In celebration of National Hispanic Heritage Month, guest host Camille Sanders, CAE, director of chapter operations and programs at ISACA and host of LeadHERship Bytes, sits down with Carlos Cardenas, CAE, AAiP, senior strategic advisor at DelCor and co-founder of Association Latinos, for a forward-thinking discussion on the future of AI in associations. Carlos shares how his personal journey with AI began during retirement planning and evolved into a passion for helping associations—especially smaller ones—use AI strategically to thrive. The conversation explores practical concepts like the “quarterback agent” for task management, the value of experiential learning, and aligning AI tools with real business goals. Together, they highlight how associations can embrace AI innovation while ensuring inclusivity and equity for Latinx members and beyond.

Check out the video podcast here:

https://youtu.be/rw3813NLPe4

This episode is sponsored by the Atlanta Convention and Visitors Bureau.

Associations NOW Presents is produced by Association Briefings.

Transcript

Camille Sanders: [00:00:00] Welcome to this month's episode of Associations NOW Presents, an original podcast series from the American Society of Association Executives. I'm your host today, Camille Sanders. In addition to my role as director of chapter operations and programs at ISACA, I also host LeadHERship Bytes, an independent podcast highlighting the career and personal journeys of inspiring women across industries.

You can find it on any major podcast platform. Now, before we dive in, we would like to thank this episode sponsor the Atlanta Convention and Visitors Bureau Now. Let's get into the really fun stuff. I'm very excited for today's conversation where we're talking about the future of AI in associations with Carlos [00:01:00] Cardenas, the senior strategic advisor at DelCor, and a co-founder of Association Latinos.

Welcome, Carlos.

Carlos Cardenas: Hi Camille. How are you? Thanks for having me.

Camille Sanders: You are very welcome. We're excited to talk with you today. So I just want to be fully transparent with our audience that I know Carlos personally and I know about. Some of your exciting writings and the things that you've been doing in terms of educating associations around experimentation and adaptation of ai.

And I know that journey started for you with something really personal around your retirement planning, and that's a really unusual journey. That's not where most of us start. So can you talk to us a little bit about what sparked that idea for you?

Carlos Cardenas: Sure. No, I will say just like everybody else, right?

October, 2022, OpenAI dropped chat GPT version 3.0, [00:02:00] and it shook the world. It shocked the world, and so everybody started to experiment and say, how does this relate to me? How can I use this? How can I leverage it? I wrote a LinkedIn article in January talking about the directions that it could go. One of 'em was a travel advisor, one of 'em was a strategic advisor.

A couple other things as well. Fast forward a couple years, right? And so I'm, I'm relatively young, I won't say my age, but I like to think about the future. I like to think about financial independence, not necessarily retirement, but financial independence. And so I started to go down the road of what does retirement, or what does financial independence look like for me?

And you've got your traditional 401k in the workplace and you've got your employer match. Outside of that, you might do Roth IRAs, you might dabble in crypto, you might have some other investment vehicles, and so I do a lot of the work myself. And so [00:03:00] I look at websites, I look at market trends, but I'm like, how can I leverage some of these tools to help me so I don't have to do a lot of the heavy lifting?

I like to experiment. I downloaded open source models of my own. I purchased an NVIDIA graphics card. I've got that installed on my home desktop computer. But I run these models and experiment with them, and I use generative I to help me build agents. So I can have one agent that does it all, and you might get to that later on, but I felt I wanted to build an ecosystem of agents to help me with these various aspects.

So what I've been able to do so far in terms of retirement or financial independence is build an agent. I'll say it's probably 70% of the way done, but it goes to the marketplace. It looks at the s and p 500, and it comes back and it gives me that information, and then it builds a dashboard for me. So I can look at my financial portfolio and I can have it send me emails.

[00:04:00] And so basically it's my assistant to say, how am I doing? The market took a downturn, or it's doing, it's on fire, right? What does that mean? What does my five year-, what does my nine year-outlook look like? Do I need to make adjustments in my 401k? So that's kind of surface level. We could talk 60 minutes about this, but I'll stop there.

Camille Sanders: Yeah. No, I love that and I thank you for that example because I think it's a really practical example to show how people can use AI in our personal lives, right? To help us with future planning. And I, for one, had never thought about that. So thank you. Thank you again for that, and I think it leads nicely into something bigger.

That you've talked about in a recent article that you wrote and published, you talked about the fact that AI isn't at this point really about innovation, it's more about survival, especially for smaller [00:05:00] associations, and you even called it this moment, a breaking point for associations, and that's a powerful.

Really strong message and I'm curious about what makes you feel that sense of urgency right now?

Carlos Cardenas: I think it's clear if you follow investments, if you follow the big, the tech bros, so to speak, right? In terms of what they're doing. Generative AI is not going anywhere. People use it on a personal level and they've been able to multiply their cap capabilities.

But when you go to the association level for us, you look at association membership and, and I'll say for ASAE membership, since we're on this podcast, I believe something around 80% of all associations, and I don't know if they're specific to ASAE, but they're small staff associations. Their annual revenues are somewhere between, uh, I'll say a million or [00:06:00] less.

And that's a wide range. So people wear multiple hats. And so now you've got, you're bogged down into the tactical things. You don't have opportunities to take that hat off and be more strategic. And so administrative overhead comes front and center and, and I think these agent AIs or these AI systems perfect candidate to be able to offload some of those administrative burdens, so to speak, to free you up for the more strategic aspects of it.

And so that's why I feel like it is critical. It's critical now, right? People are stuck in the mud, so to speak, and maybe I see it as a consultant. Their technology posture is not where it needs to be, and so I feel like this is a perfect opportunity for leadership boards to be looking at these technologies to say, how do we leverage it in the workplace?

Again, we can use it on a personal [00:07:00] level, but. How do we bring that into the workplace and bring ourselves into the future? How do we experiment? How do we build that culture of learning?

Camille Sanders: Yeah.

Carlos Cardenas: So again, that surface level answer, but that's how I thought about this.

Camille Sanders: That's really good. And in that same piece, you introduced this concept of a quarterback agent, which I think is really timely.

It's football season, and I liked the concept because again, it makes AI feel more approachable. Can you break that down a little bit for us and talk about what exactly is a quarterback agent? And why is orchestration so much more important than just having this one catchall tool?

Carlos Cardenas: I'll start from the model perspective, and you've got your chat GPT version 5.0.

You've got from Anthropic, you've got cloud version 4.0, 4.1. You've got these multiple [00:08:00] flavors that try to do it all. If you look at the open source market, you've got specialized models, more lightweight models, maybe from an energy standpoint. They do not consume as much energy. They do not need as much computation, and so rather than thinking about one person, one agent to do it all, I like to distribute that workload and think about specialty agents, and I can have multiple specialty agents.

If I'm managing, I'm at the center of them all and managing them all. I feel like I'm just perpetuating and repeating a current problem. And so therein comes the orchestrator or quarterback agent. The, it's, think of it as a digital twin, a mirror of you that you're trying to train this particular model, and that quarterback agent can work with the other one.

So let me give you an example. Now, let's just say you've got a project, project a, we'll call it. You maybe have a [00:09:00] statement of work or project deliverables, and so you could have a specialized project manager agent that can think about all of the deliverables that need to happen. Some of the project outcomes, some of the timelines.

You might have a business analyst agent where you feed some of the brick requirements from. Think of your discovery meetings. Think of a communications agent that is specialized in outreach. Drafting emails and writing letters and drafting RFPs and writing executive summaries. And so they each have their own specialty and the quarterback agent essentially activates them all.

Doesn't have to be linear, right? It could be like non-sequential, but it can say, Hey, project manager, I need you to tap into Microsoft Project to create X, Y, Z task communications. You take that output as input. Draft a letter to the client or to the internal executive leadership team. So you can see how [00:10:00] the coordination aspect of it frees you up.

Obviously there's always a human component to it where you have to have that human oversight, but conceptually, I think as the tools and the technology evolve, so too, do organizations have to evolve to figure out how do I leverage this? And I think this is one aspect in terms of Agentic AI that we're gonna hear more and more about.

Camille Sanders: Yeah. So if an association leader had to just start with one of all the agents that you described as possible, which one do you think gives the most impact right away and why?

Carlos Cardenas: Yeah. I'll say before you even go to the agents, right? Start with yourself, make sure that you're experimenting and learning. And so I think of a series of concentric circles.

With you at the center. And then I think once you master that or get comfortable with it, you can go to that next layer. And that might be the team, eventually you'll get to the organization. But I [00:11:00] think it's important to take a step back and think about what is our organization, what is the mission, and what is the vision?

What are our business goals? Instead of chasing a bright and shiny tool, and you just leave the stuff that you just bought and you put it on the shelf. What do I, what problems am I trying to solve? Gotta make sure that you align it to the business goals. Otherwise you're just tinkering and experimenting on a personal level.

That's good. On an organizational level, I think that's detrimental to the organization. There's gotta be some experimentation, but make sure it's focused experimentation. Make sure you have some concrete examples or deliverables that you know you want your team. I think if you can do that. And the tasks you find are repetitive in nature, instead of prompting each and every time to do this, I think that brings the use case for an agent to help you.

And so again, I think start with that concept. Make use of that first agent. I can't [00:12:00] tell you which one makes the most sense at in the beginning, but I feel if once you understand that and you leverage it, and then you realize those outcomes, proof of concept, then move on to the next one. It is a process.

Yeah. You can't just jump to Step Z and say, let me get a quarterback agent without doing the other ones.

Camille Sanders: Yeah, no, that makes perfect sense. Now, one thing that. I love about your writings and your guidance on this is that you have really been adamant that associations not get caught up in the hype, and you just alluded to that.

Really tying things back to business goals and back to strategy for leaders who might be feeling FOMO, that fear of missing out. On the big push around AI right now, how would they really know where to start without just chasing the shiny new thing that's happening?

Carlos Cardenas: Yeah, I would [00:13:00] say start somewhere.

You've gotta take that first step. I think. Don't let fear prevent you from taking that first step. I can tell you to use a specific tool, but I don't think we wanna do that. I think find a tool that speaks to you. Start to experiment with it, start to learn from it, and then I think you'll get more comfortable with the tool, and then I feel like eventually you'll realize what the capabilities are.

Obviously with some training, you've gotta always be learning, because this tool constantly changes and evolves. Technology is moving at a breakthrough speed. So rather than be overwhelmed with trying to chase each and everything, try to get good at a core three concepts. Try to get good at aligning it to your specific role.

Maybe you're a marketer. Maybe you need help with social media posts. Maybe you're a tinkerer, a creator and experimenter, and you need an ideation partner. Try [00:14:00] to focus on some of those things, and then I think you'll realize what the capabilities are and then harness those capabilities. Towards the business outcomes.

Sitting on the sidelines at this point is not an option. And if you are a leader, you need to look at yourself in the mirror and say, am I stifling? Not just the innovation, but the advancement of my association, my organization? And most times you won't see it, right? If you are the problem and lean on your team, have conversations, have open and honest and frank conversations.

I feel like the more you can bounce ideas from your team and off of your team, the more that you can help shape and think about the direction that your organization needs to be going in. Ultimately, I think we all need to dip our toes into generative AI in one form or another. I know we have different on-ramps, whether you have a policy or not inside the organization, guess what?

Your [00:15:00] teams are using the tool. At work, at home, they probably know more than you do. If you're an advanced user, you probably know a little bit more, but I think sitting on the sidelines, it's just not an option. At this point. You do not wanna be left behind and three years later, finally come to the realization, oh, how do we leverage this?

You'll be behind at that point.

Camille Sanders: Yeah. Yeah. I very much agree with that. And speaking of just. Moving forward and thinking forward. You have developed this very interesting concept called Innovation by parts, which you've also trademarked and. I think it's amazing how that frames AI as something that's both incremental and manageable and for smaller associations, which you've alluded to.

And I think honestly, this might be practical for even medium and large associations that have not yet [00:16:00] leaned into adapting the technology. How can they apply the innovations by parts approach in a low risk and practical way?

Carlos Cardenas: I'll give you a little bit of background in terms of the genesis of innovation by parts.

And so I go back to calculus two. Differential calculus, and so integration by parts was the concept where you take trigonometric functions as an example, and sometimes you're multiplying them and sometimes you're integrating them and it's so complex, right? And you'll probably fill pages and pages with these equations.

But there is a tool called. Integration by parts where it lets you break that down. It lets you simplify these things into additional variables to integrate that a little bit more. That's one part of it. The other part is I'm hyper visual. If I could turn my camera, you'll probably, you'll see a whiteboard on this side here.

I've got an extensive series of whiteboards, so I'm visual. I feel like if you are trying [00:17:00] to articulate, I mean, name your concept, if it's a project, if it's a mission, if it's a vision, I can talk to you about it. The message may get warped depending on how you receive it, but if we're looking at a picture and we're looking at a diagram, then we could walk through that diagram and talk through it, and I can provide you more clarity.

Maybe I'm not gifted at the gift of gab, but I think I've been blessed with being able to articulate visually. So that's the genesis of it. Behind that is you've got this complex idea, which can be daunting to some people. Look at any project, right? Whether it's an AMS implementation, sometimes you're just like, whoa, there are too many pieces to this.

But if you can break it down into smaller pieces and analyze and evaluate, then I feel like it's less daunting. It's more approachable. You can plan around those things. You can set milestones. So I guess in simple terms with maybe [00:18:00] the long-winded answer to your question, that is innovation by parts. How can organizations use it?

Go to the whiteboard. Don't assume that people understand what you're talking about. You know, whatever that project is, if you can draw it and you can understand it, then I feel like you know what you're talking about. And if you know what you're talking about in your organization, your team would be able to see your vision as well.

And don't just draw it once and look at it once, but use it as a visual tool, visual aid. As you progress along that project timeline,

Ad Read: Let's take a quick break for a word from our episode sponsor, the Atlanta Convention and Visitors Bureau. Atlanta is a leading convention city providing an exceptional environment for a wide range of events, meetings, and trade shows.

The destination strength lies in its highly accessible and walkable convention district, which is [00:19:00] seamlessly integrated with its entertainment district, home to the world's number one airport, Hartsfield Jackson, Atlanta International Airport, and a substantial inventory of more than 113,000 hotel rooms in the metro region.

The city is well equipped to accommodate small and large scale gatherings. In Atlanta, your success is our top priority. Atlanta Convention and Visitors Bureau and our close-knit hospitality community, don't just serve as a resource, we become an extension of your team. For more information, visit discover atlanta.com.

Camille Sanders: And so now Carlos, I want to shift gears. And look a little bit down the road. If associations get this right five years from now, what do you [00:20:00] think the workplace will look like? And on a personal note, what excites you most about that future?

Carlos Cardenas: I had the privilege of being the immediate past chair of the tech council last year.

We were talking about what does the IT team of the future look like? So we had a bunch of ideas in terms of what are the skill sets, what are the psychological profiles that an IT leader, but IT team members need? And the missing piece was agents, again, as agentic AI comes into the picture, how can it not be part of the future as these tools become more available and readily accessible?

Again, it's natural language processing, and so it's very accessible. You don't have to be super techy. How can that not be part of it? And so again, from the administrative overhead aspect of it, even on the board perspective in terms of, you know, strategic planning, I still hear stories [00:21:00] where boards are expected to look at 20, 30 page documents to prepare for a board meeting.

So how can you leverage these tools to be able to evaluate, analyze, and synthesize complexity? Into more simple terms. And so the board members can focus on strategy from a volunteer perspective. Associations, you know, manage various committees. I'll say you're taking meeting minutes and I think those are the simple ones.

I think we're past that part of. But in terms of keeping people organized, people are busy, right? If they're volunteering, they've got a full-time job, they've got a family, how can we lean on some of these tools? To help us with the repetitive, to help us with the mundane, to help us prepare for our next meeting, to help us look back and reflect on previous meetings, to help us look for trends and say where are we going?

Where did we come from? Are we aligning with the three year strategic plan that we created a couple years ago? Do we need to [00:22:00] pivot? Times are changing, do a landscape assessment. So all of these different tools, and even from an internal perspective, again, if I can go back to the small staff. You're wearing multiple hats.

Maybe you don't have the budget to get a technology leader. In comes your agents, and I'm not saying they're commercial off the shelf, but you might be able to train it best practices Microsoft documentation if you're a Microsoft shop. But I feel like it can augment your staff and the capabilities and maybe you don't fall behind on technology.

Maybe you can stay ahead to some extent of that technology curve. But it's a competitive advantage, and so if you don't take advantage of it, your competitor will. And what does that mean for the future of your business, your organization?

Camille Sanders: Oh, Sage advice. Sage advice there. Thank you Carlos. And I have one more question for you.

Since it's National Hispanic Heritage Month, [00:23:00] I wanna close on what I think is a really important point in this conversation, and that's that AI systems are only as good as the data. They're trained on how can associations ensure that Latinx perspectives and experiences are included so that AI outcomes are both equitable and inclusive for everyone?

Carlos Cardenas: Great question, Camille. Thank you. As I think about Hispanic Heritage Month and how associations. Can make sure they're inclusive and think about the bias of these large language models are inclusive of their Latino members, Latinx members. Some of the things that they can do, uh, and maybe I'll take a step back and say again, bias is relative.

Bias is a reflection of humanity. If you think about public libraries, they're full of bias at one point or another throughout the history of time. But people have written their perspectives. Those [00:24:00] are, think about it. Some of the models are trained on some of that data. Maybe they've been transcribed and they're digitized now, but these large language models, you'll scrape these sites and scrape these articles, and they're trained on that.

What associations can do to make sure that Latino voices are heard are accurate. You cannot take the human out of the equation. So I give an example of let's say you want to cater a Hispanic heritage marketing campaign and you wanna make sure you have an authentic voice. You've got some choices to make.

There's Spanish, the dialect, the language itself, but there are different dialects. Are you speaking to Mexican, Spanish, Argentinian, Spaniard. So I feel like it is incumbent upon the association. To be that filter to make sure that you are speaking directly to your audience. Maybe you know the differences in terms of dialect of your members, and maybe you do some extra member segmentation.[00:25:00]

I am privileged to be the incoming board president for Association Latinos. We're a local nonprofit 501(c)(3) here in Chicago. We're about four and a half years old, and so I think about exactly what I was talking about earlier. Small staff associations. Wearing multiple hats. How do we get rid of the administrative overhead?

How do we have these tools augment? And so as I assume the role, I want to incorporate some of these tools, some of these agents, some of the training, some of the upskilling into the work that we do as an organization. I think about where we are as a small nonprofit in the, I'll say ecosystem or economy of associations in general.

We're small, we've gotta outcompete, we've gotta out hustle. And so I look at these tools as a competitive advantage. I also look at it as an opportunity to train our volunteer committees, our board, [00:26:00] so that they can learn some of these tools, not just to deliver to our community, but they can take back into their own workplaces.

So if I can make 'em a stronger leader, and if I can. Get them to leverage these tools. I know that they'll go back to their organization, to their association, to their leadership roles. They're going to incorporate some of these tools and inherently make their association a stronger association.

Camille Sanders: Wonderful. Wonderful. Thank you so much Carlos, and this has been such a really insightful conversation, and I do just wanna thank you for sharing your perspective and really giving us a roadmap. On how associations can approach AI with purpose now and in the future. And also a special thank you to our listeners and viewers for tuning into Associations NOW Presents.

Each month, we bring you the conversations that are shaping associations today, [00:27:00] highlighting the challenges, the opportunities, and the real impact. That our work has on the economy, the U.S. and even the world. And a special thank you again to our episode sponsor, the Atlanta Convention and Visitors Bureau.

To learn more about planning your next event in Atlanta, please visit discoveratlanta.com. And be sure to subscribe to associations now presents on Apple Podcasts, Spotify, or wherever you listen. And for more stories and insights, visit us anytime at associationsnow.com.