Campus Technology Insider Podcast August 2024
Jordan O'Connell 08:00
That's been interesting. And some of our faculty, we've had, you know, again, since late 2022 we've been actively ringing the alarm bell, talking to faculty about this, talking to college leadership, saying a change is coming, right? We can, we can try to stay ahead of this and figure this out, or we can let this happen to us. And so we very much tried to stay in that forefront. And so we introduced these tools as quickly as we could to faculty, and some of the faculty that are still very much trying to find their way through what this means for their teaching, their style of instruction, have been ones who have been very grateful that this tool is there when they need it, in a pinch. And so they're trying to reconcile those two realities: What does this mean for the way I've always taught my course for 15 or 20 years, and how I think I'm going to have to change eventually, because now students can just use a chatbot to answer my questions. But also this tool is incredible, and it saved me a ton of time. And so they're squaring those two realities. So to answer your question more directly, Rhea, they like it when it's there. I have a group of CNA faculty right now going question by question through what was generated. But it was incredible. The CNA course I was developing would not have gotten done in time for a variety of reasons, due to some, you know, instructor health issues and things like that, but for D2L Lumi, that was available to help me generate those questions. And now faculty are reviewing those questions before we pilot the course. So they're very receptive. And they're, it's, it's obvious to those instructors how valuable it is, and they're still in control. I think that's a critical piece, right? They are still the ones going through the question library and vetting all those questions. They're still the decider. They're still the instructor of that course. But content is made available. Instead of being provided directly from a publisher, the same group of 40 exam questions, I can generate 200 exam questions that are just as good, just as aligned to the curriculum, literally aligned to the Bloom's verb, with D2L Lumi. So, in fact, I think there are a better set of questions I can generate and that they can review.
Rhea Kelly 09:51
Do you have any pointers on kind of getting the most out of AI when you're using it for these instructional design purposes? You know, what's the best way to engineer your prompts?
Shannon Brenner 10:02
That's a good question. Let me answer that a little bit more broadly before we get into prompt engineering. But I think that when using AI tools, it works best if we have open educational resources to work with, because then we don't have copyright concerns. So that's one thing, that's one tip, I guess, for instructional design or designers, is to really advocate. We'd already been advocating for using OER, right, for the student benefits, but being able to use them really increases your opportunities and the capabilities of AI tools. I also think using multiple tools and kind of knowing the strengths of each one and the limitations of each is really helpful. So for example, I had a faculty member e-mail the other day asking if I could help him with a slideshow. He's an agriculture instructor, and he just doesn't have time to work with the technology. So he sent me two articles that he'd been using to base his slideshow off of and I was able to put those into ChatGPT and have, have it make a summary of slides, a series of slides with summary bullet points, and then I took that and put it into another tool called Brisk, which was able to create a Google slideshow with images based on that content. From there we could put that, those slideshows into D2L Lumi and generate quiz questions or discussion prompts or assignment prompts from that lecture slideshow. So being able to kind of leverage all of those tools together is really how you get the most out of them.