Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
My LMS has a fever, and the only cure is more AI
281 Plays1 year ago
Phil Hill, Neil Mosley, and Glenda Morgan discuss the impact of AI on the LMS market and how vendors are incorporating it into their softwares.
Recommended
Transcript
Introduction & AI Teaser
00:00:14
Speaker
Well, welcome again to Online Education Across the Atlantic. I'm Phil Hill, and I'm here again with Neil Mosley and with Gwenda Morgan. And we're looking forward to another exciting conversation. It's exciting to us at the very least, but that's part of the reason we want to do this podcast.
AI's Role in LMS and VLE
00:00:31
Speaker
But today we did want to jump into the topic that everybody's talking about of artificial intelligence. But with that in mind, I'd like to turn it over to Neil. Well, first of all, how are you doing today?
00:00:44
Speaker
Yeah, I'm doing pretty good. I'm doing pretty good. It's been a miserable day here in the UK, so I'm hoping that this conversation perks things up a little bit. Yeah, you should be sitting in front of the fireplace, maybe. I should be. I should be, yeah. Anyway, thanks, Phil, for that. Yeah, so as Phil mentioned, we're going to be talking about the hot topic in education AI.
AI's Impact on LMS Efficiency
00:01:06
Speaker
But rather than focusing on that as a general topic and an existential thing for education, we thought it'd be useful and interesting to actually kind of narrow in on one of the, I guess, one of the most important education technologies out there for online learning and for education in general, the LMS, Learning Management System,
00:01:28
Speaker
or VLE if you're in the UK, the virtual learning environment, because it's a real EdTech anchor point. And there's a number of leading players, I guess, in the field. And they've all been making changes to their product in response to what's happening in AI.
Client Reception and Market Impact of AI in LMS
00:01:46
Speaker
I guess there's some inevitable things that we started to see, a focus on trying to
00:01:54
Speaker
and make administration tasks a little bit more efficient for teachers. There's a lot around content creation. There's a lot around assistance. And I think there's a certain degree of differentiation between products, and maybe we'll get into that as well. But I thought it would be really useful for us to consider a few things in relation to the developments that have been happening in AI.
00:02:22
Speaker
What are the impacts on the products themselves? How is AI being incorporated?
Debate on AI's Future in LMS
00:02:27
Speaker
How are companies interpreting the different possibilities that might be out there for their products? Thinking then more broadly outside of the product, thinking about the reception of those developments on existing clients and future clients, will some of the product changes that are driven by AI affect the market and the landscape?
00:02:50
Speaker
and which companies are up and which companies are down and that kind of turning the dial maybe on those kind of things. And then I guess there's maybe an angle of
00:03:00
Speaker
longer-term look of how will AI impact maybe the centrality of these kinds of systems in the UK. Periodically, questions get thrown up around the health of the LMS or the VLE and whether it's dead or alive.
AI in Course and Assessment Development
00:03:19
Speaker
We probably don't want to get into life support territory on this one,
00:03:23
Speaker
but there's probably a kind of an angle around the kind of future of the product in relation to kind of what's happening with AI. So that's kind of just a bit of an introduction of what we're going to be talking about today. So I thought I'd kind of kick off
00:03:38
Speaker
Maybe if I can pass this to you Morgan on. Just really in terms of what types of things you've been seeing amongst the kind of main LMS vendors, what kind of developments have caught the eye, how they've been using AI. It'd be really interesting just to find out the what of AI developments in this space to kind of kick off with.
00:04:01
Speaker
Sure. Thanks, Neil. And, you know, summarizing, I think, you know, especially amongst the sort of the big three commercial products there. I mean, I would say it's it's course development, assessment development, and to an extent, something of an emerging focus on tutoring or coaching, you know, so going through the topics and finding sort of ways that that that people can be helped. So that's where I sort of see most of it at the moment.
00:04:31
Speaker
And I'm, in general, a bit torn here, because I like the fact that it's within the LMS, because one of my frustrations with AI is that in some ways, it's not a thing. You know, it is a thing,
AI in Product Storytelling and Vendor Strategies
00:04:44
Speaker
but it's not its own tool or function.
00:04:49
Speaker
it adds to what's going on inside the LMS. And so I do like that sort of focus within the tool and sort of how that does that, you know, because I think it can help us sort of start to think about how it's going to evolve and emerge. I think it is still just really early days yet. So I think what we're seeing now is very different from what we're going to see in the future. But what do you think, Phil?
00:05:16
Speaker
Well, I would add, besides content creation, the items you mentioned, let's talk about what could be the biggest one, marketing. I mean, one of the biggest uses of AI is to be able to say we care about AI and we're developing AI. So there's a, that's a real, and it's natural for products. There's a lot of pressure on tech companies to have an AI story with their investors, but with customers. And so I see a lot of that happening right now. Now, to be fair,
00:05:46
Speaker
I think they're under pressure to do that. And I think it's actually a healthy process to start throwing things out there, get schools to react, figure out what makes sense, and start figuring out where AI is going to be useful in this space.
00:06:03
Speaker
And in that regard, I mean, all of the three major commercial vendors, and I would throw in Moodle there as well as like the major open source provider, they all are starting with sort of here are guiding principles. So they're putting a stake in the ground of we care about privacy, avoiding discrimination.
00:06:25
Speaker
keeping humans in the loop, etc. So they all put out their guiding principles and then are developing off of that.
00:06:34
Speaker
I definitely agree with you with where we see it going. I would sort of frame that as the LMS vendors naturally are going to try to put the solution within the LMS. They have a vested interest that it's part of their workflow. And so initially, you're seeing not just what could AI do, but what could AI do to augment our process and to augment our workflows.
00:07:03
Speaker
Some of the things that we saw or that we've seen early on fall into that range, like what's possible, what can fit within our workflow. And, you know, like if you go to some specifics, D2L came out. One of the things I found most interesting about them was having a formative assessment generator. So if you're a faculty member or a course designer,
00:07:26
Speaker
And then you get a prompt at where you can say, give me a formative assessment based on this text or this portion of the course right before there. That to me is very constrained. It's within a workflow. It makes sense. And I personally like it because that's an area where I think that learning design needs to be improved, better usage of formative assessments to help learning. Now you go.
00:07:56
Speaker
Anthology, Blackboard Learn, they seem to be taking the approach of, we're going to put it in, or at least talk about putting it in, in multiple places like it's AI throughout the product. And so they had more of their
00:08:11
Speaker
more different functional examples they're throwing in. And structure, they had a chat bot type thing. And then really what they showed at their conference was a lab where they're exploring some of the deeper topics and working with university partners about what to do.
AI's Transformation of Educational Processes
00:08:30
Speaker
And then finally, I look at Moodle. Moodle is how can you come up with a press release without actually having anything in place is what they've done.
00:08:38
Speaker
I mean, go read their press release in their discussion of AI. They reference guiding principles that we believe in them, and I think they have it. They reference here areas we're going to be looking at, but the only specifics they had were plug-ins, something somebody else has just done as a natural plug-in. So Moodle itself has no strategy on AI yet. They only have principles and a nice press release.
00:09:06
Speaker
So yeah, we're seeing it all over the map, if you will. But it's very constrained by what their natural workflows are. And you're right, it's going to change over time. But that's how I would summarize what I've seen so far. Yeah, that's interesting. I mean, I kind of echo that in that.
00:09:27
Speaker
I suppose there's been a natural focus on some of the obvious things that the obvious ways in which AI can be incorporated into their products. And there are some similarities of approaches. And like you say, Phil, having kind of looked through some of the product developments and roadmaps and things, Moodle does seem to be the one that's slightly light on.
00:09:51
Speaker
on some of the kind of developments and you know maybe that's also slightly to do with the fact that it's a slightly different flavor of LMS to some of the others but I just wondered you know maybe to that point that
00:10:06
Speaker
the vendors haven't gone really big and radical in terms of AI yet. I just wondered, in terms of maybe what we're not seeing or what you'd like to see them do and what you think the kind of possibilities might be to do something that kind of goes beyond the workflows and the kind of existing things that LMS is unknown for.
Simplifying AI Interactions in LMS
00:10:30
Speaker
I just wondered if you had any thoughts on where the products could
00:10:36
Speaker
really turn the dial in terms of their use of AI, Phil, maybe, if I could come to you? Sure. Well, I'd say the big areas of fact, if you look at what's happening, particularly much more with open AI than, say, with mid-journey, is the universal user interface. How do you actually interact with your computer program to get something done?
00:10:58
Speaker
And I joke because I keep using mid-journey and I think there's sort of a masochist side to me for doing that. I mean, you have to get on Discord and come up with a complex prompt. And then once you type in the prompt, you've lost it and you have to scroll and figure out how to find it.
00:11:15
Speaker
One of the reasons OpenAI is getting so much attention is they're going the opposite direction. They're going with, there's a new way to interact with a computer and it's not about pull-down menus, it's not about, it's just simple. Simple language, one thing, it's almost like what Google search was for finding information. OpenAI is going that direction with the very natural, simple way to interact with your computer.
00:11:43
Speaker
So when you see all of these AI tools within an
Ethical Considerations and AI Partnerships
00:11:48
Speaker
LMS, they're fitting within the pre-existing interface. It's just another button they've added to a particular function, or it's a suggestion that you accept. But it doesn't at all change the nature of how you interact with the computer and use natural language to do so. So that's the biggest one that I think that we're not even trying to deal with yet.
00:12:11
Speaker
Yeah, and I guess that's interesting in terms of, although you weren't kind of addressing it particularly, that idea of integrating with the bigger players in AI, because I think in relation to what you were saying around the different sort of
00:12:28
Speaker
ethical aspects of AI or the kind of more policy statements that companies have made. That's really interesting because I know in the design world, you have companies like Adobe that are deliberately veering away from, say, some of the image generation software out there that they know that they're
00:12:49
Speaker
their user base are going to feel strongly about because it's been trained on people's work essentially. And so I just wondered whether you think we might start to see that degree of differentiation around some of the products in terms of who they
00:13:09
Speaker
who they integrate with or who they marry up with in terms of the kind of bigger AI ecosystem. We saw in the UK, there were issues around turner-ins detection of AI.
Concerns Over AI Personalization in Education
00:13:23
Speaker
So I wonder, Morgan, whether that could be another controversial avenue or maybe even an area of differentiation for products in terms of what they don't do, I suppose.
00:13:37
Speaker
Yeah, I mean, I think that will be interesting. You know, it's been interesting this weekend, I've been listening to a lot about AI. And every time people talk about human in the loop, I see a guy walking in front of a car with a flag. And maybe I'm just too much of a techno.
00:13:52
Speaker
techno enthusiast. But you know, I think there's a lot of anxiety right now. And so I think that is driving it to an extent. Turnitin is almost the opposite, like they went gung-ho right there at the beginning and maybe made some claims that perhaps weren't sort of substantiated in the end.
00:14:16
Speaker
Right, maybe and perhaps that seems like you're giving them a lot of the benefit of the doubt there. Yeah, I'm trying to I'm working on my diplomatic skills here. It's a journey. But, you know, when you first I'm going to change the direction here, but when you first asked, you know, because I
00:14:38
Speaker
I don't have any thoughts about which sort of which sort of way that they're going to integrate in terms of actual companies. But I, when you originally asked for that question, I was thinking, Oh, sort of what sort of type of thing are they going to do? And I think one thing that they're obviously going to try to do is personalization. And I have very mixed feelings there, because personalization is one of those words that drives me around the bend.
00:15:04
Speaker
I for years have been brewing an article called personalization is bunk because most of the claims about personalization are absolutely ridiculous and and it seems like a even absurd thing to want to do you know it reminds me of bad surveys that sort of get to things but I think
00:15:26
Speaker
I do have mixed feelings here because maybe with AI, we can actually start to get there, you know, customized pathways, for example, and things like that. So I think that that actually is a productive way to think about it. So that not everybody is doing the same sort of thing without the sort of tremendous amount of work involved in actually doing real, real adaptive learning or something like that. So I think that's sort of one way that that things will go in the near term.
00:15:53
Speaker
Well, early on, like at the conferences, they were stepping all over themselves, particularly the three commercial vendors, if you will, Canvas, Brightspace, and Learn.
00:16:04
Speaker
to say, oh, we're not cutting faculty out of the loop. So one of the things they're not doing, or at least they come strongly stating, is we can't take them out of the loop. So even if there's content generated, we're going to let them. Morgan's waving her flag in front of the car. I think that's the sign that I'm getting. But yeah, they all wanted to say that. It's all, hey, we're going to let faculty determine what actually gets presented. And that is a very real issue.
00:16:33
Speaker
That wasn't a differentiation. That was what we're not going to do statement by all of the major vendors. We're not going to create content and cut faculty out of the loop. We're going to do it to let them make the final decision. Well, have you seen the same thing, Neil, from what you're looking at since you've got more of the learning design, learning instruction background on this?
00:16:58
Speaker
Yeah, I mean, I guess from a company's release perspective and kind of general ethos perspective, they're going to want to try to balance the line a little bit in terms of kind of innovating with AI, but also not hacking people off. And I think part of that is the faculty side of things. I mean, just going off on a slight related tangent, there are people experimenting with
00:17:25
Speaker
creating online courses with more and more minimal interaction from subject matters experts and academics and obviously that is that will be very very controversial but
00:17:39
Speaker
if that kind of increases in efficacy, that's going to be a challenge, but a no product is going to go and say that at this stage. So, yeah, I mean, I've seen a little bit of that. I mean, I think that's just a natural position that people are going to, products are going to take at this stage. I mean, I think
00:17:59
Speaker
to your point Morgan, the whole personalization thing I think is really fascinating because I got to a stage recently where I was just so fed up of people citing that same research paper. So the Bloom's two sigma research paper and basically say, look, this paper in the, I think it was in the eighties, AI basically fulfills the promise of this paper and to see, you know,
00:18:25
Speaker
Sal Khan on the stage and various others of kind of maybe sort of slightly lesser profiles kind of using that as the basis for kind of AI solving those kind of problems. But I mean I think it's interesting in terms of the personalization aspect because I think that is just a lot more complicated than often it's presented. One of the things that I one of the thoughts I had around personalization was there's also so much talk about
00:18:54
Speaker
lifelong learning and more flexible routes through education and so you know that made me think about the personalization aspect from that kind of thing but maybe more on a route through a degree rather than a route through a particular course and that could be an interesting kind of avenue for for AI in a product.
00:19:16
Speaker
But I was also wondering, just taking this back to the product development and the feature side of things, I remember, I think, Phil, you're talking about this idea of the walled garden of the LMS and around social media and then LMS vendors essentially incorporating things from the outside world and done badly.
00:19:39
Speaker
And I wonder whether you think there's any danger that with AI vendors may do a similar thing, although it's different, whether there would be any wall garden strategy that might hinder them, I suppose.
00:19:59
Speaker
Yeah. And just for people who haven't read any of those posters, and I'm certainly not the only one that's talked about it, but particularly before we had standards such as, well, I'll still call them IMS Global, One EdTech and the LTI standards.
00:20:17
Speaker
you know, you would say, who are the third parties innovating? Well, a lot of times it would even be social media, not an ed tech. And the initial LMS response was to say, oh, you like social media? We'll put a crappy version within our walled garden. Oh, you like a blog? We'll give you a crappy blog within the walled garden. And they just kept adding this stuff. And first of all, they were all crappy replicas of the real thing.
00:20:42
Speaker
And then you're absolutely right. It just made it a fully bloated system that was a terrible user interface. So that was more back in the 2000s when we had that, although it took years to sort of unwind that approach. So leaping forward to AI, the way that they're trying to fit it within their existing workflows with the vendors, yes, there is a danger of this. Right now, it's only an extra button or two.
00:21:12
Speaker
that we saw at the user conferences this summer. So it didn't jump out at you this summer like, ooh, that's an ugly way to plug it in. As a matter of fact, there were some nice designs that we saw. But that's just stage one. What happens when they add more complexity to it? And they're just going to have to add these controls. And if they're trying to shoehorn it into their workflow and their way of doing things, absolutely, I think that's a risk.
00:21:40
Speaker
Certainly when, as I mentioned before, I think the idea where AI can really simplify the way you interact with a computer, then the way that people assume technology can work is going to be going the opposite direction. Make this simple. Make this easy, natural for me. Now all you're doing is adding an extra button to what I already had. So it's not just,
00:22:09
Speaker
that they're designed to become hugely more complex. But even being in the same place and not taking advantage of simplification is going to meet changing expectations from the younger folk who don't take the same assumptions about the way things should be. So yeah, I do think there's a risk of it becoming additional bloat. But again, it's not so much
00:22:36
Speaker
from how they're designing it today. It's how they're wanting to fit it within their existing workflows. That's the problem. Yeah, I would agree on that. And I think there's additionally another sort of drive pushing them towards building the walled garden. There's like, if you talk or listen to Khan Academy and Duolingo and those products that are further down the line to sort of understand how they tried to build
00:23:07
Speaker
AI, and generative AI into their products, because they often the those sort of companies got got early access to it. So there's slightly further down the down the road than than than some of the LMS is to keep it safe, you know, they had to have guardrails and and things like that, you know, so that you if you're in Khan Academy, it's not going to give you some
00:23:29
Speaker
some terrible hallucinatory answer or something like that. So I think that pushes the additional push towards it being a walled garden, just keeping those guardrails in place. Also when Phil was talking about the sort of constant, you know, going back and forth about a walled garden versus open, a walled garden versus open, because maybe I'm a very strange person, my sort of head went to Jimmy Carter, you know, he was once, he once confessed in a,
00:23:59
Speaker
in an interview that he had sinned in his mind. He committed adultery in his mind. I think sometimes it reminds me a little bit about Ed Tech vendors. Even if they're not doing it now, they're committing walled gardenism in their minds maybe sometimes. And the malaise of the market, we'll have to bring that in then.
00:24:22
Speaker
Yes. I am sitting here in a in a cardigan kind of situation. So I am I'm living the Jimmy Carter world. Yeah, that's that's a we, we probably shouldn't go down the route of American presidents. I mean, in discretion should be really bad. I thought that you mentioned a few times the kind of can't me go thing. I thought that was an interesting counterpoint to the
00:24:52
Speaker
to the wall garden thing in that, you know, Canvas kind of decided an instruction kind of went down that route of maybe going to someone that was a leader to incorporate their developments into their product. And I know for Blackboard's assistant, they talked about in collaboration with Microsoft. So there seems to be, I think that's kind of a looser collaboration in terms of using there as a open AI side of things.
00:25:22
Speaker
But I guess that kind of is an interesting maybe counterpoint to this idea of wall garden, people, products kind of collaborating with companies that may be a little bit further ahead to your point, Morgan. But I wonder kind of on that point as well around differentiation and just kind of zooming out a little bit from the product side of things as well, but thinking about the market and what
00:25:50
Speaker
changes and what decisions different vendors make, how that might impact the market or whether we feel like actually there's not enough differentiation for that to have an effect. I just wondered what you're seeing now in terms of the potential for differentiation and what you think might happen in the future in terms of AI really separating products out in a meaningful way when clients come to
00:26:18
Speaker
to come to consider them Phil, I just wondered what you thought about those two things.
00:26:23
Speaker
Well, it did strike me this summer that they were taking very different approaches between the different vendors. And so for go to D2L. D2L is putting a lot more emphasis on content creation. So one of their things is Creator Plus selling alongside Brightspace LMS. Now there's some functional overlap, you know, because you can create things within the LMS, but they're going into content creation.
00:26:51
Speaker
And I find that they naturally, like the formative assessment sort of went down that route. How can we help and make it more elegant? Just like Creator Plus is intended to make more dynamic, interactive course designs, if you will. Well, now wouldn't it be neat if you could very easily add formative assessments to the mix?
00:27:13
Speaker
to make it really easy and make suggestions. That's how they did their initial demo of AI. And to me, it fit within that content creation mindset.
00:27:22
Speaker
Morgan was the anthology together, but I know one of their bigger ones was course design overall. So not the content per se, but the overall course structure. And then they also seem to be going down to, we have it plugged in here, here, here, and here, which has been the traditional Blackboard approach. We do many things in many ways. In structure, it's harder to put a
00:27:47
Speaker
label on what they do, because I didn't find that they had a coherent strategy. But they did, as you mentioned, they did have the COM-MEGO partnership that they led off strongly with. And they are really trying to work with university partners to think of the thornier problems out there, sort of a research lab approach.
00:28:08
Speaker
So and then, as I said, going back to Moodle, Moodle is talking about it, but the only work being done is by whoever's building plugins. So I see it very differentiated. Now, what does this mean long term? I guess the open question is, as we go along, if the vendors see, hey, this
00:28:29
Speaker
Our competitor is getting some real traction with that. We better do the same things ourselves That might pull them back down to a more common approach but at least out of the gate I found that their ai approaches to be highly differentiated compared to what I expected Yeah, that's interesting because I wonder morgan like in terms of the things that the things that really influence universities around these kind of choices because
00:28:57
Speaker
I guess we'd love to believe that universities always make a deeper kind of decision. They've looked across all of the different variables, but I guess we know different edtech products that sometimes the due diligence isn't done, or sometimes that there's
00:29:15
Speaker
specific things driving implementation or a desire to move away from a different product. And I just wonder how much you feel the vendors themselves are tackling the common things that come up around influencing factors for institutions changing or deciding on a new product, whether the AI developments are actually really focused in on the things that we see rise to the top when
00:29:46
Speaker
when universities and higher ed is kind of thinking about a move or a new implementation.
00:29:53
Speaker
Yeah, I think they are in a way because I think some of the big things are ease of use and savings instructor time. And those are big things that I think they are sort of ultimately aimed at in their different approaches as sort of feel nicely laid out. So I think they are sort of getting there. And in some ways, one of the things that we try and encourage
00:30:22
Speaker
universities to do is be strategic in choices. Don't get bogged down in this tweak versus that tweak, or this little widget down here versus that little widget down there. So be strategic. So I really think they need to step back and differentiate between those different products, like Phil did, and think about which one more ties in with where we're going there. But I think they are tied in. I think there's also a negative aspect of that, and partly
00:30:51
Speaker
I'm reading Brian Rosenberg's book. He's a former president of Macalester College in St. Paul. And he's got this book called Whatever It Is, I'm Against It. And the way that so much in higher ed is consensus driven. And that really stops you from being from really
00:31:10
Speaker
being innovative or being transformative. And I think that some of the vendors have just gotten infected by that same sort of thing. And so they're perhaps a bit gun shy, you know, so I would encourage them to be more bold. And I can I know that they're listening to this saying, well, the minute we're bold, Phil Hill will be honest and criticizing us for that. But I think they should not be bogged down by the by the consensus. And it's a really great book. So go read it.
00:31:36
Speaker
You've already got my interest just by the title of it. One of the vendors told me this at a recent conference is their tools or their AI stuff is getting them into the door quite a bit with schools. So even existing clients, hey, come talk to us about AI. And it almost comes across like it's not necessarily tell us about that feature you introduced and what it does do and what it doesn't do.
00:32:05
Speaker
It's more, hey, you guys seem to have some credibility and are thinking about AI. Help us think through what we want to do. So it's almost inviting vendors to be strategic consultants for the schools. And part of that, to be honest, I think represents the very poor state of what we're seeing in conferences. Morgan's going to be in Berlin next week. We'll find out if it's different.
00:32:30
Speaker
Across the board, the conferences I've seen recently, I have not seen very impressive discussions about AI that's happening. And I don't want to say it's the conference organizers fault necessarily, but there's a hunger for academics to figure out what does this mean? How should we think about it? Where does it go?
00:32:50
Speaker
And the part of the alignment is just even letting the vendors talk to the schools about that. Here's our approach. So I'm not sure how that fits in right now, but it sort of is, it's given them permission to talk about it strategically in a way that's much deeper than academics are getting in academic only circles.
00:33:16
Speaker
Yeah, and I guess that's an interesting one in terms of maybe when you think about product developments where maybe they've got a little bit more license, not necessarily to kind of throw out the shackles of being gun shy, but to take the lead a little bit in terms of what it might mean for instructors in particular around using the tools. And I guess, you know, you've seen that in terms of test creation, rubric creation, those kinds of things.
00:33:44
Speaker
I mean, I think there's an interesting balance between the kind of the innovations and the developments that are focused on instructors versus students. I mean, ultimately, you know, if an instructor has a test creation tool, then the hope is that that would benefit students. But I just wondered, to going back a little bit maybe to differentiation or what you've seen that's happening, do you feel like there's a
00:34:13
Speaker
stronger focus on instructors rather than students. We talked about that kind of mantra of the human in the loop. Do you think that there's a balance between the developments that we're seeing that are more focused to instructors versus students, if I can make that dichotomy?
00:34:35
Speaker
Oh, I think you can. And I think that's absolutely happening. When we talk about the fear or talking about them being gun shy, I think was the phrase, the primary way the vendors are being gun shy is how do we not scare faculty? That is what's driving too much of it. Some of that's a good thing. You don't want to put out bad content. So you don't want to just turn it out with no guardrails. And then all of a sudden you're giving wrong information to students.
00:35:04
Speaker
But in that balance, there are 80 to 90% focusing on the faculty side and not on the student side. Now, they'll say, I guess you could argue the chat box, tutor coaching.
00:35:20
Speaker
You know, there could be an argument saying, oh, we do care about the students and this is good for them. But it's it's almost let's put in a faculty friendly way to help students. They're not sitting back saying, hey, let's really first and foremost figure out what students need and what the possibilities are.
00:35:39
Speaker
and then figure out the guardrails. They're almost starting out with the guardrails, and most of that is driven by too much of a faculty focus. It's a spectrum, not arguing all or nothing, but I think it's very clear where the focus is right now. And do you think there's an element in that kind of focus and that kind of balance that maybe almost takes the view that in the same way we talked about the kind of the social tools and the integration
00:36:10
Speaker
It's not necessarily around poor imitations, but it's that students were going to be using stuff outside of the LMS and therefore, you know, maybe that's also kind of a hedging your bets kind of approach. I don't know if, I don't know what you think to that Morgan with the, you know, vendors are maybe focusing more on instructors as well, because they think students might look elsewhere for, for these AI tools.
00:36:36
Speaker
I think there was a really great observation on your part and a really great question. I think you're giving them much more benefit of the doubt than I would tend to. And I'm going to take a little detour into a story that I've told before on the blog, but which haunts me, which
00:36:53
Speaker
Some years ago, I was interviewing students about how they use technology, and there's one particular student. It was in a focus group, but when I think back, all I see is her. And it was at the University of Illinois, and she was talking about she was in this mathematics class, and it was terrible.
00:37:09
Speaker
absolutely awful. She went to every single class because she wanted to know what was on the exam but as soon as he introduced the topic she'd put on her headphones and she'd start googling to find other explanations of that topic because this guy could not explain Diddley Squad.
00:37:25
Speaker
And she was in the class, she was in the LMS, but being forced to go elsewhere. So we need to find a way to address that because she was spending a lot of time hunting down and evaluating things, which maybe was good for her in the end. She was an engineering student and she's now no doubt a wonderful engineer, but we need to find solutions and vendors need to find solutions for that problem without just relying on other people meeting that need.
00:37:55
Speaker
Yeah, I wonder if that almost leads us onto a more existential question about the LMS and AI, because I guess thinking about online education where there is that kind of...
00:38:07
Speaker
different relationship and different kind of distance from the university or different type of distance from the university. There is a sense in which there's kind of tools out there that enable those who are either fed up with the quality of the instructor or just have a disposition to go and find stuff out for themselves to more so than usual, maybe circumvent some of this technology. And I wonder,
00:38:33
Speaker
whether there are any existential threats to this as a product or not. Phil, I wonder what you think. Well, I mean, if there's something you can bet on, it's that...
00:38:47
Speaker
Every few years, somebody is arguing the LMS is going away. It's become a commodity and it's been a very safe bet since at least 2005. Michael Feldstein worked at SUNY at the time and they were looking at a learning management operating system. We don't need the centralized company. We need to patch together tools.
00:39:07
Speaker
So from the middle of the 2000s, there have been quote unquote existential threats. Yet every year, the LMS becomes more and more essential to university operations. And then that really jumped up during the pandemic.
00:39:21
Speaker
So it's not a wise bet to bet against the LMS market centrality anytime, even with generative AI. Having said that, keep in mind that part of what they do well is managing courses, not learners really, but courses. And it's all the administrative stuff, getting things organized, presenting grades.
00:39:46
Speaker
doing this administrative stuff. And that, for the most part, I don't see it being threatened by generative AI. I think that publishers, courseware developers, are in a much worse position or face a much more of an existential threat here because the nature of AI is to democratize content generation, to put it closer to the hands of the people who want to develop it, whether that's
00:40:13
Speaker
individual course designers who don't want to have to start with a courseware tool or textbook or faculty that want to do it themselves or academic department working together. AI is going to make their job easier and more feasible and I think a lot more of that risk is going to fall on the publisher industry.
00:40:35
Speaker
than it will on the LMS itself. And actually, I do want to turn it to Morgan. One of the demos that you saw this summer was about course design. So which sort of raises the question, does it sort of commodify the ability just to create an overall course design within an LMS? Not sure who that threatens, but you seem to have a pretty skeptical take that that's what people actually are looking for. Yeah. And in some ways, it was the
00:41:05
Speaker
The conference was in Nashville, Tennessee, the home of country music, and I saw the demo twice, and the first time they were going to construct a course about southern food, and the next time they were going to construct a course about country music. Now, there are two things I deeply care about in the world. One of those is food, especially southern food, and the other one is country music.
00:41:27
Speaker
I'm the uncool one in this trio. We've got a rock guy, we've got a jazz guy, and we've got me. So I'm the deeply uncool one. So maybe it was the content of the courses, because I know a lot about food and a little bit about country music. I was underwhelmed by the course construction thing, but it is early days yet. I'm also underwhelmed a little bit by the coaching that I've seen, but it is early days yet.
00:41:57
Speaker
But even if you look forward to a time where it does a really good job designing your course, who, if anyone... So who's the power there? Yeah. Yeah. Who actually gets threatened by that? And is it the LMS companies or do they actually? No. This allows other people to do their jobs more efficiently. So who gets threatened in that case?
00:42:16
Speaker
it would be instructors and faculty, I would say definitely or universities as well. And then also, you know, perhaps some of the there is a large but growing or, you know, a small but growing contingent of companies that sell courses to offer that that universities take and offer as as certificates or things like that. They certainly are threatened by that. So I think I mean, I think that's an interesting one to to unpack.
00:42:46
Speaker
Yeah, I'm hearing a few different things. Maybe there's that tension between other education, other aspects of the education landscape pushing forward with AI in a way that maybe circumvents the role of a traditional role of an instructor, whereas maybe in HE, sorry, there'd be
00:43:07
Speaker
you know, a bit more resistance to that. And it's similar to what you were saying, Mark, and I've seen some of the course creation stuff within LMS and I've not been overly impressed. And I've also been thinking, I'm not sure I know many, I've come across many instructors that would
00:43:25
Speaker
would necessarily want that at this stage. So I think that's an interesting dimension of the two tracks of this side of things. But I don't know what I was hearing maybe a bit from you Phil is that actually rather than threaten the LMS,
00:43:41
Speaker
this could reinvigorate the LMS and make it more central. And actually, all of those ecosystem of things like the quiz tools and the courseware tools actually might be more threatened than this kind of hub that is managing all the courses. And certainly in the UK, that component of higher ed where it's kind of unbundled and the need to manage courses feels like
00:44:10
Speaker
is not going to go away really. Yeah, I would not want to be a textbook publisher right now for this exact reason. But yeah, if a company is well managed and strategic, if an LMS company is doing that, this is an opportunity to become even more entrenched, which I know
00:44:30
Speaker
That view would not be popular with a lot of people in ed tech circles But yes, I if I had to put my money on it right now five years from now AI is a net plus for the LMS market in general maybe not for each vendor But for the market in general and how important the LMS I think it's going to actually Enhance it within the next five years now longer term things could change obviously, but yeah, that is my gut feeling
00:44:58
Speaker
We'll have to revisit this if we run this podcast for at least five years. We will, yeah. We'll have to get something in the diary for who we write about LMSs in 2028. And the big question that we haven't taken on here, though, is also the sort of challenge of things to the LMS, challenges to the LMS from products like Teams and things like that as well. I mean, and maybe they are just versions of the LMS that are
00:45:28
Speaker
nascent or emergent. Well, that might be its own episode here. Microsoft Teams has talked about it forever, and I don't see much traction. But it's particularly given Microsoft's central role as one of the hubs of generative AI development and models.
00:45:50
Speaker
that might change the equation. So I think, yeah, we might have to save that for a future episode to talk about that aspect. I'll add one other end. I know we're getting short on time, but that's the cost of doing things. We keep acting like AI is free, and it's not. AI introduces tremendous cost to pay for the server space, to pay for the models, to pay for the licensing of the different tools. AI costs money.
00:46:19
Speaker
That's gonna put a financial burden on the LMS companies. So the stronger ones financially could and let's name names I mean and structure is the vendor that is doing the best from a financial perspective They can handle this a lot better than their competitors Anthology is going through some financial challenges again
00:46:43
Speaker
And they're promising quite a few AI tools. Well, it raises the question, how are you going to pay for all of that usage? Your internal costs are going to go up. So I don't know the answer, but is this going to put pressure to increase typical LMS subscription license fees for schools?
00:47:01
Speaker
Or are they going to tend to swallow the cost internally? And I don't know the answer right now, but I'm just saying, I think we have to watch that as well. What happens to the price pressure internal to the LMS vendors as they make a lot of promises right now? Yeah, absolutely. And I guess that's just another element of
00:47:22
Speaker
where we're at in terms of not quite being clear where things might end up in a few years time. But that's been a really, really interesting discussion. Maybe if we wrap it up there, Phil, and you kind of sign us out. Oh, good. Hey, well, love the conversation again. And thanks for everybody joining us. And we will talk again in another week, another topic about online education.