Transcript Summary:
Carol:
Welcome to AI in Action! Before we begin, I want to thank Burnette Insurance for lending their office space to us to host this event. Burnett’s been a really good friend of MIS over the years, so we really appreciate it. On behalf of our speakers — Lliam Holmes, Scott Pressimone, and Matt Park — thank you all for being here today.
You’ll have an opportunity to ask questions of each of the speakers after they share their experiences with AI. After all three have spoken, we’ll bring them all up here for a panel discussion, so please don’t be shy. We might even have some gifts for those of you who participate with us.
Artificial intelligence has been a topic of conversation for many years now. Back in 2023, some of you probably remember, Lliam spoke at our annual Tech Exchange event about the emergence of AI and how it was going to be the next big thing — to change the way we do business much like the internet did in the 1990s.
So here we are two years later, and many of us are still struggling with how we’re going to use AI in our own businesses. We know we need to start, right? We need to get on board so that we’re not left behind by our competitors. But where do we start?
In conversations that Lliam has had with many of our clients, he recognized that a lot of business leaders are still just trying to figure it out. I want to emphasize that we are not selling AI tools. Our goal today is simply to share ideas, challenges, and mistakes we’ve made along the way, in hopes of inspiring you.
So hopefully we can learn from each other and create a community of like-minded people who want to take advantage of everything AI has to offer. With that, I’d like to invite Matt Park to introduce himself and kick things off.
Matt:
Thank you so much.
When Kevin first said, “Hey Matt, they’re looking for people to be on a panel about AI,” I said, we don’t do AI. But then I thought about it and realized — actually, we do. Once I really started thinking about what we use internally and what we’re using with clients, there’s a lot of AI happening.
I thought back to when I was in college at Vanderbilt as an economics major. One of my summer jobs was as a research assistant for a professor writing a book called The Third Wave. His idea was that we’d gone through the agrarian revolution, the industrial revolution, and now the technology revolution — the internet revolution.
I was thinking about that on the way here, and I realized now he’d need to write The Fourth Wave, because this is it. AI is here — not coming, not next. It’s already here. And if you’re not using it, you’re behind.
I’ve been in the copier-print business for over 20 years, and you’d think that’s about as old and stodgy an industry as you can get. But we’ve had to advance too, and there are a lot of things we’re using AI for.
I do tell my sales reps to be a little wary, though. One of our reps loves using AI to send emails, and they sound nothing like him in person! I told him, “You might want to tailor that so your email doesn’t say, ‘I was hypothesizing about…’ and then you show up saying, ‘Hey buddy, how you doing?’”
Let me share a few examples of how we’re using AI.
We do office equipment — copiers, printers, software around imaging and document management. We also work with interactive panels in school systems and businesses. One of our vendors for K-12 environments has an AI-powered lesson planning tool where a teacher can say, “I need a lesson on algebra,” choose a difficulty level, and the system builds out a lesson plan.
So AI is helping enhance the classroom experience. And as business leaders, we need to know that as kids grow up using these tools, they’ll expect to use them in the workplace. If we’re not using them, we’re behind.
Even our “dumb” copiers have AI built in — learning which buttons you push most often and re-ordering your menu automatically.
The biggest area we see, though, is document capture and document management — workflows, approvals, invoices, and automation.
Historically, people would scan an invoice and key in the invoice number, date, amount — all manually. That’s tedious and inefficient. With AI, we can teach systems to recognize invoices automatically, no matter how many vendors or formats you have.
One of my clients was processing 10,000 invoices a month with six employees. With AI, they’re now processing 20,000 invoices a month with three employees. And they didn’t lay anyone off — they moved the extra people into higher-value, customer-facing roles.
It’s real ROI. Their AI solution cost less than one full-time employee annually and doubled their productivity.
AI is also changing print management — routing print jobs to the right printers automatically, reducing waste, and improving remote printing security.
The key takeaway: AI is already here. It’s in your workflow tools, your print management, your document management — whether you realize it or not.
And one word of caution — when you’re on Teams or Zoom and you let that AI bot record the meeting, remember: everything you say is transcribed and shared. So don’t say, “Oh, that customer’s always late.” It’ll be in the transcript!
That’s all I have — thank you.
Carol:
Okay, does anyone have questions for Matt?
Scott:
We can take a few now, or you can save them for the panel.
Carol:
I actually have one. You mentioned that your client was able to move employees into other positions. Did they get any pushback from those employees?
Matt:
No, because they hated those jobs. They were tedious. Those are the kinds of employees who would leave for fifty cents more an hour, and then you’d have to retrain someone new. Now they’re in roles they enjoy — customer service, sales — and they’re adding more value. It’s been a big morale boost.
Carol:
Thank you. Anyone else?
Guest:
What system are you using for invoice AI?
Matt:
We’re resellers for several products, but one main one is PaperVision from a company called Digitech. It includes PaperVision Capture for data extraction and PaperVision Enterprise for content management and workflows.
For clients who only want data capture, we can connect to OneDrive or SharePoint. Others want full workflows and approvals, which PaperVision supports. It can even do handwriting recognition and image summarization.
The pricing varies — some are all-you-can-eat models, others charge per document or per process.
Guest:
Have you encountered fake, AI-generated invoices slipping through?
Matt:
We still keep a human in the loop. Clients do a three-way match — invoice, PO, and check number — and if anything doesn’t align, it’s flagged. AI helps extract and pre-sort, but people still verify. So far, it’s working well.
Scott:
Alright, thank you Matt. Let’s move on to marketing and AI.
(Scott’s presentation on marketing, AI, and the “human-first” approach follows — continues through website optimization, social media authenticity, and AI tools like Otter.ai, Descript, VidIQ, and Opus Clip.)
Lliam:
Thanks Scott, I appreciate that.
For those who don’t know me, I’m Lliam Holmes, CEO of MIS Solutions. I’ve been in technology for about 30 years — I’m the “bits and bytes, flashing lights” guy.
What I want to talk about is transformational — how to take everything you’ve heard today and land the plane.
Here’s my core belief: the data in your systems is now more valuable than the applications that created it.
Whether it’s your ERP, CRM, or EHR, the data — the patterns, the history, the relationships — is where your real business intelligence lives.
Most companies only use about 5% of their data. We all know the drill — you ask for a report, IT says, “Give me two weeks,” or “We’ll need to pay the vendor,” or “We don’t have that report.”
AI changes that. AI lets you ask questions of your own systems in natural language — securely — and get immediate answers.
I built a prototype chatbot that connects to multiple databases (SQL, Postgres, Oracle), interprets your question in plain English, and returns data instantly — without exposing your data to the internet.
Imagine asking:
That’s the power of unlocking your data.
We’re entering a phase where Agentic AI will flip the script — instead of you asking AI questions, AI will proactively notify you about trends, anomalies, or risks: “Hey, inventory is low,” or “Customer X hasn’t ordered recently.”
To prepare for that future, you need control of your data — not locked away in someone else’s cloud.
That’s why I believe in private cloud architectures that combine the security of on-premise systems with the flexibility of the cloud — and give you ownership of your data for AI use.
In short: your data is the most valuable thing you own.
(Lliam’s section continues through discussion of data repatriation, SaaS limitations, secure AI proxying, and Q&A.)
Carol:
Alright, let’s open the floor for questions.
(Audience Q&A about data security, ChatGPT subscriptions, medical data, anonymization, and industry compliance follows.)
Carol:
I actually have a question for each of you. For someone looking to bring AI into their business, what’s the first step you recommend?
Matt:
Start where you’ll get the biggest ROI. What’s the biggest bottleneck or pain point in your organization? That’s where AI will pay off fastest.
Scott:
I’d say start from the ground up. Talk with your staff — they know where the inefficiencies are. Involve them in identifying problems before rolling out tools or policies.
Lliam:
I agree, and I’ll add: fail small, start big. Pick a small, non-critical area to experiment. Recognize that adopting AI is as much a cultural shift as a technological one. Start with small wins, build cheerleaders, and then tackle your big, hairy problems once the team’s bought in.
Carol:
When I first started using AI for content creation, I’ll admit — I felt like I was cheating. There’s a fear factor, but once you get past it, you realize how much better and faster you can be.
(Discussion continues with audience questions on verifying AI data sources, generative “hallucinations,” prompt engineering, and the balance between structural vs. situational awareness.)
Carol:
We’ve got about ten minutes left. One final question for each of you: what’s one AI use case you’re most excited to explore next?
Lliam:
For me, it’s call sentiment analysis — taking recordings from our phone system, transcribing them, and analyzing tone and satisfaction. If a call scores below a certain threshold, it triggers a workflow for review. It’s a way to catch service issues early.
Scott:
I’m fascinated by personalized learning. AI can adapt how it teaches based on how you learn. It’s going to revolutionize employee training, language learning, and skill development.
Matt:
For us, it’s lead generation automation — analyzing client interactions, service calls, and survey responses to identify cross-selling opportunities. There’s a ton of untapped value in that data.
Carol:
Awesome. Thank you all so much. Did you guys get value from today’s session?
Audience:
Yes!
Share:
