Point of Thought - AI & Governance

AI & Governance

The End of PDFs: How AI Video is Reshaping IT Support

A look at AI video as a new service layer for support, knowledge sharing, and institutional learning.

Funny World, Familiar Problem

I've spent years trying to solve one of the most persistent (and frankly, dumbest) IT support problems on earth: helping people log in and reset their passwords. Flyers. Word docs. PDFs. You name it, we've sent it out. The result? Confusion, callbacks, and burned time. I kept thinking, There has to be a better way.

2019: Arkansas Heat Meets Password Chaos

The first lightbulb moment came in 2019 at Arkansas State University-Beebe (ASU-Beebe). We were drowning in repetitive IT support calls, walking people through the same login and password reset steps over and over. It was unsustainable in the summer heat of Arkansas, and something had to change.

So I tried something different: we made a video tutorial. It took two weeks two full weeks to create a clean, accurate walkthrough. Every time the interface changed, we had to re-record. But even with that friction, it worked. The calls dropped. The process was smoother for users. It gave us a taste of what could be possible.

2021-2022: Illinois Wesleyan and the Manual Grind

Then came Illinois Wesleyan University (IWU). By then, I knew the workflow. I hired student workers, and we broke the videos into smaller chunks for specific tasks. Production improved, but everything was still manual.

Two weeks per video was still the norm. Edits were painful; half the time, by the time we finished one, the platform's UI had already shifted under us. Still, it moved the needle. Our YouTube channels became living archives of basic support how-tos. (We even ended up running separate channels for IWU ITS and the IWU Service Desk to organize content.) It was effective, but a grind.

2023: Shaw University   All the Pieces, None of the Glue

By 2023, at Shaw University, the strategy matured. We centralised everything on a single IT tutorial page. It was clear and accessible, with all our how-to videos listed in one place (see Shaw's IT Tutorial Videos page). But it was still just a list of links.

The videos sat there, and users still had to dig through and guess which one matched their need. No AI. No dynamic delivery. Just a prettier, more centralised version of the same old problem.

Then Came HeyGen: The Tipping Point

That's when I found HeyGen. It's a generative AI tool that doesn't just host videos; it understands a user's request and helps deliver the exact clip they need, exactly when they need it. Game-changer.

Now, when someone types "reset password," they don't get a long list of how-tos or a PDF to wade through. They get walked through the task, step by step, in real time. One prompt. One video. One solution.

I still use student workers, but now they're not just filming; they're tagging, slicing, and creating modular video snippets that can be stitched together by the AI in seconds. The pressure on our support team drops instantly. Users don't need to call for help; the system answers for us. The support experience becomes on-demand and precise.

This Isn't Just a Tool: It's a Shift

What HeyGen enabled us to do isn't mere content creation; it's support re-architected. In essence, it builds on the same backbone as my other projects with generative AI: voice bots that answer phones and chat tools that resolve problems before they escalate. Now that same infrastructure is powering video support.

Where support used to be a static pile of PDFs and outdated tutorials, now it's a real-time response engine. When the user asks, the system answers with precision.

Looking Back to Look Forward

Looking back, each phase of this journey taught us something valuable. In 2019, at ASU-Beebe, the idea of video-based support was planted (you can still find those early tutorials on the ASU-Beebe ITS YouTube channel). During 2021-2022 at Illinois Wesleyan, we learned how to scale up video production manually, and our two YouTube channels became a living library of support content. By 2024, Shaw University had brought all the pieces together on a centralised tutorial site for easier access. Each step built on the last: ASU-Beebe showed the potential, IWU demonstrated the grind required, and Shaw centralised the vision.

2019 ASU-Beebe https://www.youtube.com/@asubits1587

2021-2023 Illinois Wesleyan (ITS) https://www.youtube.com/@iwuits2548

2021-2023 IWU Service Desk https://www.youtube.com/@iwuitsservicedesk4523

2024 Shaw University https://www.shawu.edu/meet-shaw/offices-services/office-of-information-technology/it-tutorial-videos/

And HeyGen? HeyGen made it scalable. It's the piece of the puzzle that finally glued all those efforts together and breathed life into them.

The Real Lesson? Time Is Expensive

We used to spend weeks making a single support video, only to watch it go stale in a few months. Now we build helpful clips in hours. The support burden doesn't just shrink; it transforms. What used to eat up countless hours of IT staff time is now handled automatically, accurately, and instantly by an AI-driven system.

So if you're a CIO, an IT director, or a university administrator:anyone overseeing a help desk, training team, or onboarding pipeline:ask yourself, why are you still sending PDFs? There's a faster, smarter way to support your users. The future isn't static, and neither should your support be.

Want to see it in action? Visit our IT tutorial video page to see how we presented these guides. Then imagine that page with an AI brain that delivers exactly what your users ask for in real time. That's where we are now, and there's no going back.

- Leon Lewis Jr.