Inside a Year of Generative AI Workshops: What Universities Asked For
For me, 2025 was the year of leading Generative AI workshops. I was fortunate to hear from many university leaders who wanted me to teach their faculty and instructional designers about all things Generative AI. Interestingly, there were several different requests related to specific aspects of AI. For today’s article and podcast, I thought it would be interesting to reflect on the most requested webinar and workshop topics.
So, here’s how they ranked based on demand:
Gen AI for Designing Learning Experiences
This one shouldn’t come as a surprise, but many universities and companies are trying to figure out how to incorporate Generative AI into the design process. Some are focusing on analysis, others on evaluation, and most are zeroing in on the actual design and development stages. What’s interesting is that many of them have already done their homework and experimented with a variety of tools, but yielded little to no results. Everyone seemed to flock to the basics, like using ChatGPT or creating an avatar video with a robotic voiceover as a first step, and then walked away unimpressed. There were very few cases where people were starting with a blank slate. Instead, they figured there had to be other ideas out there to help make their designs more innovative and useful.
During my workshops, I shared different ways to use Generative AI for storytelling, custom GPTs, voice translation, video generation, adaptive learning, flexible pathways, analysis, simulations, Universal Design for Learning (UDL), and more. Each of these topics helped educators see where AI could actually make a difference. It became much more tangible. They weren’t looking for theory or strategy. They wanted to see real AI use cases.
Ethical AI Use
A few months ago, I was invited to speak at InnovateEd, a conference held by Stanford Center for Continuing Medical Education. The event organizer mentioned that I had almost 3 hours to work with the participants and while at first, the amount of time was daunting for a virtual workshop, it gave me the opportunity to dive into both the pros and cons of AI. Everyone wants to hear about the innovative flashy ideas, which I totally understand, but there are several drawbacks of using this technology. Having this much time, I split up the workshop into half being about the positives of AI, and half being about the drawbacks. It felt like the most complete workshop I’ve hosted in a long, long time.
Deepfakes, intellectual property theft, student data, student privacy, environmental issues, and layoffs —there’s plenty to be concerned about. If AI is going to be incorporated into design practices, it needs to be done with intentionality, and it must be meticulously studied and tested. Just because something worked once doesn’t mean it will work again. There has to be a human involved at every checkpoint who understands what they’re signing up for. Multiple voices should be part of the process to weigh all the risks and provide different perspectives. And of course, it’s critical to share examples where things have gone wrong, so we can learn from real case studies.
Plagiarism and Rethinking Assessments
Plagiarism has always been a cause of concern for educators and this won’t disappear anytime soon. The problem is that AI detectors are about as accurate as lie detectors (these don’t work by the way), and many educators feel like they are fighting a losing battle. I shared an article recently about an NYU professor who is taking the stance of fighting fire with fire. It was a fascinating read and I’d encourage you to check it out.
The topic of plagiarism has moved well beyond the usual copy-and-paste prompts. Tools like ChatGPT’s Atlas and Perplexity’s Comet have introduced an entirely new way of cheating. If you haven’t seen this already, these browsers have AI agents embedded in them that can perform tasks when granted permission to control your screen. They were originally designed to make things like booking flights and planning vacations easier. Instead, they’re now being used to complete entire online courses with a single prompt. While this might not yet have a major impact on face-to-face instruction, similar technologies are emerging. Just watch this video and you can already imagine universities banning glasses from classrooms.
So, what do we do? My suggestion has been to refocus and direct our attention toward rethinking assessments. There’s quite a bit we can do to make them as AI-proof, or at least AI-resistant, as possible. I wrote an article called 5 AI-Proof Assessment Ideas and it gained quite a bit of traction. Some of the ideas were even featured in Inside Higher Ed and on various university websites.
Truthfully, there’s nothing groundbreaking or magical about my suggestions: teach-back sessions, recorded journals, interviews, reflections, critiquing AI, and community-based learning. Where these approaches make a difference is in their practicality. They are not a far stretch from what many educators are already doing. Asking a professor to tear a course down to the studs and start over is a tall order. Asking them to adapt existing assessments into a different format tends to meet much less resistance.
And to be clear, I am not blaming students. Many of the students I’ve spoken with are tired of AI, and so are educators. Honestly, I’m even tired of how much I talk about AI these days. But this is where people need the most support, so I am going to stay in this lane.
Ongoing AI Adoption
For many universities, 2025 was the year to make a decision about AI adoption, and there seemed to be three different camps: zero AI tolerance, moderate use under specific circumstances, or full AI adoption. Understandably, I didn’t receive many requests from the zero-tolerance crowd. The group I heard from most often was the moderate one. In these universities, many people were on the fence about certain ideas, or they felt comfortable with some use cases but not others. They were willing to try and experiment, but they needed help getting their faculty to see the value. An outside perspective can go a long way in these situations. These institutions were also the ones purchasing licenses for large language models and trying to encourage instructors to make better use of them.
What I found most interesting about these universities was how they approached their AI policies. In most cases, it was left to individual instructors to create and enforce their own AI policies. They were expected to include them in their syllabi and then explain them in class. I’m a bit conflicted about this approach. Personally, I prefer a more cohesive plan because the people who suffer most from inconsistency are the students. I can’t imagine how confusing it must be to take multiple courses, each with a completely different AI policy. I’ve even seen this firsthand with my own students, where they assumed AI would be completely banned in my class, only to find that I was teaching them how to use it in their designs. Every term, I get the same mix of excitement and confusion.
At the same time, some subject areas don’t have a meaningful use for AI, and forcing it in just to check a box doesn’t make sense. I’ve seen instructors in the same department debate this, and there’s unlikely to be a unified answer any time soon. I’ve also seen professors I deeply respect change their views in both directions. It’s an incredibly difficult time to make sense of this technology.
Staying Current
I co-hosted a workshop for one university’s faculty focused on using AI and UDL in their courses. I had the entire day to spend with them, and I couldn’t wait to share how I’ve been using AI to make my learning experiences more universal and accessible. To kick off the workshop, I used Poll Everywhere, asked them to scan a QR code, and respond with the first word that came to mind when they heard "Generative AI." I watched the word cloud appear, and the first two submissions were evil and retirement. I laughed nervously because this was not how I expected a six-hour workshop to begin!
Surprisingly, this sparked a valuable early conversation where faculty opened up about their fears and concerns before we even got into the material. I could tell they had been waiting for a space to express how they felt. After that discussion, I was able to shape my presentation and the activities I had planned to directly address their concerns.
By the way, you’ll see me use this exact same technique with Poll Everywhere during my next workshop with D2L. It’s all about using AI to design learning experiences. Make sure to sign up for that one that will be taking place on February 12th at 2 PM ET.
What was common to hear was that many had tried out some tools early on and then stopped keeping track. They had no idea about the new capabilities and features now available in many of today’s most common large language models. One professor shared his deep concerns about fake citations and ChatGPT. I looked at him a bit funny because that issue was mostly addressed years ago, once ChatGPT gained internet access. That’s not to say it can’t still make mistakes, but it’s nowhere near as unreliable as it was in 2022.
It really is hard to stay current with every update, and I noticed this became a recurring theme last year. Universities were looking for someone to come in and share what the latest developments were and whether they were worth paying attention to. Some tools are a complete waste of time and money. Others are absolutely worth exploring. NotebookLM and Google’s Learn Your Way are two tools you should be paying attention to.
And that, my friends, is a recap of an entire year of Generative AI workshops. What will 2026 bring? I’m not too sure, but I’m feeling both excited and a little anxious.
A tremendous thank you to our sponsors who support this show! By supporting them, you’re supporting this podcast and newsletter.
Poll Everywhere
Get Your Free Semester of Poll Everywhere 2.0
Brightspace by D2L
Try D2L Brightspace for 30 Days. Get instant access and see for yourself why millions love Brightspace
The Instructional Design, Engagement, and Support (IDEAS) website provides information and resources to help all members of the UMass Amherst community with online teaching and learning technologies.
