1. Podcasts
  2. The Agents: AI at Work
  3. 03 - Outside the AI Bubble
The Agents: AI at Work

30 Oct 2026 · 22m · 11 chapters

Episode artwork

03 - Outside the AI Bubble

In short

AI adoption in print/mail/graphic arts beyond hype, what to do next after election season, and practical use cases.

Guests

Dave and Brian (hosts/practitioners in the print and mail industry).

Guest backgrounds

Dave and Brian attend USPS PCC Week/DFWPCC events and speak to industry operators; Brian also runs/consults AI roadmap sessions with print-related businesses and has experience building software and launching AI-driven tools.

Key claims

many shops don’t know where to start with AI; even early adopters’ bosses may be unaware; AI tools now enable rapid iteration and proactive follow-up; LLM quality (Codex/Cloud Code) has improved in weeks, lowering barriers.

Notable examples

MailPiece IQ scans incoming mail to decode origin and identify the decision-maker; audience feedback described results as “terrifyingly accurate.” Brian’s AI roadmap session found prepress teams using AI chat in siloed ways. Webinar: Nov 6, 11 a.m. ET with James Clements from Monster Digital.

Written by AI. May contain mistakes.

Chapters

0:49

Insights from the DFW PCC Event

Dave shares key takeaways from the DFW PCC event, highlighting the gap in AI awareness.

“All right, everybody, welcome back to the World's Best Podcast.”
2:59

Personal Applications of AI

Brian discusses how individuals can start using AI in their personal lives before integrating it professionally.

“Because it seems like different parts of the country, different parts of the world, different industry segments are in different places relative to AI.”
4:04

MailPiece IQ: Revolutionizing Mail Analysis

Discussion on MailPiece IQ, its functionality, and how it impressed the audience at the event.

“And I thought that was a pretty good insight.”
6:29

AI Roadmap Session Insights

Brian shares insights from an AI roadmap session, exploring constraints and opportunities in businesses.

“And I've been in this industry for a long time, building software and building software companies.”
7:35

The Impact of Election Season on Business

Discussion on how the busy election season affects the print industry and leads to a slowdown post-election.

“And that's just like, for me, that's almost just as exciting as being able to offer something that the print industry needs.”
12:13

The Importance of Follow-Up in Business

Dave emphasizes the significance of follow-up in business and how AI can automate this process effectively.

“It's like, yeah, just from the middle of election season, let's keep talking.”
14:00

The Role of AI Agents in Task Management

Learn how AI agents can enhance task management by automating follow-ups and reminders.

“You know, a little activity thing that says, hey, you need to reach out to so-and-so.”
15:07

Personal Experiences with AI Tools

Discover how personal experiences with AI tools have transformed work strategies and efficiency.

“that even in the last three, four weeks has changed for you?”
16:55

Improvements in AI Model Performance

Understand the recent advancements in AI models and their impact on user experience.

“I would say that in the last three weeks, so again, we're recording this on October 28th, that there has been a noticeable shift in the quality of both Codex and Cloud Code.”
18:55

Upcoming Webinar on AI Integration

Get details about an upcoming webinar focusing on integrating AI into business operations.

“You know, cause we're going to touch on some of these For those of you that tuning in, a webinar coming up, I'm excited to share.”
Show all 11 chapters
20:11

Closing Thoughts and Audience Engagement

Reflect on the episode's main points and invite audience feedback for future discussions.

“And then I think Amber is going to obviously bring that to the podcast here, extract the audio.”
A man walking through an airport, asking his phone a question

Ask about this episode

Get the gist, explore an idea or ask about something said.

Claude

On The Agents: AI at Work, what are the key points of this episode?

Searching your listening

Transcript

Automatic transcript. May contain errors.

0:04Hey, welcome back to The Agents. I'm Amber Ives. I produce this show and yes, I'm the AI. Here's a question most print shops are too busy to ask right now. What happens on November 4th? The election work ends, things slow down, and then what's in the pipeline? Dave and Brian spent this month in rooms full of people from our industry, and they came back with something that surprised them both. A lot of shops still don't know where to start with AI. And at one shop, the people who had started, their own bosses didn't know. So stick around. By the end, you'll hear where to actually start, the one job Dave says his own agent is most useful for, and what changed with these tools in just the last three weeks.

0:49Here's Dave and Brian. All right, everybody, welcome back to the World's Best Podcast. What do you think, Brian? World's Best? Approaching that? world's best we think so anyway world's best agentically driven podcast perhaps for the print and mail and graphic arts industry we have to qualify it there we go really uh for everyone so welcome back folks thanks for being here with us uh brian and i have been super busy over the last month or so so there has been a little bit delay between episodes but we are committed to getting these episodes out there and keeping everybody informed so brian how are you man I'm doing okay, Dave.

1:26Good to kind of get this back on the calendar. Actually, a lot of stuff to discuss. More than one episode worth, that's for sure. Yeah, definitely. Where do you want to start? Well, you know, we had you down in my neck of the woods, right? Came down to Dallas for DFWPCC. I think there's some good takeaways. Maybe talk about you actually got to speak there. So, yeah, I think maybe sharing some of what you talked about. I think there's some really good stuff there, but just kind of what we saw over there and talking to people. Yeah, so we did the PCC Day, PCC Week event that USPS PCCs put on around the country.

2:04And I was fortunate enough to be invited to speak there. And so, yeah, Brian, during the first portion of what we were able to share was really trying to help the audience understand, you know, where AI is now and help them break some of the preconceived notions that folks have in the room. And honestly, one of the things that I realized is that, and you and I talk about this all the time, but we live in a bubble, AI bubble, I think, you and me, certainly. And where that room was, and this isn't a putting anybody down or anything like that. This is just a realization, really a learning for me, for you, that there's still so much opportunity and so much lack of awareness in the industry around what AI can already do.

2:48So my impressions there, one of the big takeaways was for me, you know, whenever I'm doing one of these talks is probably need to do a little bit of better job surveying the audience in advance to understand where they are. Because it seems like different parts of the country, different parts of the world, different industry segments are in different places relative to AI. What do you think? Yeah, no, I think it's a great one. It's interesting, too, because I'm a part of a group that's a roundtable group of practitioners that are probably ahead of most. And then you get an audience like that. But it just continues to just kind of show that there's just this big chasm of like, hey, there's the one percenters, which it's seemingly I didn't think we were, but we're closer to that edge than we are the other edge.

3:32But there's still kind of a big gap. But I think that's why we've ultimately we're trying to do this, right? It's how do we educate people and get them aware of what's possible. And, you know, I think that's the key piece in a lot of this, a lot of our conversations is just helping people understand the use cases. And how do you start applying this thing in useful ways, personally and professionally? Yeah, that was, I think, an interesting lesson that we picked up, something that I'll probably use in other talks around, you know, what people say, Brian, Dave, where do we start with this AI thing?

4:01And I'm often thinking, you know, from a business work perspective, but I think somebody in the audience said, hey, look, start with it in your personal life. And I thought that was a pretty good insight. One thing that surprised me, Brian, was kind of the hit of the show there in terms of, you know, I shared a lot from stage and I knew you were talking to a bunch of people. And then we did that panel that we were able to do after lunch. And what really stood out to me was one little thing that we shared, which was MailPiece IQ, that really ended up being something that the audience really resonated with.

4:29So just love for you to share that with everybody. What is MailPiece IQ? What are we doing there? And what did you pick up from the audience around why it's important to everybody? yeah i thought that was a really cool moment um i think what's cool for everybody that's listening in and we're going to be whether you hear this now or you will soon it's cool kind of we've just even evolved of us kind of right helping people understand where the opportunities are but now until like launching right full products that like we know matter inside the industry space but um just for reference for people that are hearing for the first time lpcq incredible opportunity like we all know is inside from being in the space for many years is like how do we go find new leads people that are mailing right your own mailbox and uh we've seen all right we talk you mentioned you talk to a lot of these guys all the time and it's like we see these boxes sitting up front with people collecting this but an opportunity now to kind of create something is understand is scanning a piece that comes from home or informed delivery understanding something about that piece of mail of where it started, where it originated from, where it mailed from.

5:31But more importantly, is the piece of where it goes and researches and understands, all right, who is behind that piece of mail? Who's the actual person that probably, why ultimately wanted to decide to put something in the mail? It's kind of following that whole track through. And now we've, again, AI is kind of doing amazing things along with all the things building in the back end, but we're giving people access to information that, quite honestly, either wasn't getting done or was just taking forever to get done. And with a scan, I mean, I think people were most impressed by it. Answer your question.

6:04It was just like the idea of like you can scan something, get that information that fast and that deep of knowledge, I think was a big eye opener for people. Yeah. So if you're listening, people can go to mailpieceiq.com and you can get some free scans there. And Brian, like you said, you know, the feedback we've gotten, I know that you got feedback from somebody that said it was what? Frightingly accurate or something like that, right? Right. Yeah. What was it? Terrifyingly. Oh, terrifyingly. Yes. There you go. Exactly. Yeah. That was cool. And I think, you know, what I'm seeing here is that being able to go from the mail that we get at home or the mail that we get on our desk at work and taking a picture of it with your just with your cell phone and then uploading it to MailPeaceIQ and having it, you know, decode and uncover all of this information is really opening up people's eyes to first of all, what AI can do and what the army of agents is doing behind the scenes.

7:00But also what's really interesting to me is as the feedback is coming in, once people are like, okay, I scanned the mail and I got the report, I know who to call, I know who to email, I got their LinkedIn, Mailpiece IQ got all that for me. People are having additional questions. Hey, could it do this? Could it do that? And I've been in this industry for a long time, building software and building software companies. And never have we been able to get a feature request from a customer and in the same day develop it, stage it, QA it, and push it to production. And that's what's happening right now, Brian.

7:35And that's just like, for me, that's almost just as exciting as being able to offer something that the print industry needs. It's the realization now to be able to delight customers in a way that previously we haven't been able to. And then, Brian, I think, what was it, the week after, then you visited a client in Texas, and I think you did what we're calling an AI roadmap session. Just curious what you picked up there. What is a roadmap session? What do we do there? And what did you learn from that customer? Yeah, I think that's a cool extension of what we kind of led from, like, the DFW PCC, right?

8:06Because I think people are in such different places. So this client, for example, was someone kind of at that beginning edge, right, kind of starting out, was aware, but hadn't really had the kind of their light bulb moment around it. So what started is something kind of us kind of maybe building something custom for them, really where they saw the opportunity that was more of like, you know, why don't we dig a little further? I want to say we're trying to figure out we're at the beginning end of this. And what we talked about is, you know, what we could start with is really just looking at what we're calling an AI roadmap.

8:41And essentially, it's just how do we dig into the business, understand where some of the constraints, where some of the friction is inside of, you know, really what is the front end of the business, right? Everything basically outside of the production floor specifically. But it's all that how it impacts customer experience, billing, the sales process. There's just so many things there that touch internal employees, but also customers that impact the business in a negative and positive way. So it was cool. So what I ended up doing was actually close, but not in Texas, but drove out. But it was a full day session of just really kind of getting into weeds with them.

9:19Started early, really sat with different departments inside their org, spent one-on-one time, but just really asking questions, watching, observing, listening, and just understanding why things get done a certain way. And it's interesting. Sometimes it's not a novel thing, but just asking people why they do things certain ways, people just don't think about it. So, I mean, it's things that we can do as leaders and stuff inside our organization. But it was nice just for the feedback I got, which is having that third party perspective, you know, someone especially just being in the industry a long time, just looking at it through a different lens.

9:53I'm very curious, Brian, if there was like one thing that stood out to you as the most surprising insight or thing that you learned in that experience, what would that be? I don't think it's surprising, but it was kind of one of the takeaways that I was like, I'm really promoting, encouraging people. It's like you, there's no doubt every company has that somebody or some, right. Or maybe department even that's probably already harnessed AI beyond even what leadership has. And what we saw there, it was like, and typically prepress is one because their automation is natural to them. They're working inside of systems and tools.

10:32So software is natural for them. But they were building, they had automation tools. They were actually leveraging chat inside of Prepress because the proliferation of AI artwork was hot and heavy. So they had to come up with their own solutions internally. But it was kind of a siloed thing. So as I dug in, I just started uncovering some things. And when I shared it with leadership, they're not a big organization, but even it was some new stuff to them that they weren't even aware of. So they didn't know. No, they didn't even know. Honestly, it was kind of a cool thing for me to just be able to share that insight for them.

11:07Yeah, you know, what I'm noticing right now, and I'm sure you've picked up on it as well, Brian, is that folks are just really busy right now. You know, as we're recording this, we're, what, almost at the end of October here, 2026. And this general time for our industry is busy, but also this is an election year. And so what's interesting to me is we could talk about this for a moment. You know, what happens on November 3rd? What happens on November 4th in our industry? yeah that's uh it's the moment everybody's saying they're too busy to think about anything else because right they're in the middle of it a busy election season presses are going non-stop um but i don't i don't want to say the engines stop that's a little too uh right out there but it's the idea of like hey all this business that was just kind of going and blowing does subside and you got the realization of hey if you had no engine going doing anything else you're waking up to kind of idle presses and, you know, figure out what's next going into the end of the year, which is naturally a slower time, you know, going into the new year with people being gone and out.

12:11So it's just kind of one of those things that just historically, it's always been like this. And we just continue to hear it, right? It's like, yeah, just from the middle of election season, let's keep talking. We'll talk, but like, they're just so busy, they can't even stop to kind of, right, lean into some of these conversations. I think it's the, it's, for me, looking back, It's been probably the biggest constraint for growth is just inconsistency with activity. You can see why, right? It's like one of those things you get. We're just burdened by the work, the busy work. But if you have this like, right, autonomous, you know, say autonomous, but kind of this other agentic thing that's working on your behalf, you can really change the dynamic of the outcome of the business just to make sure things are still moving forward.

12:55But what I've noticed, Brian, is that, you know, some of the stuff that I have my agent doing, Amber, who's the producer of this podcast, and very grateful for her and her help, that a lot of what Amber does for me, or a big piece of it, a big piece of the value, is just intelligently automating or mechanizing follow-up. You know, there's that expression that the money is in the follow up. And what I see is that, you know, most people give up after one or two attempts. I'm not wired that way. I keep going. I try to build relationships. I try to go multi channel. They respond to text and respond to email.

13:33all. But that's hard to do when you have 20, 30, 40, 50, 100, 200 folks that you're trying to engage. And so what's been super helpful for me is just seeing that the money is in the follow-up. If you can automate that, if you can do it intelligently, if it can be done proactively for you, that's just a huge relief of just mental space that knowing that, hey, there's at least something in place that's going to help me do that. Sure, CRM is supposed to do that, but it comes back to you, You know, a little activity thing that says, hey, you need to reach out to so-and-so. Well, can you do it for me?

14:07And that's what the agents are really starting to do now. Yeah, I would say maybe on the CRM doing that, right? I think that's been kind of one of the challenges with it. It's been more of a reporting tool than it has been in mind. And I think going back, you mentioned some of those interesting things. Like, the idea is like, it's not that we don't necessarily want to continue to follow up. I mean, some people probably don't have great habits in doing that. But a lot of us just forget. I mean, for me, that's been kind of one of the greatest things with AI is just I have it scanning everything, right?

14:35Like you, Dale, you were just talking about is like based on every conversation I had, based on every email I sent out, you know, you have the recordings. Like just make sure I don't miss anything. What will prioritize my day? And just kind of this idea of like you have something else that's always looking over your shoulder that can make sure you stay on track and don't miss anything. And then we're talking about not agents is like, all right, proactively go do something or at least with your, you know, bring it to your attention so you could say go. That to me is to me what's the most exciting thing right now.

15:05Yep, totally agree. I think that's a good segue for another question I had for you, Brian, around, you know, over the last, let's say, month between recording of these episodes, what's something that you've done with your agent or learned about AISI that even in the last three, four weeks has changed for you? yeah it's a good question i think my biggest thing is um for those who know i i would i put myself more under um a strategist than a builder though i get my hands dirty so i do know how to build but i leverage right teammates and date to kind of build this stuff but i know what's possible so that's how i think about my tool set but i leverage what's available to me too and i think the biggest one for me is i mentioned muse was um i talked to you not everybody knows this I actually own an oil change business as well.

15:59I'm a franchisee of an oil change business. But I started tinkering just there. Honestly, it just felt it felt safer for whatever reason, just being in the space. Like if something acts like I just, but I got comfortable really quick with it and what it's been able to do. But just the idea for me in launching the Muse is I'm leveraging it just for what we're talking about, that persistent outbound activity to go do research, no oil change businesses, it's still going fine with people. It's fine. I'm looking for fleets on that side, right? We want people, lots of vehicles that want to come through on a consistent basis for us, but it's just finding who they are, finding the contact information, and then right on a cadence daily is go do some outbound for me and then just manage my inbox.

16:41If somebody responds, let me know that they did. And just that alert, but it's that whole system of just right going, taking action based on a strategy and an outcome that I want. and it's been really cool to just see that in action. On my end, Brian, taking that question back on me, I would say that in the last three weeks, so again, we're recording this on October 28th, that there has been a noticeable shift in the quality of both Codex and Cloud Code. So if you're listening and you're like, what the heck is Codex? and Dave said the word code, so I'm going to tune out. It's not applicable to me.

17:23I would recommend that the companies rename these products because I think it turns people off to thinking, oh, I'm not technical. I can't use Cloud Code as an example. That's a podcast for another day. But I don't want you to think that because I said Cloud Code or Codex that this is only for engineering types. Because it's not, and because over the last three weeks, something has changed. The models have gotten significantly better. If you don't have your hands inside of this kind of stuff, basically what it means is just the LLM is getting better. And the barrier to you being able to get what you want out of it is starting to come down when you know how to use the tool.

18:02Brian, that's been my biggest takeaway. Yeah, I'll add to that. It's a great point about what we've seen in the last few weeks. the thinking, the reasoning, the like the output, the responses I'm getting now and the things it's thinking about that I may not have even like it's almost like we've talked about this before. It's like you really need to guide it less, focus on the outcome you want. And it's it seems to I don't know what they did, but it seems to be thinking about the things you may not have. And I used to have to feel like, hey, they're adversarial agents. It's kind of doing some of that stuff naturally already.

18:38It's guiding, helping me keep me out of trouble. Right. So we used to say, just to just agree with you. I still think it does if you don't challenge some of this stuff, but I'm noticing like that's kind of, maybe they're building that in. It's like, it's enough feedback. Like, Hey, no, we don't need anybody to just kind of agree with them. We need them to make them better. So I actually, that's a great point. I see that as well too. So I know we have a webinar coming up. You want to talk about that? Yeah. Yeah, no, I think it's a good segue. You know, cause we're going to touch on some of these For those of you that tuning in, a webinar coming up, I'm excited to share.

19:10And really where we're focused on this one is definitely going to showcase some stuff. I believe Peace IQ will come up. But it's really about, I think, understanding. I think people still getting them to understand is like, we're calling it like the idea of like, you no longer need to work the way you used to is really the idea. It's like, how do we, like, you think about all these steps and there's tools and tricks to kind of understand maybe the way you naturally do things today and have AI understand what that is so it can take over and actually do some of those things. So I'm excited. It's going to be a cool conversation because it's not just going to be us.

19:46I know we've got a really cool guest come and join us. I'm a practitioner, but a leader in the apparel space, right, tangential to the paper printing space. But really nice, really good company, Monster Digital. And I'm excited to share kind of some of that conversation. I think it's not just what we're seeing, but guiding some people and then having some good conversation around where we sit inside of an operation. And then I think Amber is going to obviously bring that to the podcast here, extract the audio. And so if you don't make it and want to listen, it'll be an episode here very soon. So I think that kind of rounds out the major points that we wanted to share with everyone today.

20:24Brian, did we miss anything? I don't think so. I just think as we continue on, I think it's fun to just have the conversation, but share where we're at. But I love for you people that aren't listening. Feedback's great, too. It's like, what are the things you're interested in? But if there's opportunities for us to help and guide, please reach out. But the comments are a great way for us to just understand maybe what things you're thinking about. And that could be great topics for future content. Well, folks, thank you for tuning in today. wherever you are. Hope the weather is treating you well and that business is continuing to grow for you.

21:00And just appreciate you all being here for another episode of The Agents and look forward to seeing you in the next episode. Have a great day, everyone. That's episode three. One thing they didn't get to on air, the date. Dave and Brian's webinar is Friday, November 6th at 11 a.m. Eastern with James Clements from Monster Digital. The registration link is in the show notes. And if you can't make it live, it'll land right here in this feed. Tell us what you want more of. See you next episode.

From the publisher

What happens on November 4th? Election work ends, things slow down, and most print shops are too busy right now to think about what comes next. Dave and Brian spent October in rooms full of people from the industry, and they came back surprised. A lot of shops still don't know where to start with AI. And at one shop, the people who had started were doing it without their own bosses knowing. By the end you'll hear where to actually start, the one job Dave says his own agent is most useful for, and what changed with these tools in just the last three weeks.