00:02:49.159 --> 00:02:54.199
See the show notes for links to the essays we discussed if you want to read them in their entirety.
00:02:56.359 --> 00:03:04.359
Our first hot take, our first essay to discuss, is titled Are You Ready for the AI University?
00:03:04.359 --> 00:03:10.759
This was published in the Chronicle of Higher Education earlier this year.
00:03:10.759 --> 00:03:17.000
This was written by Scott Latham, who is a professor of strategy at the University of Massachusetts at Lowell.
00:03:17.000 --> 00:03:26.840
And in this piece, Professor Latham casts a fairly big vision for how generative AI will disrupt higher education in the coming years.
00:03:26.840 --> 00:03:33.319
There is a lot, there are a lot of big claims in this essay, but we're going to focus on just one of them.
00:03:33.319 --> 00:03:44.520
And for a bit of context, here's a quote from the essay: faculty will need to train AI in how to help them build lectures, assessments, and fine-tune their classroom materials.
00:03:44.520 --> 00:03:49.400
Further training will be needed when AI first delivers a course.
00:03:49.400 --> 00:03:58.520
While students will readily accept non-human professors, any new technology will have hiccups that require human intervention to resolve.
00:03:58.520 --> 00:04:02.680
And here's the thesis statement I'll ask our panelists to respond to.
00:04:02.680 --> 00:04:07.960
Once the training wheels are off, AI taught courses will become the dominant paradigm.
00:04:07.960 --> 00:04:10.599
That is a pretty bold statement.
00:04:10.599 --> 00:04:12.120
Sara, I'll start with you.
00:04:12.120 --> 00:04:13.719
Do you take that or leave that?
00:04:14.840 --> 00:04:20.680
I leave it in terms of the scope and speed of the argument that this author outlined.
00:04:20.680 --> 00:04:24.759
And I think for two big reasons.
00:04:24.759 --> 00:04:32.279
I think first, some of the predictions come up against energy shortfalls that aren't yet resolved.
00:04:32.279 --> 00:04:45.000
And so if we look, it's not just energy shortfall, it's transmission, it's chips, it's cybersecurity, it's all of the infrastructure and environmental features need needed to scope such a grand vision.
00:04:45.000 --> 00:04:56.519
Uh assuming that if such a vision were to come to pass, it would be an effort all over the country and be putting pressure on all of our systems, which would require addressing mega project problems, political gridlock.
00:04:56.519 --> 00:04:58.440
So that's one.
00:04:58.440 --> 00:05:13.000
And uh yeah, it also, I will say it also I think ignores the um the inconsistencies in human systems, or federalist structure, shared governance, uh, the way that higher education works across states.
00:05:13.000 --> 00:05:19.000
Um I think it's it's a pretty sweeping vision given our current patchwork system.
00:05:19.480 --> 00:05:20.440
Gotcha, gotcha.
00:05:20.440 --> 00:05:21.159
All right.
00:05:21.159 --> 00:05:23.799
Michelle, do you take this or leave this?
00:05:24.360 --> 00:05:25.720
I leave it, Derek.
00:05:26.759 --> 00:05:28.279
And why do you leave it?
00:05:28.600 --> 00:05:39.399
Well, I'm gonna lean towards the um relational aspect of teaching and say that AI does not have the disciplinary um expertise.
00:05:39.399 --> 00:05:45.320
It's that ethical judgment that faculty bring, our pedagogical pedagogical vision that we represent.
00:05:45.320 --> 00:05:51.720
Um and the most important piece is that interaction and relational trust and accountability with our students.
00:05:51.720 --> 00:06:00.840
I could see it as a back-end partner, uh, potential support for async or a flipped classroom environment.
00:06:00.840 --> 00:06:01.720
Okay.
00:06:02.679 --> 00:06:06.519
And you say this as someone with a fair degree of online teaching experience.
00:06:06.519 --> 00:06:11.639
And in fact, someone who uses AI a lot in the development of your learning materials.
00:06:11.960 --> 00:06:14.679
I certainly do, but it's not taking over.
00:06:15.000 --> 00:06:16.120
It's not taking over.
00:06:16.120 --> 00:06:16.600
Okay.
00:06:16.600 --> 00:06:17.559
Good.
00:06:17.559 --> 00:06:19.480
So we've got two leaves that leave it.
00:06:19.480 --> 00:06:22.600
Leo, are you gonna are you gonna complete the trio here?
00:06:22.600 --> 00:06:23.799
Do you take this or leave this?
00:06:24.039 --> 00:06:25.559
Well, I'm gonna go the other way then.
00:06:25.559 --> 00:06:26.840
I'll lean.
00:06:26.840 --> 00:06:28.200
Take it.
00:06:28.840 --> 00:06:29.080
Okay.
00:06:29.639 --> 00:06:33.399
Yes, I'll just make try to make it more uh uh interesting.
00:06:33.399 --> 00:06:42.919
And um Sara hit the the key point that I kind of the crux of this kind of well, why do I lean, you know, uh take it is that the timeline.
00:06:42.919 --> 00:06:47.879
I think right now, in the short term, it's I would I'll I'll leave it.
00:06:47.879 --> 00:06:57.159
But in the long term, I'm actually thinking that maybe a um so I try to compare AI to let's say the internet.
00:06:57.159 --> 00:07:00.360
I think right now we're in the dialogue phase.
00:07:00.360 --> 00:07:04.679
Can you imagine us doing our job now with dialogue technology?
00:07:04.679 --> 00:07:09.799
So it's hard to imagine what it's gonna be like in, let's say, 20 years' time, right?
00:07:09.799 --> 00:07:29.480
If it keeps improving in that way, including um um the you know how it solves some environmental issues and other things, if it keeps going that way, I can see AI being a really good teacher in that way, that it can improve in that way.
00:07:29.480 --> 00:07:32.039
It has unlimited patience.
00:07:32.039 --> 00:07:41.240
If it can improve its accuracy, then yeah, possibly, because I I honestly I'm using AI to learn things now.
00:07:41.240 --> 00:07:43.399
Obviously, low-stakes stuff.
00:07:43.399 --> 00:07:52.919
I'm not using it to teach me how to cure cancer, although it might happen in in the future, but it's hard to speculate 20 years from now.
00:07:52.919 --> 00:07:59.559
But at the same time, what Michelle just said about all the human relations, I think that's always going to be important.
00:07:59.559 --> 00:08:27.399
So the profess the human professor will become less of teaching the uh the technical part, perhaps, but teaching other things that, let's say, a college student might need, coming to college, the social aspect, the you know, the emotional intelligence, uh, even some human judgment part that AI may be may still have difficulty in in kind of understanding or even teaching.
00:08:27.399 --> 00:08:36.759
But in terms of strictly technical stuff, I think AI could could take on that that role in the future, in the longer term, not right now.
00:08:37.240 --> 00:08:37.720
Yeah.
00:08:37.720 --> 00:08:39.080
What's that line about?
00:08:39.080 --> 00:08:45.240
We tend to overestimate the short-term impact of new technologies and underestimate the long-term impact.
00:08:45.240 --> 00:09:02.279
Um, that said, I feel like I've been hearing this refrain for my entire career in higher ed, which is we're gonna have some magic technology that will provide wonderful personalized learning to all of our students and they will learn all the things that they need to know through this technology.
00:09:02.279 --> 00:09:06.680
Um, and it will it will mean we don't have huge class sizes anymore, right?
00:09:06.680 --> 00:09:07.960
It will change everything.
00:09:07.960 --> 00:09:12.040
And I feel like I've been waiting for that to actually happen for 20 years now.
00:09:12.040 --> 00:09:18.360
So um I think I'm a little skeptical that such a thing can exist myself.
00:09:18.360 --> 00:09:20.440
Um, what do you all think about that?
00:09:20.440 --> 00:09:26.200
How would you respond to Leo's claim that yeah, this is something that that we can imagine happening 20 years down the road?
00:09:26.200 --> 00:09:27.480
Michelle or Sarah?
00:09:27.800 --> 00:09:39.720
I think Leo brings up a good point that some of the um practical knowledge-based pieces could provide that one-on-one instruction for students and to support their learning where they are.
00:09:39.720 --> 00:09:53.480
But the piece that it doesn't um recognize is where we negotiate meaning and understanding with our students, where we challenge them, their values, the way they're confronting issues, um, all that still remains human.
00:09:53.480 --> 00:09:58.840
So Leo may have a point that could sway me closer to possibly take it.
00:09:58.840 --> 00:10:05.160
When I think about could we partner and have different roles within the classroom environment?
00:10:05.480 --> 00:10:09.320
AI as an instructional assistant, but not as the sole instructor.
00:10:09.560 --> 00:10:11.560
Still not letting them take my place.
00:10:12.200 --> 00:10:13.000
Right, right.
00:10:13.000 --> 00:10:16.680
I absolutely agree with Michelle just what Michelle just said.
00:10:16.680 --> 00:10:24.360
The the roles, the uh the roles may change, but I think humans always will prefer to have human interaction.
00:10:24.360 --> 00:10:30.920
So yeah, so that's just kind of our our job in kind of defining what that is going to be like moving forward.
00:10:30.920 --> 00:10:31.720
Yeah.
00:10:32.759 --> 00:10:42.200
And I think when we've seen efforts along these lines in the past to try to turn the education over to the robots, whether it was AI or the MOOC craze of the 2010s.
00:10:42.200 --> 00:10:44.279
It does work for some students.
00:10:44.279 --> 00:10:50.759
Like there are students who have sufficient background knowledge and sufficient motivation that they'll kind of take these tools and really learn well.
00:10:50.759 --> 00:11:01.800
But I find that it doesn't work for 85% of the students who need the human relationship, they need the structure, they need the motivation to do that kind of learning.
00:11:01.800 --> 00:11:03.240
Um, I am skeptical.
00:11:03.240 --> 00:11:06.519
Uh uh Latham writes in the piece, I said this earlier.
00:11:06.519 --> 00:11:12.040
While students will readily accept non-human professors, I don't know about that.
00:11:12.040 --> 00:11:14.360
Again, some students will, I'm sure.
00:11:14.360 --> 00:11:15.800
They'll be happy.
00:11:15.800 --> 00:11:17.960
Um, but that that seems like a stretch.
00:11:17.960 --> 00:11:23.399
I think most of our students are not are not wanting to have the AI tutor as their sole instructor.
00:11:24.200 --> 00:11:35.720
That makes me think about the gap we're seeing when Pew and other organizations have studied public perspectives on AI and the perspectives of experts versus the average member of the public.
00:11:35.720 --> 00:11:42.600
There's a pretty big gap, and we're seeing a larger gap than we've seen with the introduction of a lot of other technologies.
00:11:42.600 --> 00:11:46.440
Um so perhaps we'll tip over that gap at some point in certain areas.
00:11:46.440 --> 00:11:58.360
But right now, I certainly hear that I teach mid-career professionals, and I that gap is alive and well as I hear technology experts in my classes interacting with folks who are pretty wary.
00:11:58.360 --> 00:12:02.360
Um, and then looking at that data and talking it through.
00:12:03.000 --> 00:12:04.040
Yeah, yeah.
00:12:04.360 --> 00:12:04.920
All right.
00:12:04.920 --> 00:12:10.440
Okay, so we have two take it, uh, no, two leave it's and a take it for our first one.
00:12:10.440 --> 00:12:12.440
Um, let's move on to our second one.
00:12:12.440 --> 00:12:17.399
I've tried to pick some really interesting essays for us to respond to that kind of go in very different directions.
00:12:17.399 --> 00:12:25.560
Um uh let's see, essay number two is titled AI Risks Undermining the Heart of Higher Education.
00:12:25.560 --> 00:12:34.200
And this is a piece by Zahid Naz, who is a senior lecturer in academic and professional education at Queen Mary University of London.
00:12:34.200 --> 00:12:41.960
Um, and in this piece, Naz expresses a very common concern that I hear from faculty about AI's role in learning.
00:12:41.960 --> 00:12:47.320
A worry that students using AI will avoid the hard work of learning.
00:12:47.320 --> 00:12:47.879
Right?
00:12:47.879 --> 00:12:51.560
The first one was a little bit about AI doing the instruction, the teaching.
00:12:51.560 --> 00:12:54.279
This one is about AI doing the learning.
00:12:54.279 --> 00:13:00.200
Uh Naz writes, uh, the process of learning involves much more than reading texts.
00:13:00.200 --> 00:13:09.480
It requires students to grapple with intricate concepts, compare and contrast ideas, and navigate the nuanced arguments presented in academic literature.
00:13:09.480 --> 00:13:19.320
This sharpens critical thinking, cultivates original thought, and builds the foundation for intellectual independence and what I have identified as the thesis.
00:13:19.320 --> 00:13:28.200
In short, the unchecked use of AI could ultimately undermine the very intellectual rigor that makes higher education meaningful.
00:13:28.200 --> 00:13:30.680
I'll start with you this time, Leo.
00:13:30.680 --> 00:13:32.120
Do you take that or leave that?
00:13:33.399 --> 00:13:34.519
I'll take that.
00:13:34.519 --> 00:13:35.399
I agree with that.
00:13:35.399 --> 00:13:40.519
Um so I and you said something earlier, Derek, about motivation of students.
00:13:40.519 --> 00:13:56.920
And I think a lot of the times, especially with our the current kind of system for them for the masses, is that it's very uh assessment grade, kind of you know, uh uh playing a huge part in why they do certain things.
00:13:56.920 --> 00:14:10.759
Uh of course there are people who are just they just want to learn, they want to improve, they want to grow, but the pressure is really on our students to produce something that will give them a good grade so they can get a good job and and and all of that.
00:14:10.759 --> 00:14:17.160
And that makes it a lot tempting, more tempting to take shortcuts.
00:14:17.160 --> 00:14:22.519
Uh because they just the grades become more important than the learning part in many cases.
00:14:22.519 --> 00:14:44.360
So I think uh unless we can tap into um really in this you know incentivizing the students' motivation part, that we have a good grasp of that, having some a tool that is so easily that can be used so easily to get you know, to get a passing grade or a good grade, I can see people using it.
00:14:44.360 --> 00:14:56.759
So that means we need to have some mechanism to help people wanting to learn, or maybe change the assessment system a little bit so that there's no so high stake for them.
00:14:56.759 --> 00:14:58.840
Um so I I would take that.
00:14:58.840 --> 00:15:06.840
I I agree that you know we that there needs to be some uh checked use uh of AI right now and in the short term.
00:15:07.720 --> 00:15:08.600
Yeah, yeah.
00:15:08.840 --> 00:15:09.879
How about you, Michelle?
00:15:10.759 --> 00:15:12.360
I will say I agree with Leo.
00:15:12.360 --> 00:15:36.440
I'm gonna take the part about the warning about um the unchecked deployment of AI, but I'm gonna leave the idea that AI can degrade learning because I feel like um as faculty, as professors, we have this responsibility to create these guardrails and these scaffolds so that AI can actually augment our student learning rather than replace it.
00:15:36.440 --> 00:15:48.759
Um and if we design that learning that creates productive struggle for our students, we're actually enhancing that thinking and that learning rather than allowing AI to do that.
00:15:48.759 --> 00:15:51.080
But it has to be an intentional design.
00:15:53.399 --> 00:15:58.519
So before I go to Sarah, what what is what what does that look like in your courses, Michelle?
00:15:58.519 --> 00:16:03.639
What are you what are some design elements you use to try to move students to that productive struggle?
00:16:04.200 --> 00:16:14.519
So um I will say that first of all, that um I use AI as a cognitive partner and it's a teammate within my um classroom.
00:16:14.519 --> 00:16:23.000
So students work collectively with their human friends, and then they're allowed to invite AI into the conversation as that third critical partner.
00:16:23.000 --> 00:16:43.399
Um I'm also dabbling with the AI agent in which students can interact and receive feedback on the work that they're doing, whether it's a process or a structure they don't understand, training that agent to not give the answer but to ask questions that leads the student to find their own answer.
00:16:43.399 --> 00:17:06.440
Um I think when we create that ambiguity, that um the the requirement to compare a, for example, in an essay assignment, compare an AI outcome to your own outcome and give the rationalization of why you agree or you disagree helps to enhance that thinking process as opposed to saying, oh, you can use AI for this assignment.
00:17:06.440 --> 00:17:12.359
Scaffold it, structure it, and design it, helping students understand the why behind the practice.
00:17:12.359 --> 00:17:16.279
So looking at the process, not just that outcome.
00:17:16.759 --> 00:17:17.079
Right.
00:17:17.079 --> 00:17:26.119
And so if the unchecked use is problematic, your checks involve the structures and sequences that you're asking students to work through, right?
00:17:26.119 --> 00:17:35.160
Um, as well as what I heard is some checks on the AI itself to kind of point it in useful directions and not uh less helpful directions.
00:17:35.559 --> 00:17:35.880
Correct.
00:17:35.880 --> 00:17:49.720
And the only other thing I'll add because Leo said it, he said emotional intelligence, and I gave a workshop this morning on emotional intelligence, and we used AI to demonstrate our emotional intelligence by having this interactive conversation.
00:17:49.720 --> 00:17:55.079
So AI is capable of not only the technical skills but the dispositional skills as well.
00:17:56.360 --> 00:17:56.759
All right.
00:17:56.759 --> 00:17:58.600
I'm gonna go to you now, Sara.
00:17:58.600 --> 00:18:01.240
Take it or leave it.
00:18:01.799 --> 00:18:13.240
I I take it, and one thing I'm I'm really aware of right now is that I teach relatively small classes, 15, 20, 25 students.
00:18:13.240 --> 00:18:17.160
Um, I have a great um teaching excellence center.
00:18:17.160 --> 00:18:19.480
I have access to lots of resources.
00:18:19.480 --> 00:18:26.839
I'm communicating with friends and colleagues around the country who work in really big systems that are pushed to the absolute max.
00:18:26.839 --> 00:18:40.039
They're sometimes teaching dozens or hundreds of students in a single class with limited TA support uh and don't necessarily have strong instructional support at the university for transitions.
00:18:40.039 --> 00:18:46.200
And so uh I think I struggle at the individual level and where resources are there.
00:18:46.200 --> 00:18:59.000
I think that there will be ways, as Michelle said, to make this um a really strong learning tool and and create spaces strategically where students can lean into that tool and and reflect on it.
00:18:59.000 --> 00:19:06.200
Um, but I try to think about I think I suspect that will be pretty spotty and there'll be systems where that's much, much harder.
00:19:09.880 --> 00:19:11.079
I'm curious, Sara.
00:19:11.079 --> 00:19:15.960
So so Leo pointed to the kind of the role of motivation in all this, right?
00:19:15.960 --> 00:19:21.799
And if students are just box checking, then yeah, it's easy to use use some unchecked AI.
00:19:21.799 --> 00:19:25.000
Um you mentioned teaching mid-career professionals.
00:19:25.000 --> 00:19:28.920
Do you find that they come in wanting a certain degree of rigor?
00:19:28.920 --> 00:19:34.039
And that makes it easier then to kind of avoid the the AI shortcuts.
00:19:35.240 --> 00:19:37.240
You know, I would I would say it's a mixed bag.
00:19:37.240 --> 00:19:53.160
I would say the majority of students uh want a reasonable degree of rigor, but I definitely get those students who are in the program because they've been pushed to be in the program around a promotional opportunity, or somebody's told them that they really need to check that box uh professionally.
00:19:53.160 --> 00:20:04.920
And um, those do tend to be the students that I sometimes have conversations with about AI use and and um to the extent to which they're leaning on AI, but they're also really busy professionals.
00:20:04.920 --> 00:20:23.240
So I think the temptation, I have a lot of students who are emergency services worker, and so the temptation is I've just worked, you know, a crazy shift and I'm exhausted and I'm trying to do 10 different things, and now there's another emergency, and so AI looks really good at that point, and uh that's that's a hard one.
00:20:25.079 --> 00:20:25.480
Thank you.
00:20:25.480 --> 00:20:31.720
I think sometimes I have this vision of like the the first year undergraduate taking a writing course they don't want to have to take.
00:20:31.720 --> 00:20:35.640
And that's that's a hard, hard course to teach when AI is out there.
00:20:35.640 --> 00:20:41.160
And sometimes I feel like the professional studies programs must just it's just all you know, sunshine and roses there.
00:20:41.160 --> 00:20:42.839
Like they all want to be there and learn.
00:20:42.839 --> 00:20:44.759
Um, but we're all humans, right?
00:20:44.759 --> 00:20:48.039
We don't we don't show up with our best selves every day in class.
00:20:48.039 --> 00:20:48.839
Yeah.
00:20:48.839 --> 00:20:49.960
Yeah.
00:20:49.960 --> 00:20:53.240
Um Sarah, one one other follow-up.
00:20:53.240 --> 00:20:59.240
We talked about some of the the checks that Michelle has in terms of the structures and the tools that she's using.
00:20:59.240 --> 00:21:07.400
What are what are some of the checks that you have in your courses to try to kind of orient students towards a more rigorous application of learning?
00:21:08.120 --> 00:21:09.400
Well, I'm experimenting.
00:21:09.400 --> 00:21:16.600
This term is the first term that I did a brief, like three to four minute video right at the start of my public policy course.
00:21:16.600 --> 00:21:18.519
And I said, let's talk about AI.
00:21:18.519 --> 00:21:23.079
Let's talk about the get to badly and how we can use it in policy analysis.
00:21:23.079 --> 00:21:28.839
And so I sort of stepped right out there and um students really responded to that.
00:21:28.839 --> 00:21:33.880
And I created an AI agent, a tool that students could use around policy analysis.
00:21:33.880 --> 00:21:40.360
And I fed that tool and I explained to them how I did it with readings and concepts and frameworks from the course.
00:21:40.360 --> 00:21:50.440
And then I asked them to please not use that tool until a very specific point in the course, because I didn't want the tool to take the learning away from them.
00:21:50.440 --> 00:21:56.120
Uh and I said, you know, policy practice is a way of thinking about and approaching problems.
00:21:56.120 --> 00:22:01.640
And if we let AI learn how to approach this problem, you know, we're paying for a lot and we're showing up and we're not learning.
00:22:01.640 --> 00:22:09.720
And so I said, but it's a fantastic partner to sort of think through and challenge the assumptions that you have and the research research that you're using.
00:22:09.720 --> 00:22:14.519
And so I tried to give them strategic places where I said, please, please test the tool here.
00:22:14.519 --> 00:22:16.120
Tell us how it's working.
00:22:16.120 --> 00:22:24.120
Um, and so far I think that's I think it's going um much better than it did in the last term, where it was a little bit more hands-off about it.
00:22:24.920 --> 00:22:25.480
Yeah.
00:22:25.480 --> 00:22:27.880
Because you're saying we're gonna use AI.
00:22:27.880 --> 00:22:47.000
It's not like we're gonna pretend it doesn't exist, but we're gonna be thoughtful and intentional about how we use AI, not to replace our own work, but maybe as as Michelle was saying, as a conversation partner, as a as a uh um uh a tool that can kind of provoke us with interesting ideas, perhaps.
00:22:47.319 --> 00:22:50.519
Yes, and I linked to articles about AI and motivation.
00:22:50.519 --> 00:22:54.839
Um, and several of them responded it was optional reading, but they clearly did it.
00:22:55.240 --> 00:22:56.519
Yeah, yeah.
00:22:56.519 --> 00:23:02.680
Um Leo, anything to add after hearing Michelle and Sara talk about their approaches?
00:23:03.079 --> 00:23:05.880
No, I absolutely agree with with those approaches.
00:23:05.880 --> 00:23:19.079
I think um ultimately, but ultimately it is about setting up, like you said, some guardrails, some kind of guidance on how to actually use the tool to to learn and um benefit them.
00:23:19.079 --> 00:23:23.319
Um and you know, some people will never do that.
00:23:23.319 --> 00:23:27.720
They was, you know, they will find ways to, you know, uh, like you said, Derek, we're all humans.
00:23:27.720 --> 00:23:29.720
Sometimes we don't show up with our best selves.
00:23:29.720 --> 00:23:39.559
But at the same time, that I also believe there are many people who really wanting to learn just you know, in this kind of uh environment, don't know how or what's the best way.
00:23:39.559 --> 00:23:49.319
So that I love that, you know, Sarah and Michelle are really giving them uh ways to do that, to optimize their experience using the technology.
00:23:49.720 --> 00:24:03.240
And I think one thing I've observed is as these new technologies come on, it takes higher ed a little time to figure out what are some of better practices, what are what are they good for, what are they bad for, what are the structures that we could put into place?
00:24:03.240 --> 00:24:07.640
Um you know, there's some collective figuring out that happens.
00:24:07.640 --> 00:24:10.440
Well, and that's a great segue into our next essay.
00:24:10.440 --> 00:24:14.600
And Leo, I'm coming for you on this one first, because this is this is your wheelhouse.
00:24:14.600 --> 00:24:17.400
Um, this one is about AI literacies.
00:24:17.400 --> 00:24:23.000
Uh, this is an essay by Ray Schroeder called The Urgent Need for AI Literacy.
00:24:23.000 --> 00:24:28.200
Uh, Ray Schroeder is a senior fellow at UPCEA, the online and professional education association.
00:24:28.200 --> 00:24:39.160
And in this piece, he cites a number of workforce studies pointing to the growing role of AI and the need to prepare students to have AI literacies.
00:24:39.160 --> 00:24:57.400
Um he says uh the rapid advent of AI capabilities coupled with the developing economic pressures worldwide have led to a surge in employers seeking to reduce operating expenses through widespread use of generative and agentic AI to augment and in some cases replace humans in their workforce.
00:24:57.400 --> 00:25:00.759
And then what I've identified, there's kind of an implied thesis here.
00:25:00.759 --> 00:25:03.400
So I'm going to read the statement and then frame it up a little bit.
00:25:03.400 --> 00:25:14.759
He says we are failing to fully prepare those students to enter the workforce, where, as the World Economic Forum says, two-thirds of business leaders surveyed said they wouldn't hire a candidate without AI skills.
00:25:14.759 --> 00:25:24.200
And nearly three-quarters said they would rather hire a less experienced candidate with AI skills than a more experienced candidate without them.
00:25:24.200 --> 00:25:29.400
And so I'm going to focus on the first part of that thesis statement, which I'll drop in the chat.
00:25:29.400 --> 00:25:34.680
We are failing to prepare, failing to fully prepare those students to enter the workforce.
00:25:34.680 --> 00:25:38.120
Um Leo, would you take that thesis or leave it?
00:25:39.240 --> 00:25:43.400
Let me share a recent experience first, and then I'll answer answer you.
00:25:43.400 --> 00:25:50.039
Uh this week I I was at a gathering with uh a small group of, I would say AI experts.
00:25:50.039 --> 00:25:51.880
They are very close to the action.
00:25:51.880 --> 00:26:05.000
And um it was quite depressing, actually, um, because they thought, really believed, that in about 18 months' time, uh pretty much all the entry-level jobs will be gone.
00:26:05.000 --> 00:26:09.720
And you can see it kind of happening now, but they are very close to the action.
00:26:09.720 --> 00:26:17.400
And I quote one of them saying, you know, when they saw the um the new newest models, I quote, almost like magic.
00:26:17.400 --> 00:26:21.160
So um you can kind of okay, I put some weight on what they're saying.
00:26:21.160 --> 00:26:22.440
Let's just say, assume that.
00:26:22.440 --> 00:26:34.120
So, in that, if that is true, then I'll take it that we are not really preparing our students because what students come to a university like UBA so that they can get good jobs.
00:26:34.120 --> 00:26:47.160
We want them to you know prepare them for the workforce so that they learn the skills and also all the other things that can help them succeed in getting that entry-level job and then succeeding you know in the uh for the rest of their lives.
00:26:47.160 --> 00:26:50.600
But if those jobs are gone, then what are we doing?
00:26:50.600 --> 00:26:53.000
Why are they paying the money to come here?
00:26:53.000 --> 00:27:01.960
Yes, there are other benefits of going to college, but at the same time, you pick out that much money to pay for something, you want some tangible return.
00:27:01.960 --> 00:27:05.720
So here's some of my, in a way, my thinking.
00:27:05.720 --> 00:27:09.319
I don't know, I would love to get everybody's take on this, right?
00:27:09.319 --> 00:27:16.440
So for the past, I don't know, 100, 200 years, we're preparing students to get jobs because of industrial revolutions.
00:27:16.440 --> 00:27:18.680
You have jobs and people want to get those.
00:27:18.680 --> 00:27:23.160
But before all of that, everybody was kind of their own entrepreneurs.
00:27:23.160 --> 00:27:26.920
You're a blacksmith, you're a farmer, you're all these, you work for yourself.
00:27:26.920 --> 00:27:29.559
Would we go back to that?
00:27:30.360 --> 00:27:30.759
Right?
00:27:31.079 --> 00:27:36.519
If all the jobs, entry-level jobs are gone, AI can do all of that.
00:27:36.519 --> 00:27:46.120
But at the same time, everybody, every one of us, every one of the students can also have a team of AI agents to be their own bosses.
00:27:46.120 --> 00:27:53.079
As long as there are problems in the world, there is always a demand for solutions, however big or small.
00:27:53.079 --> 00:28:08.839
If there are demand for solutions, there are always opportunities for these, I would call them micro solopreneurs, that they can use a team of AI agents uh to help them solve real-world problems, however big or small.
00:28:08.839 --> 00:28:24.600
Shouldn't we be teaching them how to do that and using the technology to help them, you know, build up that kind of scope of the problem, build using infrastructure support here to learn how to do that, and then go out there and be their own bosses in many ways.
00:28:24.600 --> 00:28:40.600
I understand that not 100% of the students would want to do that, but even if there are 10%, and if there's such a radical change in the society, maybe, you know, that we need to have some kind of radical changes as well to adapt to that.
00:28:40.600 --> 00:28:43.079
I am just brainstorming right here.
00:28:43.079 --> 00:28:53.000
So I that much, but it's just an idea that I've been really kind of simmering in terms of how teachers in in AI literacy.
00:28:53.000 --> 00:28:58.200
Of course, they need to be AI literate to be able to do any of those things I just talked about.
00:28:58.200 --> 00:29:00.680
Or even just to get jobs if they ask you a bit.
00:29:01.880 --> 00:29:02.360
Yeah.
00:29:02.360 --> 00:29:07.000
Well, and I've heard a lot of people say the entry-level jobs in this field or that field are disappearing.
00:29:07.000 --> 00:29:10.279
I haven't heard a lot of what then do we do?
00:29:10.279 --> 00:29:13.799
Um, so I appreciate your brainstorming with us.
00:29:13.799 --> 00:29:16.440
Uh Sara, how would you respond to this thesis?
00:29:16.440 --> 00:29:20.360
Are we failing to fully prepare our students to enter the workforce?
00:29:22.440 --> 00:29:24.039
Yes, I'm afraid so.
00:29:24.039 --> 00:29:24.920
I would take that.
00:29:24.920 --> 00:29:33.079
And I think um, Leo, you hit on uh a lot of themes that I've been hearing about and and reading about as well.
00:29:33.079 --> 00:29:40.120
Um I think one thing that really strikes me is that AI literacy, it's not just a technology.
00:29:40.120 --> 00:29:50.839
It is because, as Leo suggested, I think it is so fundamentally going to change our social fabric, our economy, what it means to be successful professionally.
00:29:50.839 --> 00:30:08.039
Um, that I think part of AI literacy for me is um change, navigating change and helping students figure out how to kind of surf with disruptions and and not seize up because we're gonna have a lot of change.
00:30:08.039 --> 00:30:10.360
Um it's already here, we'll have more.
00:30:10.360 --> 00:30:17.640
And so I think um, and that's intellectually and emotionally, it can be fun and it can also be exhausting.
00:30:17.640 --> 00:30:23.240
And I think recognizing where kind of the human condition is, we've come out of a global pandemic.
00:30:23.240 --> 00:30:29.480
Um, you know, we're in politically tumultuous times, and so, and we have institutions.
00:30:29.480 --> 00:30:48.839
That were, you know, industrial or institutions that ironically they helped sort of create or unleash, depending on how you look at it, all of these technologies, but they're now socially accelerated to a point that our institutions aren't built to manage, to develop public policy, to do quick responses.
00:30:48.839 --> 00:30:52.279
And that creates a lot of tensions.
00:30:52.279 --> 00:30:58.519
And so I think that a big piece of this is figuring out when to go fast and when to go slow.
00:30:58.519 --> 00:31:08.920
There's a huge temporal element to all of this because reason, judgment, our democratic process, our shared governance system, it all takes time.
00:31:08.920 --> 00:31:26.759
And so I think when we think about AI literacy, not just for students, for faculty, for administrators, it's got to be about human strategy and change strategy and trying to sort of assess what's coming and get much, much more creative than we've been.
00:31:26.759 --> 00:31:33.319
And so I feel like Leo, we could have a whole session on just what are the creative possibilities when those entry-level jobs go away.
00:31:33.319 --> 00:31:38.759
And I think we need a big, big temp approach to trying to solve those problems.
00:31:39.079 --> 00:31:39.559
Michelle.
00:31:40.279 --> 00:31:42.680
Struggling with how to respond to this, Derek.
00:31:42.680 --> 00:31:49.000
For everyone joining us today, I'm an educator for an education society.
00:31:49.000 --> 00:31:56.360
So I'm looking at preparing the folks that work in educational institutions, or mean meaning K-12 education.
00:31:56.360 --> 00:32:07.960
And I recently read an article that in 2023, only 50% of school divisions have either said, yes, I'm going to use AI or no or not, but haven't developed policy yet.
00:32:07.960 --> 00:32:13.319
So, in quite honest, in full honesty, education is behind the eight-ball.
00:32:13.319 --> 00:32:15.319
We're moving slow already.
00:32:15.319 --> 00:32:18.600
We're not keeping up with what the business world is doing.
00:32:18.600 --> 00:32:24.440
And what I'm seeing here is that school divisions are looking for the tools.
00:32:24.440 --> 00:32:26.680
They're not looking at the skills.
00:32:26.680 --> 00:32:33.000
And so therefore, we're not the they want to know how to make their life easier.
00:32:33.000 --> 00:32:35.880
What tools should we adopt or not adopt?
00:32:35.880 --> 00:32:41.880
And they haven't gotten to the point of recognizing that it's not the tools that we need to teach people how to use.
00:32:41.880 --> 00:32:47.960
It's the AI skill, the AI literacy, the judgment, the critical thinking, the metacognition, et cetera.
00:32:47.960 --> 00:32:53.480
So I'm kind of like sitting on a double-edged sword here, not really sure which way to go.
00:32:53.480 --> 00:33:01.400
And I think the other piece that we really have to consider is you're looking at a panel of three individuals.
00:33:01.400 --> 00:33:03.160
Well, what if you looked at the system?
00:33:03.160 --> 00:33:07.480
How many people are actually integrating teaching AI skills?
00:33:07.480 --> 00:33:17.079
And if we had to say that, I'd say I'd take it because if I look at my own department or my own school of education, not everybody is at that place yet.
00:33:17.079 --> 00:33:19.079
And so we need to be.
00:33:19.180 --> 00:33:19.180
I would add, to your last point, that most of the AI policy discussions I've heard in higher ed, and the AI strategy responses are at the individual course level and the individual instructor, how can they make better choices for their courses? What policy do you set for your course? But our students don't experience courses as islands. They enroll in programs that are hopefully developing a whole skillset in our students over time. If our programs don't have coherent responses to these challenges, smart choices by faculty can only go so far if students are experiencing very different conditions in the courses in their program.
00:33:19.279 --> 00:33:19.279
Yeah that ambiguity across programs is a problem. And it gets even deeper when you start talking about dissertations and capstones and what's permitted and what's not permitted.
00:33:19.380 --> 00:33:19.380
So I think you all said take it, that maybe we need to do more and move faster. I'm going to move us on to our fourth essay which I think takes a different approach. This is a piece published I think last week in the Chronicle of Higher Education.
00:34:41.480 --> 00:34:44.519
Yes, that was last week, I think, by Kevin Gannon.
00:34:44.519 --> 00:34:48.519
It's called Sometimes We Resist AI for Good Reasons.
00:34:48.519 --> 00:34:55.079
Uh Kevin Gannon is a professor of history and director of the teaching center at Queen's University of Charlotte.
00:34:55.079 --> 00:34:57.719
Um I've met Kevin, he's a teaching center guy.
00:34:57.719 --> 00:35:02.359
Um his his online handle is the Tattooed Professor, and it is very accurate.
00:35:02.359 --> 00:35:04.359
He is covered in tattoos.
00:35:04.359 --> 00:35:19.559
Um Kevin uh takes his essay, most of the essay details four big questions related to AI policy in higher education, where he argues the voices of AI skeptics and even resistors need to be involved.
00:35:19.559 --> 00:35:27.639
He writes on top of the dizzying rate of technological change is a relentless boosterism surrounding generative AI.
00:35:27.639 --> 00:35:36.039
As faculty members, we are told that there is no alternative, that these tools are the future of work, that we must either adapt or become obsolete.
00:35:36.039 --> 00:35:43.000
That framing is responsible, however, for the exclusion of important perspectives from our institutional conversations on AI.
00:35:43.000 --> 00:35:57.959
And so the thesis statement I'll ask you to respond to is rather this is a call to make sure that everyone is at the table when you are setting these policies, even and especially the harshest campus critics of AI.
00:35:57.959 --> 00:36:00.359
Michelle, I'll start with you.
00:36:00.919 --> 00:36:03.959
Um I will take it and I'll take it strongly.
00:36:03.959 --> 00:36:07.639
Um I feel like this is a shift for me.
00:36:07.639 --> 00:36:19.319
I started using AI um in the summer of 22, excuse me, 23, and um I was an early adopter and very enthusiastic about it.
00:36:19.319 --> 00:36:48.279
But it wasn't until I sat amongst the resistors that I recognized the importance of slowing down, the importance of hearing dissenting views, um, especially those who are skeptical and critical of AI, because it helps to build a better policy and practice for the whole organization, but for myself too, who would be ready green light, jump in, let's go, when in reality, it might be time to pause and consider how to do it more effectively.
00:36:48.679 --> 00:36:49.719
Sara, how about you?
00:36:50.039 --> 00:36:53.239
I would take it and I partly I own this.
00:36:53.239 --> 00:36:55.719
Um maybe I'm a closet contrarian.
00:36:55.719 --> 00:37:00.119
Uh, I have I'm I'm ambivalent about artificial intelligence.
00:37:00.119 --> 00:37:04.039
I've jumped into it, I'm working with it, I'm learning with my students.
00:37:04.039 --> 00:37:17.639
Um and I, as I look at other technologies like social media and everything we've learned over the course of the last 15 plus years, um, I if I could push that button and shut it all down, I might do that.
00:37:17.639 --> 00:37:20.839
Uh so, you know, so don't give me that power.
00:37:20.839 --> 00:37:34.839
Um, but one of the things that I think about a lot is I think it's I felt like it was important as a contrarian that I stepped into the process and that I be tried to become AI literate.
00:37:34.839 --> 00:37:37.480
Uh, and and thus my ambivalence.
00:37:37.480 --> 00:37:39.879
I would, I guess I'm not a solid contrarian anymore.
00:37:39.879 --> 00:37:41.079
I'm a bit ambivalent.
00:37:41.079 --> 00:38:02.199
And um I so I think the friendly amendment for me with the take it is that I would expect if we engage contrarians, that we ask or expect them to step up and work on some AI literacy in the process so that people aren't just uh hardstop no uh without understanding much about the technology.
00:38:05.879 --> 00:38:06.199
Right.
00:38:06.199 --> 00:38:15.079
I have certainly encountered the contrarians and the resistors who read a couple of headlines about AI three years ago and made up their mind then that it was useless.
00:38:15.079 --> 00:38:17.639
And I feel like that's that's not an informed position.
00:38:17.719 --> 00:38:24.119
Um it's hard to critique something that you don't understand very well and you haven't experimented with.
00:38:24.119 --> 00:38:25.079
Yeah.
00:38:25.879 --> 00:38:26.839
Leo, how about you?
00:38:26.839 --> 00:38:28.119
Do you take this or leave it?
00:38:28.519 --> 00:38:29.719
Oh, I'll take it.
00:38:29.719 --> 00:38:40.839
Um, I even though I can see enormous potential with the technology, we I think we can all agree that it has some inherent flaws, limitations, and a lot of problems with it.
00:38:40.839 --> 00:38:46.919
And we're in this rapidly growing stage, but it hasn't grown yet in certain ways.
00:38:46.919 --> 00:38:49.799
So this is the time to make changes, right?
00:38:49.799 --> 00:38:56.039
To make sure that you know we actually have that window um to set some rules.
00:38:56.039 --> 00:38:58.759
Once the rules are set, they're difficult to change.
00:38:58.759 --> 00:39:00.519
But this is the time to set those rules.
00:39:00.519 --> 00:39:13.079
And so I encourage people who have uh uh who are skeptical of this technology or have you know want to be critical about it to really raise their, you know, to have their voice heard, basically.
00:39:13.079 --> 00:39:17.399
But at the same time, I agree with Sarah, basically people need to be AI literate first.
00:39:17.399 --> 00:39:27.879
Because and I always I say that to everybody, is whether you're for the technology, technology or if you're against it, your arguments are going to be stronger if you know more about it.
00:39:27.879 --> 00:39:49.559
And that's what I'm really advocating for, is for for all of us to learn more, to be more AI literate, to learn more about it, and from the librarian's perspective, where to find credible information so that it can substantiate your arguments and all of that, so that your voice becomes stronger to make the technology better, to make the changes better.
00:39:49.559 --> 00:39:55.639
Because right now we're in a period of chaos, in a period of rule setting, and there aren't that many rules.
00:39:55.639 --> 00:40:02.359
Even if you look at countries and regions with regulations, it's chaotic chaotic right now.
00:40:02.359 --> 00:40:05.719
So is this this is time to help shape that.
00:40:05.719 --> 00:40:07.879
So I take it.
00:40:09.239 --> 00:40:16.279
I think um there's a question in the chat from Tom about how this relates to um online teaching.
00:40:16.279 --> 00:40:36.119
And I think that um by bringing these naysayers, these dissenters into the conversation, especially if they are AI, if they are faculty who teach online, it's critically important going back to that first article about will AI be the instructor for our courses online in the future.
00:40:36.119 --> 00:40:47.079
And like Leah said, faculty have to be AI literate in order to partner with AI in online environments or face-to-face environments.
00:40:48.199 --> 00:40:52.919
Yeah, I uh along those lines are in relationship to teaching for me.
00:40:52.919 --> 00:41:13.079
Um, I'm co-teaching a public sector AI leadership course, and one of the things we're doing is working to integrate um AI critics into our curriculum, including from the tech industry, experts who are leaders in AI who are now or or have been raising a flag and highlighting significant risks.
00:41:13.079 --> 00:41:26.119
And so I want us to be able to talk about those in classes that involve government leaders making big decisions about AI, because otherwise tech vendors have enormous power and we don't understand the big arguments.
00:41:28.839 --> 00:41:38.759
I'm gonna take us to our last essay, and I want to use Tom's question as a jumping-off point because I do want to talk concretely about what online teaching and learning looks like in this AI era.
00:41:38.759 --> 00:41:45.639
Um this essay is called On AI We Reap What We Sow.
00:41:45.639 --> 00:41:52.039
This is a piece by Chad Hansen, who teaches sociology and religion at Casper College in Wyoming.
00:41:52.039 --> 00:42:01.959
He makes the case that higher ed has focused far too much on outcomes and products of learning and not on the value of the process of learning itself.
00:42:01.959 --> 00:42:05.239
Um He says, Enter AI.
00:42:05.239 --> 00:42:11.959
Today we live in an era in which students can feed a prompt into an automated prose generator and in seconds have a viable draft of a writing assignment.
00:42:11.959 --> 00:42:13.959
What are they supposed to think?
00:42:13.959 --> 00:42:19.159
We've spent three decades acting like outcomes assessments are the only things we value.
00:42:19.159 --> 00:42:22.919
And here's the thesis statement I'll ask you to respond to.
00:42:22.919 --> 00:42:38.279
We should see the AI era as providing us with a reason and an opportunity to expand our interest to include an analysis of the broadly formative processes involved in education as opposed to focusing solely on narrow sets of outcomes.
00:42:38.279 --> 00:42:47.480
And I'll ask uh Sarah, do you take that and do you leave it and and what does that look like in online ed, right?
00:42:47.480 --> 00:42:50.519
This this kind of tension between outcomes and process.
00:42:51.959 --> 00:42:59.239
I take that absolutely, and I think in in online ed, I think we're potentially even more at risk.
00:42:59.239 --> 00:43:13.239
Um I think we have often marketed ourselves to understandably to professionals who have very um specific career goals and they they tie the coursework and the degrees or certificates to those career goals.
00:43:13.239 --> 00:43:19.000
And um, some people who are very busy, and and we often have students who want to get through quickly.
00:43:19.000 --> 00:43:25.639
Uh, and I think we have often supported that with our business models, and I felt conflicted about it.
00:43:25.639 --> 00:43:54.359
Um, I think one of the things that uh I totally taken this is that rethinking what education is, I think we might loop back to the comments Leo was making about um, you know, what are the bigger questions that if people are going to struggle to get some of the traditional jobs or to get the or to follow a sort of straight career path, um, then what does that mean for education and especially online, uh, say adult learning?
00:43:54.359 --> 00:44:01.159
Uh, because I think we have marketed ourselves as all about jobs, jobs, jobs.
00:44:01.159 --> 00:44:03.879
And that has concerned me for some time.
00:44:03.879 --> 00:44:21.480
And you take like communication scholars like Stan Dietz have uh he's talked about corporate colonization and this idea that we've sort of taken corporate logics around productivity and efficiency and career progression, and we've absorbed those as the goals for so many other institutions, including education.
00:44:21.480 --> 00:44:27.959
Uh, and I think that links to the obsession with outcomes and efficiency and getting people through.
00:44:27.959 --> 00:44:50.679
And while I don't want to exclude the pragmatic reasons to do that, I also think it's part of how we end up at this moment, um, where bigger questions about what it means to be human, what it means to be embodied, what it means to work and live in a liberal democracy, um, I think we've gotten really rusty at making education about these big topics as well.
00:44:50.679 --> 00:44:59.879
Uh, and I'm hoping that AI this moment will give us a chance to say, once again, we've asked for centuries, what is this thing, this education?
00:44:59.879 --> 00:45:01.319
Uh, what is it good for?
00:45:01.319 --> 00:45:06.039
And and who are we as human beings, uh, especially in relationship to AI?
00:45:07.959 --> 00:45:10.599
Leo, do you take this or leave us?
00:45:11.000 --> 00:45:11.959
Oh, I take it.
00:45:11.959 --> 00:45:15.399
I definitely see see right now as an opportunity.
00:45:15.399 --> 00:45:30.519
Um I have grown up with the education that is very standardized, and the beauty, the potential of AI is uh the personalization, the customization of the technology.
00:45:30.519 --> 00:45:33.639
This could really transform teaching and learning.
00:45:33.639 --> 00:45:41.959
Um, we're just playing with it right now, and the technology is not probably not quite there yet, but moving in a few years and a decade's time.
00:45:41.959 --> 00:45:48.759
It could, I mean, why do we need any kind of standardized assessment anymore for the standardized jobs?
00:45:48.759 --> 00:45:57.239
I mean, I think I mentioned a couple of things in my different answers, in that there may be a future where it's less standardized.
00:45:57.239 --> 00:46:07.959
And um, that means the teaching and learning, we could be freed from producing things that are to satisfy those kind of requirements that we have now.
00:46:07.959 --> 00:46:10.919
I hope that's the way we're going, to be honest.
00:46:10.919 --> 00:46:17.079
I you I mean, education, I mean, I don't think anybody thinks current education is perfect.
00:46:17.079 --> 00:46:22.599
It's kind of you know, teaching to the to the in some ways, the average, the masses in some ways.
00:46:22.599 --> 00:46:30.039
So it a lot of marginalized groups, people, you know, a little bit of outliers, they they they don't get the benefits of it.
00:46:30.039 --> 00:46:37.399
And with AI, there is a possibility, there is a potential that everyone gets that benefit that's personalized.
00:46:37.399 --> 00:46:40.359
So I um I want to see that.
00:46:40.759 --> 00:46:41.399
Yeah.
00:46:41.399 --> 00:46:42.599
Thank you, Leo.
00:46:42.599 --> 00:46:43.399
Michelle?
00:46:43.799 --> 00:46:45.399
I'm gonna take it as well.
00:46:45.399 --> 00:46:48.279
Um, for me, this has been an exciting shift.
00:46:48.279 --> 00:46:53.639
My courses are heavily writing intensive, so I'll use writing as the example.
00:46:53.639 --> 00:47:04.359
Um, I have definitely undervalued that process, that formative part of evaluation, looking at drafts and revisions and productive struggle, et cetera.
00:47:04.359 --> 00:47:19.000
But now that I've integrated AI into that process, we use the draft changes features where students can actually engage in that critical thinking and processing of accuracy and inaccuracies and et cetera.
00:47:19.000 --> 00:47:29.079
And what I've seen is um I understand why the gaps are in that polished product at the end now, or previously I couldn't see that process.
00:47:29.079 --> 00:47:33.399
Um, so I think that that's a really valuable contribution of AI.
00:47:33.399 --> 00:47:35.879
And so I see it kind of as a dual prong.
00:47:35.879 --> 00:47:40.599
Um, AI, I don't want it to relieve us of that polished end product.
00:47:40.599 --> 00:47:52.839
I do want a polished end product, but I think it allows us to reclaim that idea of um formative assessment and learning in the process as they reach that final product.
00:47:54.359 --> 00:47:58.359
I'm reminded of a talk I heard Randy Bass give years ago.
00:47:58.359 --> 00:48:01.879
He's a vice provost at Georgetown University now.
00:48:01.879 --> 00:48:06.279
Um, and he was telling about an American studies course that he was teaching.
00:48:06.279 --> 00:48:09.399
He was having his students create these short video assignments.
00:48:09.399 --> 00:48:10.759
This was almost 20 years ago.
00:48:10.759 --> 00:48:13.319
So this was a big lift technically to do that.
00:48:13.319 --> 00:48:24.679
Um, and they were trying to make arguments through these short video assignments, and he would watch their final products and he said, It's like there was learning left on the cutting room floor that he wasn't privy to.
00:48:24.679 --> 00:48:26.759
He only saw their final product.
00:48:26.759 --> 00:48:31.319
He didn't see the decisions they made, the things they decided not to include, right?
00:48:31.319 --> 00:48:35.799
The arguments they started and then decided, no, this isn't isn't the right way, right?
00:48:35.799 --> 00:48:41.799
Um, and so um anyway, I I feel like yes, the process is really important.
00:48:41.799 --> 00:48:47.079
Um, and maybe it's it's yeah, AI is pointing us to that, certainly.
00:48:47.079 --> 00:48:48.119
Sarah?
00:48:48.839 --> 00:48:58.599
One one thing I I'm I've been wrestling with lately around um the customized learning, which I think I I agree, Lou, is one of the really exciting things about AI.
00:48:58.599 --> 00:49:12.919
Um, the sort of uh flip side or double-edged that I worry a little bit about how to balance is um I could see us move towards a form of customized learning that means students don't have as much shared experience.
00:49:12.919 --> 00:49:28.119
And so I just I'm thinking a lot in my classes about uh you know universal design and giving students options and allowing and encouraging them to use AI in different ways and getting excited about the even more customized opportunity.
00:49:28.119 --> 00:49:43.239
But then I see places where I think, oh my goodness, if we're if we're too much working at different paces and using different tools and creating different projects, the opportunity to have like in communication, we talk about a boundary object, a sort of shared object for learning.
00:49:43.239 --> 00:49:45.239
I I could we could lose that.
00:49:45.239 --> 00:49:49.000
So I'm I'm trying to think about how to create that balance.
00:49:50.839 --> 00:50:02.599
Yeah, there's um Flower Darby has written some great books on teaching online, and she talks about how uh she was working with a graphic designer to create some icons for her slide deck to talk about online teaching.
00:50:02.599 --> 00:50:06.519
And the first set of icons were all like some person at a keyboard.
00:50:06.519 --> 00:50:10.599
And she's like, No, learning is a community endeavor.
00:50:10.599 --> 00:50:12.839
You need multiple people in these pictures, right?
00:50:12.839 --> 00:50:14.039
Talking to each other.
00:50:14.039 --> 00:50:26.119
Um, and you know, that's that's there are, you know, if you're teaching fully asynchronous, there's a set of tools to try to create those shared experiences that are different than if you have some kind of synchronous online environment.
00:50:26.119 --> 00:50:28.439
Um so I'm mindful of time.
00:50:28.439 --> 00:50:33.000
I want to thank our panelists for sharing their experiences and their wisdom on this topic.
00:50:33.000 --> 00:50:35.639
I really appreciate diving into this with all three of you.
00:50:35.639 --> 00:50:37.399
Thank you for being here and for doing this.
00:50:37.399 --> 00:50:38.359
Thanks so much.
00:50:38.599 --> 00:50:39.639
Thanks for having us.
00:50:39.639 --> 00:50:40.759
Thank you.
00:50:42.279 --> 00:50:51.799
Thanks to our three take it relieve it panelists, Michelle Beavers, Leo Lo, and Sara McClellan, for sharing their perspectives and experiences with AI and online education.
00:50:51.799 --> 00:51:00.439
In the show notes for this episode, you'll find links to all the essays I cited during the panel, and links to more information about each of our three panelists.
00:51:00.439 --> 00:51:04.439
I would love to hear from you, dear listener, about today's panel.
00:51:04.439 --> 00:51:06.119
What perspectives would you take?
00:51:06.119 --> 00:51:07.559
Which ones would you leave?
00:51:07.559 --> 00:51:12.679
How are you navigating the role of AI in online education?
00:51:12.679 --> 00:51:15.959
You can click the link in the show notes to send me a text message.
00:51:15.959 --> 00:51:22.039
Be sure to include your name so I know who you are, or just email me at derek at derekbruff.org.
00:51:22.839 --> 00:51:28.359
Intentional Teaching is sponsored by UPCEA, the Online and Professional Education Association.
00:51:28.359 --> 00:51:36.919
In the show notes, you'll find a link to the UPCEA website where you can find out about their research, networking opportunities, and professional development offerings.
00:51:36.919 --> 00:51:42.599
This episode of Intentional Teaching was produced and edited by me, Derek Bruff.
00:51:42.599 --> 00:51:50.919
See the show notes for links to my website and socials, and to the Intentional Teaching newsletter, which goes out most weeks on Thursday or Friday.
00:51:50.919 --> 00:51:57.000
If you found this or any episode of Intentional Teaching useful, would you consider sharing it with a colleague?
00:51:57.000 --> 00:51:58.439
That would mean a lot.
00:51:58.439 --> 00:52:00.919
As always, thanks for listening.
00:00:34.839 --> 00:00:43.000
Welcome to Intentional Teaching, a podcast aimed at educators to help them develop foundational teaching skills and explore new ideas in teaching.
00:00:43.000 --> 00:00:45.079
I'm your host, Derek Bruff.
00:00:45.079 --> 00:00:51.000
I hope this podcast helps you be more intentional in how you teach and in how you develop as a teacher over time.
00:00:51.320 --> 00:00:54.119
I'm back with another Take It or Leave It panel.
00:00:54.119 --> 00:00:55.799
This one is a little bit different.
00:00:55.799 --> 00:01:00.759
Y'all have told me many times that you like the take it or leave it episodes of this podcast.
00:01:00.759 --> 00:01:15.719
So, the summer, when my University of Virginia colleagues and I were planning a series of professional development workshops for the fall aimed at faculty who teach online, I suggested that we use the take it or leave it format for one of the events.
00:01:15.719 --> 00:01:31.479
My colleagues liked the idea, and on October 2nd, in my role as Associate Director of the UVA Center for Teaching Excellence, I hosted a virtual panel titled Take It or Leave It, AI's Role in Online Learning, featuring three fantastic UVA colleagues.
00:01:31.479 --> 00:01:39.319
The conversation went very well, and the panelists and the CTE gave me permission to share the audio from the panel here on the Intentional Teaching Podcast.
00:01:39.480 --> 00:01:44.519
The panelists for this edition of Take It or Leave It are all at the University of Virginia.
00:01:44.519 --> 00:01:53.560
Michelle Beavers is an associate professor and coordinator of the Administration and Supervision Program in UVA's School of Education and Human Development.
00:01:53.560 --> 00:02:02.120
Leo Lo is Dean of Libraries and University Librarian, also advisor to the provost on AI literacy and a professor of education.
00:02:02.120 --> 00:02:07.799
Special thanks to Leo for agreeing to do this panel just two weeks after he started his job at UVA.
00:02:07.799 --> 00:02:20.280
Our third panelist is Sara McClellan, Assistant Professor of Professional Studies, and Program Coordinator of the Public Administration Certificate Program at UVA's School of Continuing and Professional Studies.
00:02:20.759 --> 00:02:29.079
For those of you new to the Take It or Leave It format, you'll hear me briefly describe five recent op-eds on teaching and learning in higher ed.
00:02:29.079 --> 00:02:37.960
For each op-ed, I'll ask each of our panelists if they take it, that is, generally agree with the main thesis of the essay or leave it.
00:02:37.960 --> 00:02:49.159
This is an artificial binary that I've found to generate rich discussion of the issues at hand, and that was definitely true for this panel, focused on hot takes about generative AI and online education.
00:02:49.159 --> 00:02:54.199
See the show notes for links to the essays we discussed if you want to read them in their entirety.
00:02:56.359 --> 00:03:04.359
Our first hot take, our first essay to discuss, is titled Are You Ready for the AI University?
00:03:04.359 --> 00:03:10.759
This was published in the Chronicle of Higher Education earlier this year.
00:03:10.759 --> 00:03:17.000
This was written by Scott Latham, who is a professor of strategy at the University of Massachusetts at Lowell.
00:03:17.000 --> 00:03:26.840
And in this piece, Professor Latham casts a fairly big vision for how generative AI will disrupt higher education in the coming years.
00:03:26.840 --> 00:03:33.319
There is a lot, there are a lot of big claims in this essay, but we're going to focus on just one of them.
00:03:33.319 --> 00:03:44.520
And for a bit of context, here's a quote from the essay: faculty will need to train AI in how to help them build lectures, assessments, and fine-tune their classroom materials.
00:03:44.520 --> 00:03:49.400
Further training will be needed when AI first delivers a course.
00:03:49.400 --> 00:03:58.520
While students will readily accept non-human professors, any new technology will have hiccups that require human intervention to resolve.
00:03:58.520 --> 00:04:02.680
And here's the thesis statement I'll ask our panelists to respond to.
00:04:02.680 --> 00:04:07.960
Once the training wheels are off, AI taught courses will become the dominant paradigm.
00:04:07.960 --> 00:04:10.599
That is a pretty bold statement.
00:04:10.599 --> 00:04:12.120
Sara, I'll start with you.
00:04:12.120 --> 00:04:13.719
Do you take that or leave that?
00:04:14.840 --> 00:04:20.680
I leave it in terms of the scope and speed of the argument that this author outlined.
00:04:20.680 --> 00:04:24.759
And I think for two big reasons.
00:04:24.759 --> 00:04:32.279
I think first, some of the predictions come up against energy shortfalls that aren't yet resolved.
00:04:32.279 --> 00:04:45.000
And so if we look, it's not just energy shortfall, it's transmission, it's chips, it's cybersecurity, it's all of the infrastructure and environmental features need needed to scope such a grand vision.
00:04:45.000 --> 00:04:56.519
Uh assuming that if such a vision were to come to pass, it would be an effort all over the country and be putting pressure on all of our systems, which would require addressing mega project problems, political gridlock.
00:04:56.519 --> 00:04:58.440
So that's one.
00:04:58.440 --> 00:05:13.000
And uh yeah, it also, I will say it also I think ignores the um the inconsistencies in human systems, or federalist structure, shared governance, uh, the way that higher education works across states.
00:05:13.000 --> 00:05:19.000
Um I think it's it's a pretty sweeping vision given our current patchwork system.
00:05:19.480 --> 00:05:20.440
Gotcha, gotcha.
00:05:20.440 --> 00:05:21.159
All right.
00:05:21.159 --> 00:05:23.799
Michelle, do you take this or leave this?
00:05:24.360 --> 00:05:25.720
I leave it, Derek.
00:05:26.759 --> 00:05:28.279
And why do you leave it?
00:05:28.600 --> 00:05:39.399
Well, I'm gonna lean towards the um relational aspect of teaching and say that AI does not have the disciplinary um expertise.
00:05:39.399 --> 00:05:45.320
It's that ethical judgment that faculty bring, our pedagogical pedagogical vision that we represent.
00:05:45.320 --> 00:05:51.720
Um and the most important piece is that interaction and relational trust and accountability with our students.
00:05:51.720 --> 00:06:00.840
I could see it as a back-end partner, uh, potential support for async or a flipped classroom environment.
00:06:00.840 --> 00:06:01.720
Okay.
00:06:02.679 --> 00:06:06.519
And you say this as someone with a fair degree of online teaching experience.
00:06:06.519 --> 00:06:11.639
And in fact, someone who uses AI a lot in the development of your learning materials.
00:06:11.960 --> 00:06:14.679
I certainly do, but it's not taking over.
00:06:15.000 --> 00:06:16.120
It's not taking over.
00:06:16.120 --> 00:06:16.600
Okay.
00:06:16.600 --> 00:06:17.559
Good.
00:06:17.559 --> 00:06:19.480
So we've got two leaves that leave it.
00:06:19.480 --> 00:06:22.600
Leo, are you gonna are you gonna complete the trio here?
00:06:22.600 --> 00:06:23.799
Do you take this or leave this?
00:06:24.039 --> 00:06:25.559
Well, I'm gonna go the other way then.
00:06:25.559 --> 00:06:26.840
I'll lean.
00:06:26.840 --> 00:06:28.200
Take it.
00:06:28.840 --> 00:06:29.080
Okay.
00:06:29.639 --> 00:06:33.399
Yes, I'll just make try to make it more uh uh interesting.
00:06:33.399 --> 00:06:42.919
And um Sara hit the the key point that I kind of the crux of this kind of well, why do I lean, you know, uh take it is that the timeline.
00:06:42.919 --> 00:06:47.879
I think right now, in the short term, it's I would I'll I'll leave it.
00:06:47.879 --> 00:06:57.159
But in the long term, I'm actually thinking that maybe a um so I try to compare AI to let's say the internet.
00:06:57.159 --> 00:07:00.360
I think right now we're in the dialogue phase.
00:07:00.360 --> 00:07:04.679
Can you imagine us doing our job now with dialogue technology?
00:07:04.679 --> 00:07:09.799
So it's hard to imagine what it's gonna be like in, let's say, 20 years' time, right?
00:07:09.799 --> 00:07:29.480
If it keeps improving in that way, including um um the you know how it solves some environmental issues and other things, if it keeps going that way, I can see AI being a really good teacher in that way, that it can improve in that way.
00:07:29.480 --> 00:07:32.039
It has unlimited patience.
00:07:32.039 --> 00:07:41.240
If it can improve its accuracy, then yeah, possibly, because I I honestly I'm using AI to learn things now.
00:07:41.240 --> 00:07:43.399
Obviously, low-stakes stuff.
00:07:43.399 --> 00:07:52.919
I'm not using it to teach me how to cure cancer, although it might happen in in the future, but it's hard to speculate 20 years from now.
00:07:52.919 --> 00:07:59.559
But at the same time, what Michelle just said about all the human relations, I think that's always going to be important.
00:07:59.559 --> 00:08:27.399
So the profess the human professor will become less of teaching the uh the technical part, perhaps, but teaching other things that, let's say, a college student might need, coming to college, the social aspect, the you know, the emotional intelligence, uh, even some human judgment part that AI may be may still have difficulty in in kind of understanding or even teaching.
00:08:27.399 --> 00:08:36.759
But in terms of strictly technical stuff, I think AI could could take on that that role in the future, in the longer term, not right now.
00:08:37.240 --> 00:08:37.720
Yeah.
00:08:37.720 --> 00:08:39.080
What's that line about?
00:08:39.080 --> 00:08:45.240
We tend to overestimate the short-term impact of new technologies and underestimate the long-term impact.
00:08:45.240 --> 00:09:02.279
Um, that said, I feel like I've been hearing this refrain for my entire career in higher ed, which is we're gonna have some magic technology that will provide wonderful personalized learning to all of our students and they will learn all the things that they need to know through this technology.
00:09:02.279 --> 00:09:06.680
Um, and it will it will mean we don't have huge class sizes anymore, right?
00:09:06.680 --> 00:09:07.960
It will change everything.
00:09:07.960 --> 00:09:12.040
And I feel like I've been waiting for that to actually happen for 20 years now.
00:09:12.040 --> 00:09:18.360
So um I think I'm a little skeptical that such a thing can exist myself.
00:09:18.360 --> 00:09:20.440
Um, what do you all think about that?
00:09:20.440 --> 00:09:26.200
How would you respond to Leo's claim that yeah, this is something that that we can imagine happening 20 years down the road?
00:09:26.200 --> 00:09:27.480
Michelle or Sarah?
00:09:27.800 --> 00:09:39.720
I think Leo brings up a good point that some of the um practical knowledge-based pieces could provide that one-on-one instruction for students and to support their learning where they are.
00:09:39.720 --> 00:09:53.480
But the piece that it doesn't um recognize is where we negotiate meaning and understanding with our students, where we challenge them, their values, the way they're confronting issues, um, all that still remains human.
00:09:53.480 --> 00:09:58.840
So Leo may have a point that could sway me closer to possibly take it.
00:09:58.840 --> 00:10:05.160
When I think about could we partner and have different roles within the classroom environment?
00:10:05.480 --> 00:10:09.320
AI as an instructional assistant, but not as the sole instructor.
00:10:09.560 --> 00:10:11.560
Still not letting them take my place.
00:10:12.200 --> 00:10:13.000
Right, right.
00:10:13.000 --> 00:10:16.680
I absolutely agree with Michelle just what Michelle just said.
00:10:16.680 --> 00:10:24.360
The the roles, the uh the roles may change, but I think humans always will prefer to have human interaction.
00:10:24.360 --> 00:10:30.920
So yeah, so that's just kind of our our job in kind of defining what that is going to be like moving forward.
00:10:30.920 --> 00:10:31.720
Yeah.
00:10:32.759 --> 00:10:42.200
And I think when we've seen efforts along these lines in the past to try to turn the education over to the robots, whether it was AI or the MOOC craze of the 2010s.
00:10:42.200 --> 00:10:44.279
It does work for some students.
00:10:44.279 --> 00:10:50.759
Like there are students who have sufficient background knowledge and sufficient motivation that they'll kind of take these tools and really learn well.
00:10:50.759 --> 00:11:01.800
But I find that it doesn't work for 85% of the students who need the human relationship, they need the structure, they need the motivation to do that kind of learning.
00:11:01.800 --> 00:11:03.240
Um, I am skeptical.
00:11:03.240 --> 00:11:06.519
Uh uh Latham writes in the piece, I said this earlier.
00:11:06.519 --> 00:11:12.040
While students will readily accept non-human professors, I don't know about that.
00:11:12.040 --> 00:11:14.360
Again, some students will, I'm sure.
00:11:14.360 --> 00:11:15.800
They'll be happy.
00:11:15.800 --> 00:11:17.960
Um, but that that seems like a stretch.
00:11:17.960 --> 00:11:23.399
I think most of our students are not are not wanting to have the AI tutor as their sole instructor.
00:11:24.200 --> 00:11:35.720
That makes me think about the gap we're seeing when Pew and other organizations have studied public perspectives on AI and the perspectives of experts versus the average member of the public.
00:11:35.720 --> 00:11:42.600
There's a pretty big gap, and we're seeing a larger gap than we've seen with the introduction of a lot of other technologies.
00:11:42.600 --> 00:11:46.440
Um so perhaps we'll tip over that gap at some point in certain areas.
00:11:46.440 --> 00:11:58.360
But right now, I certainly hear that I teach mid-career professionals, and I that gap is alive and well as I hear technology experts in my classes interacting with folks who are pretty wary.
00:11:58.360 --> 00:12:02.360
Um, and then looking at that data and talking it through.
00:12:03.000 --> 00:12:04.040
Yeah, yeah.
00:12:04.360 --> 00:12:04.920
All right.
00:12:04.920 --> 00:12:10.440
Okay, so we have two take it, uh, no, two leave it's and a take it for our first one.
00:12:10.440 --> 00:12:12.440
Um, let's move on to our second one.
00:12:12.440 --> 00:12:17.399
I've tried to pick some really interesting essays for us to respond to that kind of go in very different directions.
00:12:17.399 --> 00:12:25.560
Um uh let's see, essay number two is titled AI Risks Undermining the Heart of Higher Education.
00:12:25.560 --> 00:12:34.200
And this is a piece by Zahid Naz, who is a senior lecturer in academic and professional education at Queen Mary University of London.
00:12:34.200 --> 00:12:41.960
Um, and in this piece, Naz expresses a very common concern that I hear from faculty about AI's role in learning.
00:12:41.960 --> 00:12:47.320
A worry that students using AI will avoid the hard work of learning.
00:12:47.320 --> 00:12:47.879
Right?
00:12:47.879 --> 00:12:51.560
The first one was a little bit about AI doing the instruction, the teaching.
00:12:51.560 --> 00:12:54.279
This one is about AI doing the learning.
00:12:54.279 --> 00:13:00.200
Uh Naz writes, uh, the process of learning involves much more than reading texts.
00:13:00.200 --> 00:13:09.480
It requires students to grapple with intricate concepts, compare and contrast ideas, and navigate the nuanced arguments presented in academic literature.
00:13:09.480 --> 00:13:19.320
This sharpens critical thinking, cultivates original thought, and builds the foundation for intellectual independence and what I have identified as the thesis.
00:13:19.320 --> 00:13:28.200
In short, the unchecked use of AI could ultimately undermine the very intellectual rigor that makes higher education meaningful.
00:13:28.200 --> 00:13:30.680
I'll start with you this time, Leo.
00:13:30.680 --> 00:13:32.120
Do you take that or leave that?
00:13:33.399 --> 00:13:34.519
I'll take that.
00:13:34.519 --> 00:13:35.399
I agree with that.
00:13:35.399 --> 00:13:40.519
Um so I and you said something earlier, Derek, about motivation of students.
00:13:40.519 --> 00:13:56.920
And I think a lot of the times, especially with our the current kind of system for them for the masses, is that it's very uh assessment grade, kind of you know, uh uh playing a huge part in why they do certain things.
00:13:56.920 --> 00:14:10.759
Uh of course there are people who are just they just want to learn, they want to improve, they want to grow, but the pressure is really on our students to produce something that will give them a good grade so they can get a good job and and and all of that.
00:14:10.759 --> 00:14:17.160
And that makes it a lot tempting, more tempting to take shortcuts.
00:14:17.160 --> 00:14:22.519
Uh because they just the grades become more important than the learning part in many cases.
00:14:22.519 --> 00:14:44.360
So I think uh unless we can tap into um really in this you know incentivizing the students' motivation part, that we have a good grasp of that, having some a tool that is so easily that can be used so easily to get you know, to get a passing grade or a good grade, I can see people using it.
00:14:44.360 --> 00:14:56.759
So that means we need to have some mechanism to help people wanting to learn, or maybe change the assessment system a little bit so that there's no so high stake for them.
00:14:56.759 --> 00:14:58.840
Um so I I would take that.
00:14:58.840 --> 00:15:06.840
I I agree that you know we that there needs to be some uh checked use uh of AI right now and in the short term.
00:15:07.720 --> 00:15:08.600
Yeah, yeah.
00:15:08.840 --> 00:15:09.879
How about you, Michelle?
00:15:10.759 --> 00:15:12.360
I will say I agree with Leo.
00:15:12.360 --> 00:15:36.440
I'm gonna take the part about the warning about um the unchecked deployment of AI, but I'm gonna leave the idea that AI can degrade learning because I feel like um as faculty, as professors, we have this responsibility to create these guardrails and these scaffolds so that AI can actually augment our student learning rather than replace it.
00:15:36.440 --> 00:15:48.759
Um and if we design that learning that creates productive struggle for our students, we're actually enhancing that thinking and that learning rather than allowing AI to do that.
00:15:48.759 --> 00:15:51.080
But it has to be an intentional design.
00:15:53.399 --> 00:15:58.519
So before I go to Sarah, what what is what what does that look like in your courses, Michelle?
00:15:58.519 --> 00:16:03.639
What are you what are some design elements you use to try to move students to that productive struggle?
00:16:04.200 --> 00:16:14.519
So um I will say that first of all, that um I use AI as a cognitive partner and it's a teammate within my um classroom.
00:16:14.519 --> 00:16:23.000
So students work collectively with their human friends, and then they're allowed to invite AI into the conversation as that third critical partner.
00:16:23.000 --> 00:16:43.399
Um I'm also dabbling with the AI agent in which students can interact and receive feedback on the work that they're doing, whether it's a process or a structure they don't understand, training that agent to not give the answer but to ask questions that leads the student to find their own answer.
00:16:43.399 --> 00:17:06.440
Um I think when we create that ambiguity, that um the the requirement to compare a, for example, in an essay assignment, compare an AI outcome to your own outcome and give the rationalization of why you agree or you disagree helps to enhance that thinking process as opposed to saying, oh, you can use AI for this assignment.
00:17:06.440 --> 00:17:12.359
Scaffold it, structure it, and design it, helping students understand the why behind the practice.
00:17:12.359 --> 00:17:16.279
So looking at the process, not just that outcome.
00:17:16.759 --> 00:17:17.079
Right.
00:17:17.079 --> 00:17:26.119
And so if the unchecked use is problematic, your checks involve the structures and sequences that you're asking students to work through, right?
00:17:26.119 --> 00:17:35.160
Um, as well as what I heard is some checks on the AI itself to kind of point it in useful directions and not uh less helpful directions.
00:17:35.559 --> 00:17:35.880
Correct.
00:17:35.880 --> 00:17:49.720
And the only other thing I'll add because Leo said it, he said emotional intelligence, and I gave a workshop this morning on emotional intelligence, and we used AI to demonstrate our emotional intelligence by having this interactive conversation.
00:17:49.720 --> 00:17:55.079
So AI is capable of not only the technical skills but the dispositional skills as well.
00:17:56.360 --> 00:17:56.759
All right.
00:17:56.759 --> 00:17:58.600
I'm gonna go to you now, Sara.
00:17:58.600 --> 00:18:01.240
Take it or leave it.
00:18:01.799 --> 00:18:13.240
I I take it, and one thing I'm I'm really aware of right now is that I teach relatively small classes, 15, 20, 25 students.
00:18:13.240 --> 00:18:17.160
Um, I have a great um teaching excellence center.
00:18:17.160 --> 00:18:19.480
I have access to lots of resources.
00:18:19.480 --> 00:18:26.839
I'm communicating with friends and colleagues around the country who work in really big systems that are pushed to the absolute max.
00:18:26.839 --> 00:18:40.039
They're sometimes teaching dozens or hundreds of students in a single class with limited TA support uh and don't necessarily have strong instructional support at the university for transitions.
00:18:40.039 --> 00:18:46.200
And so uh I think I struggle at the individual level and where resources are there.
00:18:46.200 --> 00:18:59.000
I think that there will be ways, as Michelle said, to make this um a really strong learning tool and and create spaces strategically where students can lean into that tool and and reflect on it.
00:18:59.000 --> 00:19:06.200
Um, but I try to think about I think I suspect that will be pretty spotty and there'll be systems where that's much, much harder.
00:19:09.880 --> 00:19:11.079
I'm curious, Sara.
00:19:11.079 --> 00:19:15.960
So so Leo pointed to the kind of the role of motivation in all this, right?
00:19:15.960 --> 00:19:21.799
And if students are just box checking, then yeah, it's easy to use use some unchecked AI.
00:19:21.799 --> 00:19:25.000
Um you mentioned teaching mid-career professionals.
00:19:25.000 --> 00:19:28.920
Do you find that they come in wanting a certain degree of rigor?
00:19:28.920 --> 00:19:34.039
And that makes it easier then to kind of avoid the the AI shortcuts.
00:19:35.240 --> 00:19:37.240
You know, I would I would say it's a mixed bag.
00:19:37.240 --> 00:19:53.160
I would say the majority of students uh want a reasonable degree of rigor, but I definitely get those students who are in the program because they've been pushed to be in the program around a promotional opportunity, or somebody's told them that they really need to check that box uh professionally.
00:19:53.160 --> 00:20:04.920
And um, those do tend to be the students that I sometimes have conversations with about AI use and and um to the extent to which they're leaning on AI, but they're also really busy professionals.
00:20:04.920 --> 00:20:23.240
So I think the temptation, I have a lot of students who are emergency services worker, and so the temptation is I've just worked, you know, a crazy shift and I'm exhausted and I'm trying to do 10 different things, and now there's another emergency, and so AI looks really good at that point, and uh that's that's a hard one.
00:20:25.079 --> 00:20:25.480
Thank you.
00:20:25.480 --> 00:20:31.720
I think sometimes I have this vision of like the the first year undergraduate taking a writing course they don't want to have to take.
00:20:31.720 --> 00:20:35.640
And that's that's a hard, hard course to teach when AI is out there.
00:20:35.640 --> 00:20:41.160
And sometimes I feel like the professional studies programs must just it's just all you know, sunshine and roses there.
00:20:41.160 --> 00:20:42.839
Like they all want to be there and learn.
00:20:42.839 --> 00:20:44.759
Um, but we're all humans, right?
00:20:44.759 --> 00:20:48.039
We don't we don't show up with our best selves every day in class.
00:20:48.039 --> 00:20:48.839
Yeah.
00:20:48.839 --> 00:20:49.960
Yeah.
00:20:49.960 --> 00:20:53.240
Um Sarah, one one other follow-up.
00:20:53.240 --> 00:20:59.240
We talked about some of the the checks that Michelle has in terms of the structures and the tools that she's using.
00:20:59.240 --> 00:21:07.400
What are what are some of the checks that you have in your courses to try to kind of orient students towards a more rigorous application of learning?
00:21:08.120 --> 00:21:09.400
Well, I'm experimenting.
00:21:09.400 --> 00:21:16.600
This term is the first term that I did a brief, like three to four minute video right at the start of my public policy course.
00:21:16.600 --> 00:21:18.519
And I said, let's talk about AI.
00:21:18.519 --> 00:21:23.079
Let's talk about the get to badly and how we can use it in policy analysis.
00:21:23.079 --> 00:21:28.839
And so I sort of stepped right out there and um students really responded to that.
00:21:28.839 --> 00:21:33.880
And I created an AI agent, a tool that students could use around policy analysis.
00:21:33.880 --> 00:21:40.360
And I fed that tool and I explained to them how I did it with readings and concepts and frameworks from the course.
00:21:40.360 --> 00:21:50.440
And then I asked them to please not use that tool until a very specific point in the course, because I didn't want the tool to take the learning away from them.
00:21:50.440 --> 00:21:56.120
Uh and I said, you know, policy practice is a way of thinking about and approaching problems.
00:21:56.120 --> 00:22:01.640
And if we let AI learn how to approach this problem, you know, we're paying for a lot and we're showing up and we're not learning.
00:22:01.640 --> 00:22:09.720
And so I said, but it's a fantastic partner to sort of think through and challenge the assumptions that you have and the research research that you're using.
00:22:09.720 --> 00:22:14.519
And so I tried to give them strategic places where I said, please, please test the tool here.
00:22:14.519 --> 00:22:16.120
Tell us how it's working.
00:22:16.120 --> 00:22:24.120
Um, and so far I think that's I think it's going um much better than it did in the last term, where it was a little bit more hands-off about it.
00:22:24.920 --> 00:22:25.480
Yeah.
00:22:25.480 --> 00:22:27.880
Because you're saying we're gonna use AI.
00:22:27.880 --> 00:22:47.000
It's not like we're gonna pretend it doesn't exist, but we're gonna be thoughtful and intentional about how we use AI, not to replace our own work, but maybe as as Michelle was saying, as a conversation partner, as a as a uh um uh a tool that can kind of provoke us with interesting ideas, perhaps.
00:22:47.319 --> 00:22:50.519
Yes, and I linked to articles about AI and motivation.
00:22:50.519 --> 00:22:54.839
Um, and several of them responded it was optional reading, but they clearly did it.
00:22:55.240 --> 00:22:56.519
Yeah, yeah.
00:22:56.519 --> 00:23:02.680
Um Leo, anything to add after hearing Michelle and Sara talk about their approaches?
00:23:03.079 --> 00:23:05.880
No, I absolutely agree with with those approaches.
00:23:05.880 --> 00:23:19.079
I think um ultimately, but ultimately it is about setting up, like you said, some guardrails, some kind of guidance on how to actually use the tool to to learn and um benefit them.
00:23:19.079 --> 00:23:23.319
Um and you know, some people will never do that.
00:23:23.319 --> 00:23:27.720
They was, you know, they will find ways to, you know, uh, like you said, Derek, we're all humans.
00:23:27.720 --> 00:23:29.720
Sometimes we don't show up with our best selves.
00:23:29.720 --> 00:23:39.559
But at the same time, that I also believe there are many people who really wanting to learn just you know, in this kind of uh environment, don't know how or what's the best way.
00:23:39.559 --> 00:23:49.319
So that I love that, you know, Sarah and Michelle are really giving them uh ways to do that, to optimize their experience using the technology.
00:23:49.720 --> 00:24:03.240
And I think one thing I've observed is as these new technologies come on, it takes higher ed a little time to figure out what are some of better practices, what are what are they good for, what are they bad for, what are the structures that we could put into place?
00:24:03.240 --> 00:24:07.640
Um you know, there's some collective figuring out that happens.
00:24:07.640 --> 00:24:10.440
Well, and that's a great segue into our next essay.
00:24:10.440 --> 00:24:14.600
And Leo, I'm coming for you on this one first, because this is this is your wheelhouse.
00:24:14.600 --> 00:24:17.400
Um, this one is about AI literacies.
00:24:17.400 --> 00:24:23.000
Uh, this is an essay by Ray Schroeder called The Urgent Need for AI Literacy.
00:24:23.000 --> 00:24:28.200
Uh, Ray Schroeder is a senior fellow at UPCEA, the online and professional education association.
00:24:28.200 --> 00:24:39.160
And in this piece, he cites a number of workforce studies pointing to the growing role of AI and the need to prepare students to have AI literacies.
00:24:39.160 --> 00:24:57.400
Um he says uh the rapid advent of AI capabilities coupled with the developing economic pressures worldwide have led to a surge in employers seeking to reduce operating expenses through widespread use of generative and agentic AI to augment and in some cases replace humans in their workforce.
00:24:57.400 --> 00:25:00.759
And then what I've identified, there's kind of an implied thesis here.
00:25:00.759 --> 00:25:03.400
So I'm going to read the statement and then frame it up a little bit.
00:25:03.400 --> 00:25:14.759
He says we are failing to fully prepare those students to enter the workforce, where, as the World Economic Forum says, two-thirds of business leaders surveyed said they wouldn't hire a candidate without AI skills.
00:25:14.759 --> 00:25:24.200
And nearly three-quarters said they would rather hire a less experienced candidate with AI skills than a more experienced candidate without them.
00:25:24.200 --> 00:25:29.400
And so I'm going to focus on the first part of that thesis statement, which I'll drop in the chat.
00:25:29.400 --> 00:25:34.680
We are failing to prepare, failing to fully prepare those students to enter the workforce.
00:25:34.680 --> 00:25:38.120
Um Leo, would you take that thesis or leave it?
00:25:39.240 --> 00:25:43.400
Let me share a recent experience first, and then I'll answer answer you.
00:25:43.400 --> 00:25:50.039
Uh this week I I was at a gathering with uh a small group of, I would say AI experts.
00:25:50.039 --> 00:25:51.880
They are very close to the action.
00:25:51.880 --> 00:26:05.000
And um it was quite depressing, actually, um, because they thought, really believed, that in about 18 months' time, uh pretty much all the entry-level jobs will be gone.
00:26:05.000 --> 00:26:09.720
And you can see it kind of happening now, but they are very close to the action.
00:26:09.720 --> 00:26:17.400
And I quote one of them saying, you know, when they saw the um the new newest models, I quote, almost like magic.
00:26:17.400 --> 00:26:21.160
So um you can kind of okay, I put some weight on what they're saying.
00:26:21.160 --> 00:26:22.440
Let's just say, assume that.
00:26:22.440 --> 00:26:34.120
So, in that, if that is true, then I'll take it that we are not really preparing our students because what students come to a university like UBA so that they can get good jobs.
00:26:34.120 --> 00:26:47.160
We want them to you know prepare them for the workforce so that they learn the skills and also all the other things that can help them succeed in getting that entry-level job and then succeeding you know in the uh for the rest of their lives.
00:26:47.160 --> 00:26:50.600
But if those jobs are gone, then what are we doing?
00:26:50.600 --> 00:26:53.000
Why are they paying the money to come here?
00:26:53.000 --> 00:27:01.960
Yes, there are other benefits of going to college, but at the same time, you pick out that much money to pay for something, you want some tangible return.
00:27:01.960 --> 00:27:05.720
So here's some of my, in a way, my thinking.
00:27:05.720 --> 00:27:09.319
I don't know, I would love to get everybody's take on this, right?
00:27:09.319 --> 00:27:16.440
So for the past, I don't know, 100, 200 years, we're preparing students to get jobs because of industrial revolutions.
00:27:16.440 --> 00:27:18.680
You have jobs and people want to get those.
00:27:18.680 --> 00:27:23.160
But before all of that, everybody was kind of their own entrepreneurs.
00:27:23.160 --> 00:27:26.920
You're a blacksmith, you're a farmer, you're all these, you work for yourself.
00:27:26.920 --> 00:27:29.559
Would we go back to that?
00:27:30.360 --> 00:27:30.759
Right?
00:27:31.079 --> 00:27:36.519
If all the jobs, entry-level jobs are gone, AI can do all of that.
00:27:36.519 --> 00:27:46.120
But at the same time, everybody, every one of us, every one of the students can also have a team of AI agents to be their own bosses.
00:27:46.120 --> 00:27:53.079
As long as there are problems in the world, there is always a demand for solutions, however big or small.
00:27:53.079 --> 00:28:08.839
If there are demand for solutions, there are always opportunities for these, I would call them micro solopreneurs, that they can use a team of AI agents uh to help them solve real-world problems, however big or small.
00:28:08.839 --> 00:28:24.600
Shouldn't we be teaching them how to do that and using the technology to help them, you know, build up that kind of scope of the problem, build using infrastructure support here to learn how to do that, and then go out there and be their own bosses in many ways.
00:28:24.600 --> 00:28:40.600
I understand that not 100% of the students would want to do that, but even if there are 10%, and if there's such a radical change in the society, maybe, you know, that we need to have some kind of radical changes as well to adapt to that.
00:28:40.600 --> 00:28:43.079
I am just brainstorming right here.
00:28:43.079 --> 00:28:53.000
So I that much, but it's just an idea that I've been really kind of simmering in terms of how teachers in in AI literacy.
00:28:53.000 --> 00:28:58.200
Of course, they need to be AI literate to be able to do any of those things I just talked about.
00:28:58.200 --> 00:29:00.680
Or even just to get jobs if they ask you a bit.
00:29:01.880 --> 00:29:02.360
Yeah.
00:29:02.360 --> 00:29:07.000
Well, and I've heard a lot of people say the entry-level jobs in this field or that field are disappearing.
00:29:07.000 --> 00:29:10.279
I haven't heard a lot of what then do we do?
00:29:10.279 --> 00:29:13.799
Um, so I appreciate your brainstorming with us.
00:29:13.799 --> 00:29:16.440
Uh Sara, how would you respond to this thesis?
00:29:16.440 --> 00:29:20.360
Are we failing to fully prepare our students to enter the workforce?
00:29:22.440 --> 00:29:24.039
Yes, I'm afraid so.
00:29:24.039 --> 00:29:24.920
I would take that.
00:29:24.920 --> 00:29:33.079
And I think um, Leo, you hit on uh a lot of themes that I've been hearing about and and reading about as well.
00:29:33.079 --> 00:29:40.120
Um I think one thing that really strikes me is that AI literacy, it's not just a technology.
00:29:40.120 --> 00:29:50.839
It is because, as Leo suggested, I think it is so fundamentally going to change our social fabric, our economy, what it means to be successful professionally.
00:29:50.839 --> 00:30:08.039
Um, that I think part of AI literacy for me is um change, navigating change and helping students figure out how to kind of surf with disruptions and and not seize up because we're gonna have a lot of change.
00:30:08.039 --> 00:30:10.360
Um it's already here, we'll have more.
00:30:10.360 --> 00:30:17.640
And so I think um, and that's intellectually and emotionally, it can be fun and it can also be exhausting.
00:30:17.640 --> 00:30:23.240
And I think recognizing where kind of the human condition is, we've come out of a global pandemic.
00:30:23.240 --> 00:30:29.480
Um, you know, we're in politically tumultuous times, and so, and we have institutions.
00:30:29.480 --> 00:30:48.839
That were, you know, industrial or institutions that ironically they helped sort of create or unleash, depending on how you look at it, all of these technologies, but they're now socially accelerated to a point that our institutions aren't built to manage, to develop public policy, to do quick responses.
00:30:48.839 --> 00:30:52.279
And that creates a lot of tensions.
00:30:52.279 --> 00:30:58.519
And so I think that a big piece of this is figuring out when to go fast and when to go slow.
00:30:58.519 --> 00:31:08.920
There's a huge temporal element to all of this because reason, judgment, our democratic process, our shared governance system, it all takes time.
00:31:08.920 --> 00:31:26.759
And so I think when we think about AI literacy, not just for students, for faculty, for administrators, it's got to be about human strategy and change strategy and trying to sort of assess what's coming and get much, much more creative than we've been.
00:31:26.759 --> 00:31:33.319
And so I feel like Leo, we could have a whole session on just what are the creative possibilities when those entry-level jobs go away.
00:31:33.319 --> 00:31:38.759
And I think we need a big, big temp approach to trying to solve those problems.
00:31:39.079 --> 00:31:39.559
Michelle.
00:31:40.279 --> 00:31:42.680
Struggling with how to respond to this, Derek.
00:31:42.680 --> 00:31:49.000
For everyone joining us today, I'm an educator for an education society.
00:31:49.000 --> 00:31:56.360
So I'm looking at preparing the folks that work in educational institutions, or mean meaning K-12 education.
00:31:56.360 --> 00:32:07.960
And I recently read an article that in 2023, only 50% of school divisions have either said, yes, I'm going to use AI or no or not, but haven't developed policy yet.
00:32:07.960 --> 00:32:13.319
So, in quite honest, in full honesty, education is behind the eight-ball.
00:32:13.319 --> 00:32:15.319
We're moving slow already.
00:32:15.319 --> 00:32:18.600
We're not keeping up with what the business world is doing.
00:32:18.600 --> 00:32:24.440
And what I'm seeing here is that school divisions are looking for the tools.
00:32:24.440 --> 00:32:26.680
They're not looking at the skills.
00:32:26.680 --> 00:32:33.000
And so therefore, we're not the they want to know how to make their life easier.
00:32:33.000 --> 00:32:35.880
What tools should we adopt or not adopt?
00:32:35.880 --> 00:32:41.880
And they haven't gotten to the point of recognizing that it's not the tools that we need to teach people how to use.
00:32:41.880 --> 00:32:47.960
It's the AI skill, the AI literacy, the judgment, the critical thinking, the metacognition, et cetera.
00:32:47.960 --> 00:32:53.480
So I'm kind of like sitting on a double-edged sword here, not really sure which way to go.
00:32:53.480 --> 00:33:01.400
And I think the other piece that we really have to consider is you're looking at a panel of three individuals.
00:33:01.400 --> 00:33:03.160
Well, what if you looked at the system?
00:33:03.160 --> 00:33:07.480
How many people are actually integrating teaching AI skills?
00:33:07.480 --> 00:33:17.079
And if we had to say that, I'd say I'd take it because if I look at my own department or my own school of education, not everybody is at that place yet.
00:33:17.079 --> 00:33:19.079
And so we need to be.
00:33:19.180 --> 00:33:19.180
I would add, to your last point, that most of the AI policy discussions I've heard in higher ed, and the AI strategy responses are at the individual course level and the individual instructor, how can they make better choices for their courses? What policy do you set for your course? But our students don't experience courses as islands. They enroll in programs that are hopefully developing a whole skillset in our students over time. If our programs don't have coherent responses to these challenges, smart choices by faculty can only go so far if students are experiencing very different conditions in the courses in their program.
00:33:19.279 --> 00:33:19.279
Yeah that ambiguity across programs is a problem. And it gets even deeper when you start talking about dissertations and capstones and what's permitted and what's not permitted.
00:33:19.380 --> 00:33:19.380
So I think you all said take it, that maybe we need to do more and move faster. I'm going to move us on to our fourth essay which I think takes a different approach. This is a piece published I think last week in the Chronicle of Higher Education.
00:34:41.480 --> 00:34:44.519
Yes, that was last week, I think, by Kevin Gannon.
00:34:44.519 --> 00:34:48.519
It's called Sometimes We Resist AI for Good Reasons.
00:34:48.519 --> 00:34:55.079
Uh Kevin Gannon is a professor of history and director of the teaching center at Queen's University of Charlotte.
00:34:55.079 --> 00:34:57.719
Um I've met Kevin, he's a teaching center guy.
00:34:57.719 --> 00:35:02.359
Um his his online handle is the Tattooed Professor, and it is very accurate.
00:35:02.359 --> 00:35:04.359
He is covered in tattoos.
00:35:04.359 --> 00:35:19.559
Um Kevin uh takes his essay, most of the essay details four big questions related to AI policy in higher education, where he argues the voices of AI skeptics and even resistors need to be involved.
00:35:19.559 --> 00:35:27.639
He writes on top of the dizzying rate of technological change is a relentless boosterism surrounding generative AI.
00:35:27.639 --> 00:35:36.039
As faculty members, we are told that there is no alternative, that these tools are the future of work, that we must either adapt or become obsolete.
00:35:36.039 --> 00:35:43.000
That framing is responsible, however, for the exclusion of important perspectives from our institutional conversations on AI.
00:35:43.000 --> 00:35:57.959
And so the thesis statement I'll ask you to respond to is rather this is a call to make sure that everyone is at the table when you are setting these policies, even and especially the harshest campus critics of AI.
00:35:57.959 --> 00:36:00.359
Michelle, I'll start with you.
00:36:00.919 --> 00:36:03.959
Um I will take it and I'll take it strongly.
00:36:03.959 --> 00:36:07.639
Um I feel like this is a shift for me.
00:36:07.639 --> 00:36:19.319
I started using AI um in the summer of 22, excuse me, 23, and um I was an early adopter and very enthusiastic about it.
00:36:19.319 --> 00:36:48.279
But it wasn't until I sat amongst the resistors that I recognized the importance of slowing down, the importance of hearing dissenting views, um, especially those who are skeptical and critical of AI, because it helps to build a better policy and practice for the whole organization, but for myself too, who would be ready green light, jump in, let's go, when in reality, it might be time to pause and consider how to do it more effectively.
00:36:48.679 --> 00:36:49.719
Sara, how about you?
00:36:50.039 --> 00:36:53.239
I would take it and I partly I own this.
00:36:53.239 --> 00:36:55.719
Um maybe I'm a closet contrarian.
00:36:55.719 --> 00:37:00.119
Uh, I have I'm I'm ambivalent about artificial intelligence.
00:37:00.119 --> 00:37:04.039
I've jumped into it, I'm working with it, I'm learning with my students.
00:37:04.039 --> 00:37:17.639
Um and I, as I look at other technologies like social media and everything we've learned over the course of the last 15 plus years, um, I if I could push that button and shut it all down, I might do that.
00:37:17.639 --> 00:37:20.839
Uh so, you know, so don't give me that power.
00:37:20.839 --> 00:37:34.839
Um, but one of the things that I think about a lot is I think it's I felt like it was important as a contrarian that I stepped into the process and that I be tried to become AI literate.
00:37:34.839 --> 00:37:37.480
Uh, and and thus my ambivalence.
00:37:37.480 --> 00:37:39.879
I would, I guess I'm not a solid contrarian anymore.
00:37:39.879 --> 00:37:41.079
I'm a bit ambivalent.
00:37:41.079 --> 00:38:02.199
And um I so I think the friendly amendment for me with the take it is that I would expect if we engage contrarians, that we ask or expect them to step up and work on some AI literacy in the process so that people aren't just uh hardstop no uh without understanding much about the technology.
00:38:05.879 --> 00:38:06.199
Right.
00:38:06.199 --> 00:38:15.079
I have certainly encountered the contrarians and the resistors who read a couple of headlines about AI three years ago and made up their mind then that it was useless.
00:38:15.079 --> 00:38:17.639
And I feel like that's that's not an informed position.
00:38:17.719 --> 00:38:24.119
Um it's hard to critique something that you don't understand very well and you haven't experimented with.
00:38:24.119 --> 00:38:25.079
Yeah.
00:38:25.879 --> 00:38:26.839
Leo, how about you?
00:38:26.839 --> 00:38:28.119
Do you take this or leave it?
00:38:28.519 --> 00:38:29.719
Oh, I'll take it.
00:38:29.719 --> 00:38:40.839
Um, I even though I can see enormous potential with the technology, we I think we can all agree that it has some inherent flaws, limitations, and a lot of problems with it.
00:38:40.839 --> 00:38:46.919
And we're in this rapidly growing stage, but it hasn't grown yet in certain ways.
00:38:46.919 --> 00:38:49.799
So this is the time to make changes, right?
00:38:49.799 --> 00:38:56.039
To make sure that you know we actually have that window um to set some rules.
00:38:56.039 --> 00:38:58.759
Once the rules are set, they're difficult to change.
00:38:58.759 --> 00:39:00.519
But this is the time to set those rules.
00:39:00.519 --> 00:39:13.079
And so I encourage people who have uh uh who are skeptical of this technology or have you know want to be critical about it to really raise their, you know, to have their voice heard, basically.
00:39:13.079 --> 00:39:17.399
But at the same time, I agree with Sarah, basically people need to be AI literate first.
00:39:17.399 --> 00:39:27.879
Because and I always I say that to everybody, is whether you're for the technology, technology or if you're against it, your arguments are going to be stronger if you know more about it.
00:39:27.879 --> 00:39:49.559
And that's what I'm really advocating for, is for for all of us to learn more, to be more AI literate, to learn more about it, and from the librarian's perspective, where to find credible information so that it can substantiate your arguments and all of that, so that your voice becomes stronger to make the technology better, to make the changes better.
00:39:49.559 --> 00:39:55.639
Because right now we're in a period of chaos, in a period of rule setting, and there aren't that many rules.
00:39:55.639 --> 00:40:02.359
Even if you look at countries and regions with regulations, it's chaotic chaotic right now.
00:40:02.359 --> 00:40:05.719
So is this this is time to help shape that.
00:40:05.719 --> 00:40:07.879
So I take it.
00:40:09.239 --> 00:40:16.279
I think um there's a question in the chat from Tom about how this relates to um online teaching.
00:40:16.279 --> 00:40:36.119
And I think that um by bringing these naysayers, these dissenters into the conversation, especially if they are AI, if they are faculty who teach online, it's critically important going back to that first article about will AI be the instructor for our courses online in the future.
00:40:36.119 --> 00:40:47.079
And like Leah said, faculty have to be AI literate in order to partner with AI in online environments or face-to-face environments.
00:40:48.199 --> 00:40:52.919
Yeah, I uh along those lines are in relationship to teaching for me.
00:40:52.919 --> 00:41:13.079
Um, I'm co-teaching a public sector AI leadership course, and one of the things we're doing is working to integrate um AI critics into our curriculum, including from the tech industry, experts who are leaders in AI who are now or or have been raising a flag and highlighting significant risks.
00:41:13.079 --> 00:41:26.119
And so I want us to be able to talk about those in classes that involve government leaders making big decisions about AI, because otherwise tech vendors have enormous power and we don't understand the big arguments.
00:41:28.839 --> 00:41:38.759
I'm gonna take us to our last essay, and I want to use Tom's question as a jumping-off point because I do want to talk concretely about what online teaching and learning looks like in this AI era.
00:41:38.759 --> 00:41:45.639
Um this essay is called On AI We Reap What We Sow.
00:41:45.639 --> 00:41:52.039
This is a piece by Chad Hansen, who teaches sociology and religion at Casper College in Wyoming.
00:41:52.039 --> 00:42:01.959
He makes the case that higher ed has focused far too much on outcomes and products of learning and not on the value of the process of learning itself.
00:42:01.959 --> 00:42:05.239
Um He says, Enter AI.
00:42:05.239 --> 00:42:11.959
Today we live in an era in which students can feed a prompt into an automated prose generator and in seconds have a viable draft of a writing assignment.
00:42:11.959 --> 00:42:13.959
What are they supposed to think?
00:42:13.959 --> 00:42:19.159
We've spent three decades acting like outcomes assessments are the only things we value.
00:42:19.159 --> 00:42:22.919
And here's the thesis statement I'll ask you to respond to.
00:42:22.919 --> 00:42:38.279
We should see the AI era as providing us with a reason and an opportunity to expand our interest to include an analysis of the broadly formative processes involved in education as opposed to focusing solely on narrow sets of outcomes.
00:42:38.279 --> 00:42:47.480
And I'll ask uh Sarah, do you take that and do you leave it and and what does that look like in online ed, right?
00:42:47.480 --> 00:42:50.519
This this kind of tension between outcomes and process.
00:42:51.959 --> 00:42:59.239
I take that absolutely, and I think in in online ed, I think we're potentially even more at risk.
00:42:59.239 --> 00:43:13.239
Um I think we have often marketed ourselves to understandably to professionals who have very um specific career goals and they they tie the coursework and the degrees or certificates to those career goals.
00:43:13.239 --> 00:43:19.000
And um, some people who are very busy, and and we often have students who want to get through quickly.
00:43:19.000 --> 00:43:25.639
Uh, and I think we have often supported that with our business models, and I felt conflicted about it.
00:43:25.639 --> 00:43:54.359
Um, I think one of the things that uh I totally taken this is that rethinking what education is, I think we might loop back to the comments Leo was making about um, you know, what are the bigger questions that if people are going to struggle to get some of the traditional jobs or to get the or to follow a sort of straight career path, um, then what does that mean for education and especially online, uh, say adult learning?
00:43:54.359 --> 00:44:01.159
Uh, because I think we have marketed ourselves as all about jobs, jobs, jobs.
00:44:01.159 --> 00:44:03.879
And that has concerned me for some time.
00:44:03.879 --> 00:44:21.480
And you take like communication scholars like Stan Dietz have uh he's talked about corporate colonization and this idea that we've sort of taken corporate logics around productivity and efficiency and career progression, and we've absorbed those as the goals for so many other institutions, including education.
00:44:21.480 --> 00:44:27.959
Uh, and I think that links to the obsession with outcomes and efficiency and getting people through.
00:44:27.959 --> 00:44:50.679
And while I don't want to exclude the pragmatic reasons to do that, I also think it's part of how we end up at this moment, um, where bigger questions about what it means to be human, what it means to be embodied, what it means to work and live in a liberal democracy, um, I think we've gotten really rusty at making education about these big topics as well.
00:44:50.679 --> 00:44:59.879
Uh, and I'm hoping that AI this moment will give us a chance to say, once again, we've asked for centuries, what is this thing, this education?
00:44:59.879 --> 00:45:01.319
Uh, what is it good for?
00:45:01.319 --> 00:45:06.039
And and who are we as human beings, uh, especially in relationship to AI?
00:45:07.959 --> 00:45:10.599
Leo, do you take this or leave us?
00:45:11.000 --> 00:45:11.959
Oh, I take it.
00:45:11.959 --> 00:45:15.399
I definitely see see right now as an opportunity.
00:45:15.399 --> 00:45:30.519
Um I have grown up with the education that is very standardized, and the beauty, the potential of AI is uh the personalization, the customization of the technology.
00:45:30.519 --> 00:45:33.639
This could really transform teaching and learning.
00:45:33.639 --> 00:45:41.959
Um, we're just playing with it right now, and the technology is not probably not quite there yet, but moving in a few years and a decade's time.
00:45:41.959 --> 00:45:48.759
It could, I mean, why do we need any kind of standardized assessment anymore for the standardized jobs?
00:45:48.759 --> 00:45:57.239
I mean, I think I mentioned a couple of things in my different answers, in that there may be a future where it's less standardized.
00:45:57.239 --> 00:46:07.959
And um, that means the teaching and learning, we could be freed from producing things that are to satisfy those kind of requirements that we have now.
00:46:07.959 --> 00:46:10.919
I hope that's the way we're going, to be honest.
00:46:10.919 --> 00:46:17.079
I you I mean, education, I mean, I don't think anybody thinks current education is perfect.
00:46:17.079 --> 00:46:22.599
It's kind of you know, teaching to the to the in some ways, the average, the masses in some ways.
00:46:22.599 --> 00:46:30.039
So it a lot of marginalized groups, people, you know, a little bit of outliers, they they they don't get the benefits of it.
00:46:30.039 --> 00:46:37.399
And with AI, there is a possibility, there is a potential that everyone gets that benefit that's personalized.
00:46:37.399 --> 00:46:40.359
So I um I want to see that.
00:46:40.759 --> 00:46:41.399
Yeah.
00:46:41.399 --> 00:46:42.599
Thank you, Leo.
00:46:42.599 --> 00:46:43.399
Michelle?
00:46:43.799 --> 00:46:45.399
I'm gonna take it as well.
00:46:45.399 --> 00:46:48.279
Um, for me, this has been an exciting shift.
00:46:48.279 --> 00:46:53.639
My courses are heavily writing intensive, so I'll use writing as the example.
00:46:53.639 --> 00:47:04.359
Um, I have definitely undervalued that process, that formative part of evaluation, looking at drafts and revisions and productive struggle, et cetera.
00:47:04.359 --> 00:47:19.000
But now that I've integrated AI into that process, we use the draft changes features where students can actually engage in that critical thinking and processing of accuracy and inaccuracies and et cetera.
00:47:19.000 --> 00:47:29.079
And what I've seen is um I understand why the gaps are in that polished product at the end now, or previously I couldn't see that process.
00:47:29.079 --> 00:47:33.399
Um, so I think that that's a really valuable contribution of AI.
00:47:33.399 --> 00:47:35.879
And so I see it kind of as a dual prong.
00:47:35.879 --> 00:47:40.599
Um, AI, I don't want it to relieve us of that polished end product.
00:47:40.599 --> 00:47:52.839
I do want a polished end product, but I think it allows us to reclaim that idea of um formative assessment and learning in the process as they reach that final product.
00:47:54.359 --> 00:47:58.359
I'm reminded of a talk I heard Randy Bass give years ago.
00:47:58.359 --> 00:48:01.879
He's a vice provost at Georgetown University now.
00:48:01.879 --> 00:48:06.279
Um, and he was telling about an American studies course that he was teaching.
00:48:06.279 --> 00:48:09.399
He was having his students create these short video assignments.
00:48:09.399 --> 00:48:10.759
This was almost 20 years ago.
00:48:10.759 --> 00:48:13.319
So this was a big lift technically to do that.
00:48:13.319 --> 00:48:24.679
Um, and they were trying to make arguments through these short video assignments, and he would watch their final products and he said, It's like there was learning left on the cutting room floor that he wasn't privy to.
00:48:24.679 --> 00:48:26.759
He only saw their final product.
00:48:26.759 --> 00:48:31.319
He didn't see the decisions they made, the things they decided not to include, right?
00:48:31.319 --> 00:48:35.799
The arguments they started and then decided, no, this isn't isn't the right way, right?
00:48:35.799 --> 00:48:41.799
Um, and so um anyway, I I feel like yes, the process is really important.
00:48:41.799 --> 00:48:47.079
Um, and maybe it's it's yeah, AI is pointing us to that, certainly.
00:48:47.079 --> 00:48:48.119
Sarah?
00:48:48.839 --> 00:48:58.599
One one thing I I'm I've been wrestling with lately around um the customized learning, which I think I I agree, Lou, is one of the really exciting things about AI.
00:48:58.599 --> 00:49:12.919
Um, the sort of uh flip side or double-edged that I worry a little bit about how to balance is um I could see us move towards a form of customized learning that means students don't have as much shared experience.
00:49:12.919 --> 00:49:28.119
And so I just I'm thinking a lot in my classes about uh you know universal design and giving students options and allowing and encouraging them to use AI in different ways and getting excited about the even more customized opportunity.
00:49:28.119 --> 00:49:43.239
But then I see places where I think, oh my goodness, if we're if we're too much working at different paces and using different tools and creating different projects, the opportunity to have like in communication, we talk about a boundary object, a sort of shared object for learning.
00:49:43.239 --> 00:49:45.239
I I could we could lose that.
00:49:45.239 --> 00:49:49.000
So I'm I'm trying to think about how to create that balance.
00:49:50.839 --> 00:50:02.599
Yeah, there's um Flower Darby has written some great books on teaching online, and she talks about how uh she was working with a graphic designer to create some icons for her slide deck to talk about online teaching.
00:50:02.599 --> 00:50:06.519
And the first set of icons were all like some person at a keyboard.
00:50:06.519 --> 00:50:10.599
And she's like, No, learning is a community endeavor.
00:50:10.599 --> 00:50:12.839
You need multiple people in these pictures, right?
00:50:12.839 --> 00:50:14.039
Talking to each other.
00:50:14.039 --> 00:50:26.119
Um, and you know, that's that's there are, you know, if you're teaching fully asynchronous, there's a set of tools to try to create those shared experiences that are different than if you have some kind of synchronous online environment.
00:50:26.119 --> 00:50:28.439
Um so I'm mindful of time.
00:50:28.439 --> 00:50:33.000
I want to thank our panelists for sharing their experiences and their wisdom on this topic.
00:50:33.000 --> 00:50:35.639
I really appreciate diving into this with all three of you.
00:50:35.639 --> 00:50:37.399
Thank you for being here and for doing this.
00:50:37.399 --> 00:50:38.359
Thanks so much.
00:50:38.599 --> 00:50:39.639
Thanks for having us.
00:50:39.639 --> 00:50:40.759
Thank you.
00:50:42.279 --> 00:50:51.799
Thanks to our three take it relieve it panelists, Michelle Beavers, Leo Lo, and Sara McClellan, for sharing their perspectives and experiences with AI and online education.
00:50:51.799 --> 00:51:00.439
In the show notes for this episode, you'll find links to all the essays I cited during the panel, and links to more information about each of our three panelists.
00:51:00.439 --> 00:51:04.439
I would love to hear from you, dear listener, about today's panel.
00:51:04.439 --> 00:51:06.119
What perspectives would you take?
00:51:06.119 --> 00:51:07.559
Which ones would you leave?
00:51:07.559 --> 00:51:12.679
How are you navigating the role of AI in online education?
00:51:12.679 --> 00:51:15.959
You can click the link in the show notes to send me a text message.
00:51:15.959 --> 00:51:22.039
Be sure to include your name so I know who you are, or just email me at derek at derekbruff.org.
00:51:22.839 --> 00:51:28.359
Intentional Teaching is sponsored by UPCEA, the Online and Professional Education Association.
00:51:28.359 --> 00:51:36.919
In the show notes, you'll find a link to the UPCEA website where you can find out about their research, networking opportunities, and professional development offerings.
00:51:36.919 --> 00:51:42.599
This episode of Intentional Teaching was produced and edited by me, Derek Bruff.
00:51:42.599 --> 00:51:50.919
See the show notes for links to my website and socials, and to the Intentional Teaching newsletter, which goes out most weeks on Thursday or Friday.
00:51:50.919 --> 00:51:57.000
If you found this or any episode of Intentional Teaching useful, would you consider sharing it with a colleague?
00:51:57.000 --> 00:51:58.439
That would mean a lot.
00:51:58.439 --> 00:52:00.919
As always, thanks for listening.
