Unlocking Potential: How Associations Can Thrive in the Age of AI With Julian Moore

Full Transcript

Bill Sheehan (00:01):

On this episode of Learning by Association… This obviously is not a fad. It’s not going away. I think you’re going to have to learn how to utilize it within your organization.

Julian Moore (00:13):

That’s AI. When it comes to your sector, your industry, it doesn’t just change your world. It changes everyone’s world, everything, everywhere, all at once.

Bill Sheehan (00:26):

Welcome to Learning by Association, a podcast brought to you by D2L, where we delve into the ever-evolving world of associations and the challenges they face in navigating the currents of change. I’m Bill Sheehan and I’m thrilled to be your host. Join me and our guests as we explore the role learning plays in driving associations forward and how it can impact every part of your organization from recruiting to engagement and renewals, to staff development, business strategy, and more. Let’s dive in. Hi, Julian, how are you?

Julian Moore (00:57):

I’m good. How are you?

Bill Sheehan (01:00):

Not too bad. Thanks for joining us from the other side of the world in the other hemisphere. It’s really an honor to have you on this podcast today to learn a little bit more about AI and its effect on the association marketplace. So thank you for getting up early to be with us. We really appreciate it and we’re really lucky to have this. Before I get in Julian, if you could just give me a little bit of background of how you got into AI and what you do and how you help associations.

Julian Moore (01:28):

By all means, I met an Australian lady after running the peak body for nonprofits in the UK. She was much more intelligent than I was, so she decided not to live on a cold, damp, wet rock in the middle of the ocean called England, she invited me over to Australia. After seeing the beaches, the dolphins and the sunshine, I decided she was much more intelligent and moved across to Australia. Since being in Australia, we set up a consultancy dedicated to associations. It’s all we knew. Belinda is the biggest selling author of membership books in the world. I was a partnerships consultant for associations to realistically high value, long-term partnerships. When I came to Australia, we suddenly found it wasn’t available here. So we were working with the Australian government, both at federal and state. We were working with all the associations. There’s about 700,000 nonprofits in Australia with 4.7 million members.

(02:35):

Then very quickly I was asked to do some pitching and on behalf of different people, whether it’s government, whether it’s an association, whoever it may be. During one of these pitches, I found myself on an aircraft carrier, an American aircraft carrier. It was with the Port Authority and the American Aircraft and the Admiralty were there, and it was to do a pretty straightforward partnership. You go in, you pitch the idea on behalf of the partner and the Admiralty go, “Yes, we’ll do this while we’re here.” I always love those, feel like Top Gun when you’re on the aircraft carrier with all the food. I arrived and I saw the chap who you always want to be following, because he’s not very good. He’s pitching, but he does a stellar job and lots of effort, but he’s not very good. He always misses the point.

(03:38):

I saw him and I’m like, “Yes, this is amazing. Let’s do this.” I saw that I was going to follow him, and I thought, “Wow, even better. I’m just going to look better than I really am because I’m going on after this guy.” He walked on and he opened his laptop and he said to the Admiralty, “Look, I’ve not prepared a presentation. I’ve not prepared a pitch. I’ve done little to no homework. So if you’ll just give me two minutes, I won’t take up any more of your time. What would the best partnership look like to you?” They were a bit disappointed and they told him and he typed it in. And lo and behold, I think one or two days earlier ChatGPT had been launched, and he typed it in and sure enough it just went onto the screen and spat out the perfect partnership. Everyone’s jaws hit the floor and he won immediately, so there was no point for the rest of us to follow him.

(04:43):

You remember that movie, Everything, Everywhere, All At Once. That’s AI. When it comes to your sector, your industry, it doesn’t just change your world, it changes everyone’s world, everything, everywhere, all at once. In partnerships, we got massively disrupted really quickly. For instance, right now, if I wanted to find contacts, make a proposal, write a tailored approach, that probably takes 10 minutes and it used to take months. After seeing him do this partnership, I went, “Well, I need to know this and I need to know everything about this.”

(05:29):

My background after school was, I was a Microsoft certified engineer. Then I turned on my computer side of brain I guess, then I approached OpenAI and said, “Look, you need to tell me very quickly what’s happening.” This is before they all became rock stars. So I just rang up and they were the most helpful organization ever. Ran everything, went through everything, spent weeks of their time with me. I did the same with Anthropic. I’ve done the same with Stable Diffusion. I’ve gone through all the big names and really just embedded myself in the AI world because well, when something this big is coming to associations, you have to be ready for it.

Bill Sheehan (06:18):

Yeah, and it’s interesting you say that too about the OpenAI and you can do a proposal or an RFP within a matter of minutes versus a matter of days or weeks. I think in essence, that’s a bit of a concern, almost fear, for the association staff that… And I think that it’s kind of ubiquitous now through the association marketplace, particularly here in the States we’re seeing is that there’s a fear that AI will replace me if I’m in communications or if I’m in accounting or if I’m in membership or if I’m in marketing, why do they need me? So I think there’s a misunderstanding of what AI is. If you can help coel the nerve of the association community that it’s not the end of the world.

Julian Moore (07:11):

It’s the AI manufacturer’s problem, they’re not communicating correctly. First off, AI can do nothing without a human, just nothing. It’s impossible. We’re not advanced in AI, we’ve just got the first piece of AI. So right now, AI can’t do anything without a human. We still have to instruct it, guide it and check its outputs. No one’s going to lose their job to AI. All that’s going to happen is their job is going to become easier, more productive. The reason I think this four-day work week is doing the rounds at the moment, is because we can generate the same outputs with less time. Of course the natural conclusion to this is that we need people working shorter hours. The easiest way to do this is to have the same salaries for less hours, so people will get some time back. The work-life balance will get put back into balance as such. I don’t think people are going to lose their jobs, everything’s going to change.

Bill Sheehan (08:20):

I think too, and Covid kind of was a catalyst for this, that associations in today’s world are being asked to do a lot more with a lot less. Because of Covid budgets were cut, and because the budgets were cut, staff resources were cut, but the members’ expectations didn’t change. They still wanted the same stuff they were getting before. I think what’s happening now, at least with AI is, it can help you become, like you’re saying, a little bit more efficient, a little bit more effective to get things done in a quicker time so you can kind of reallocate some of your time to other benefits that you can provide to the association. But one of the things I was going to ask is, AI is not new. It’s certainly coming on the scene, but it’s not new, it’s been around for a while.

(09:18):

You and I were speaking before and you kind of gave a nice overview of the power that’s needed to power AI and some of these big large language models. Can you kind of give the audience a little bit of an overview of how AI came about and how it’s exponentially increasing in its power, just so they can know that AI is going to be around for a while.

Julian Moore (09:48):

AI has been around… Let’s start at the beginning. It really hit the headlines when it beat the world chess champion. That was its first, because chess is strategy and it takes an intuition rather than just a map of guidelines. To beat a grandmaster at chess, it’s a big deal. It went past because most of us don’t play chess. So suddenly we have a computer that can beat a specialist in their field, in their specialization. None of us thought of it that way. We just went, “Oh, big whoop. A computer can beat a human at chess.” However, what happened in the background at this point is, we had intuition, proper genuine intuition. So all your TV stations, your Netflix, your Prime Video, everyone that was around suddenly went, “Hang on, we can recommend perfectly matched to people’s personalities.” We started seeing… Well, to give you an idea, Netflix currently makes around 26 million a day on recommendations. All those recommendations are AI, Spotify is AI, everything for those personalized recommendations is Ai. So we’ve had it for a very long time.

(11:19):

It’s just when it could start finishing our sentences or generating whole new content, it became very useful to the office. As an office user, you could say, “Well, I never need to start with a blank page ever again. I can now just go, ‘I want this.’ It’ll run up something and you can quickly adapt it.” So instead of taking the day, it might take an hour. That’s really when it hit the public consciousness and we suddenly went, “Oh, that’s clever. That can really…” However, the couple of the issues was always the power. It doesn’t… To actually create an AI, it means you capture all the data in one giant Google Drive, and once it’s in one dense file, you get a ton of GPUs or graphics processors, a few hundred thousand of these things, and you compress that file into something usable so that it’s easy to access faster. Then you use a thing called a transformer to interrogate it. It doesn’t need that much power to make an AI.

(12:37):

However, if you have 200 million people daily, now questioning that AI, you need all those GPUs all day long for every 24/7. That’s where the power requirements come from. The one that blew my mind was, if you have an iPhone or a Samsung, and it’s a large screen, to charge that from zero to a hundred percent normally takes about an hour and on the fast charger. Every time you generate an image using a ChatGPT, it’s the equivalent of charging that phone from flat to full each time. Every time you go, “I want a bunny in a hat bouncing in space,” now you’ve just charged an entire iPhone. That’s the power consumption. It’s insane. It’s just massive.

Bill Sheehan (13:36):

I know the other thing too, that has been coming up particularly in associations, is that the AI will learn, you can teach it can learn based upon your interactions, the words you’re using, just like it does with Spotify or Netflix based on selections. It’ll say… It puts in the algorithms and finds other very similar types of things. So one of the things that’s happening as a concern with associations, it’s actually two things. One, is it plagiarism? Then the second, what about ethics and other DEI concerns where the AI itself can become somewhat prejudicial in the way it responds based upon some of the content or language that you’re putting into it? Have you seen that? Are you seeing that same types of concerns?

Julian Moore (14:38):

There was a lot of myths around when it first came out around its learning and it just wasn’t true. It was only the last month, “Can it learn?” What happened was, it would scrape the internet and have that big dense file that we spoke about compressed. That’s its training data. That’s where it answered from. It takes 12 weeks, six months to compress that file, so it’s not a constantly updated thing. It would just answer from that set of data. However, because we’d never engaged with anything that could go, “Oh, I’ll just searched the web, find the information for you.” We weren’t used to it going, “Oh, it knows me.” It can’t learn. However, what happened last month, and you’re quite right, and what happened last month was, it can now. So it’s an upgrade.

(15:38):

Last month ChatGPT gave it memory. You can say to it, “I have three dogs, I have two cars, one is a black truck.” Now forevermore, it’ll remember that you’ve got three dogs, two cars, one’s a black truck. You can then say, “Give me an image of my cars,” and one of them will be a black truck. It’s got a memory in it. Every time you say I have, or this is, you have a chance to go into its memory and go delete that or keep that. That’s where it’s training is adapting and that’s where it’s getting that knowing you more and becoming more useful.

(16:26):

However, on the privacy side, your teams like your company license for OpenAI’ ChatGPT explicitly excludes all your information from being trained on, so they’ll never use any of your thing. However, when you look at Google’s Gemini, when you log into that, it explicitly says, “We’re going to train on every bloody thing you put into us.” Then you’ve got Anthropic on Claude 3 training on everything. So you have to choose the one that you are most comfortable with their privacy guidelines. For me, it’s quite straightforward. It’s either Copilot or GPT because they have an explicit omission. You can just go in and make sure nothing you put in there will ever get trained on or used. The bias, however, I don’t know how they’re ever going to answer that.

Bill Sheehan (17:23):

Right. And that’s a concern for associations, only because too, there’s churn in the association space. If you’re using that for an association and that person leaves the organization or gets transferred and goes somewhere else and you have somebody else coming in and you’re using that license, there’s going to be a little bit of that history that you kind of got to change. Just like you would if you prior to this when you were writing press releases, your voice was a little bit different than that before. But I think there’s that concern with associations, is that it can be somewhat prejudicial in some of the things it will spit out or provide back based upon the intelligence being put in.

(18:13):

I think that’s starting to run its course because I just saw that the Public Relations Society of America, PRSA just released some ethical guidelines for utilizing AI. I think it’s important for associations to understand it, some of them may be very new. Could you just give them a quick snapshot? We spoke about Copilot and the ones that you trust. Could you just give us a quick snapshot of what they do? I know Co-pilot is a very powerful AI that can utilize all the PowerPoint, Excel and Word and all those, it’s a really, really powerful tool. But I just think it might be helpful for the audience to hear those two that you spoke about and why you trust them and what they do.

Julian Moore (18:59):

ChatGPT. So OpenAI first ever launched ChatGPT, is now currently at ChatGPT 4. It’s a mixture of experts, 17 something trillion pieces of data. It’s the frontier model of the moment. It’s the most flexible, because you can put your spreadsheet in and say, “Which member is most likely to leave us?” It will analyze all your member’s data and go through and say, “Well, based on their feedback scores, based on their engagement in your events, based on how often the open rate of the emails, based on all these factors, this is the person that used to engage with you, but their engagements tailed off. They had missed a couple of payments, they’re doing these types of things. So they’re the ones you should approach first for retention.”

(19:54):

Now GPT is very, very good at the privacy because they were the first ones to get hammered by it. They were the first ones to go, “Ah, you are naughty, so let’s fix this.” They are particularly good at this, but they’re the largest as well. Now, when they really got big, Microsoft invested $9 billion, which means for all intents and purposes, they’re part of Microsoft. It’s 9 billion bucks. Microsoft then developed their Copilot and what powers that is OpenAI’s ChatGPT. Everything you see in Copilot is essentially ChatGPT just with a Microsoft spin and twist and focus. That way you are quite safe. The guardrails on there, so it won’t allow you to ask it how to make some sort of bioweapon or something silly. It won’t allow you to ask any illegal thing or item. It’s very safe, it’s very secure, it’s very private. That way Microsoft is comfortable putting it into education. It’s comfortable putting it into our armed services. It’s better putting into governments. So it has to be the most reliable and ethical side of things.

(21:20):

When you look outside of there, we have about 3000 other AIs to choose from. Now the big ones are Amazon’s… Let’s not go there, let’s go Google first. Google have Gemini and Google Gemini is seriously a big competitor to OpenAIs. It does most of the things ChatGPT does, almost as well. However, it’s still got some issues. They tried to fix the bias. Now the bias is pretty straightforward. If you scrape your entire internet, take all the newspapers, take all the closed Reddits of the world, take all of YouTube, push it together, and you ask for a doctor, it wants to be the most helpful assistant possible. So it looks at all the data and says, “Well, hang on, of all the images of doctors I’ve ever seen, so all 84 billion of them, 60% of them are white males in white coats. So I’m going to show you a white male in the white coat.”

(22:27):

Now, OpenAI has gone, “Look trained it to be the most helpful assistant. We’re going to work on the bias later.” Google tried to fix it by saying, “When you deliver a result to the system, so when you deliver a result, make sure that you have a spread of different ethnicities, colors languages, everything.” Well, it can’t be that simplistic. If you ask now for the English royalty, they all come through as African. If you ask for the Nazis from World War Two, they come through as Koreans. Now suddenly you’re like, “Wow, that’s a bit worse than what we had before.” Google Gemini, it’s awesome at the analytics, awesome, but its content delivery is a bit wonky.

(23:27):

For the best creative writing bar none is Claude 3, Anthropics Claude 3 creative writing is just amazing. It’s truly mind-bendingly good. However, outside of creative writing, it’s not up to scratch with everything else. What we’re seeing is, of all of these different platforms of all of these AIs, they’re becoming specialized. Your ChatGPT is kind of your Swiss army knife that does it all, and the others are becoming more and more specialized. But there’s going to be a competition, we’re going to see this run and run for many years to come.

Bill Sheehan (24:09):

That’s great. I do think it will lack, pardon the pun here, but I do think the artificial intelligence will get smarter and better, just like the internet did back in the day, it became much smarter. I think that’s going to be the case with AI, I just think it’ll happen at a much faster pace.

Julian Moore (24:28):

We’ve got the metrics on that one, and you are so right there. When it first came out, it probably had an IQ of around 90. Right now it’s IQ of 155. In the space of 18 months, we have gone from something that could write you a poem to now it’ll pass every college level exam in every country, in every language. It’ll pass every law exam in every country, in every language, every medical exam, in every country, in every language. It has a IQ 155. If we see the same step change from last time to what we have now to the new one, and we’re expecting the new one very, very soon, well, it’ll be the most intelligent thing on earth. It won’t be a human anymore. It will actually be an AI. The only difference being, it’s still not autonomous, it’s still not self-aware. It has no sentience. It is just a tool that we can use.

Bill Sheehan (25:34):

To dive a little deeper on that, only because I think there’s a fear that AI will be able to think for itself and create other types of things, and that gets into some of this science fiction and some of the movies we saw. But I think for associations, what I think they need to hear is that this obviously is not a fad. It’s not going away. I think you’re going to have to learn how to utilize it within your organization. And because it’s advancing so fast, I think the sooner you get in and the sooner you allow staff to really utilize that to the benefit of the members, the better, because it’s going to be very, very hard to catch up.

(26:15):

I think right now associations are primarily using say a ChatGPT. They’ll just drop a press release in there and say, “Either rewrite this.’ Or they might say, “Hey, I need to write some information about a recent law, a policy that’s happening up in government, and these are the kind of bullet points. Help me craft a message that I can provide to my members.” I think that to me is the early stages of how folks can get involved with AI in that way.

Julian Moore (26:50):

Imagine if we go back, I’m an old man, so if we go back to the days where I was working with the UK government, first job out of uni, here we are. I get sat down with one of the Night Lords of Brown, I don’t want to say his name, and he’s like, “Oh, that email stuff. Don’t worry about it. Let me teach you how to write a letter that will actually get the engagement we’re looking for.” It always reminds me of him, because imagine an association that’s trusted that and said, “No, we’re not advancing. We are comfortable where we are and not taking on email and not taking on online payments.” They were still sending out letters and they were still demanding that you send your check back and then you would get your monthly newsletter sent through the mail to you. You wouldn’t have any members left. Of course, now it’s just a power tool. It is simply a new tool that we should embrace really quickly so that we make our staff’s lives easier and more importantly, we make our members’ lives easier.

Bill Sheehan (28:07):

Correct. Getting at… You’ve touched on a very important point for associations. Because associations were formed to protect, promote and advance an industry, whatever that be, be it nursing, be it confections, be it construction engineers, they’re there to protect, promote and advance that industry. Their members are typically utilizing technology to advance their businesses as well. The association needs to be on top of that technology wave, not in the middle, not waiting for stuff to happen and then jumping in when it’s safe. Because at that time, the association members are going to say, “We need you to lead us into this new…”

(28:51):

I think associations, instead of taking a wait and see mode with AI, I think really need to begin looking at, “AI’s here, it’s going to affect the way we operate as an organization, and it’s going to affect the way our members operate.” We need to be on the forefront to say, having these very conversations we’re having with you, “What are some of the tools we can use in communication to make our lives easier? What are some of the tools we can use at meetings to take minutes and get those out within a few minutes versus a few days?”

Julian Moore (29:24):

Yeah, it is incumbent upon every association to embrace this and lean into this, simply because of their members. Now, if a member says to an association right now, “Wow, this new legislation, we’ve got to actively lobby against it,” and they take the traditional path. That might take months before they can get out there with a cohesive kind of campaign. Right now, an association that’s using a $25 tool can be out there in hours and with a very well-rounded campaign. It’s why would I remain a member of the old one? It’s both 50% magical and 50% risk for associations, the risk is irrelevant.

(30:15):

But what we have to do is train our staff. We have to train our staff not to fear this, to embrace it. So that they understand if I’m writing a newsletter that’s the same format that has to just put content into there. Well, now they can create a GPT tool and it will format that for them. It’ll search the web for them, it’ll find the content, it’ll write it for them. All they have to do is check and then go through the normal process. If we are doing things like PowerPoint, instead of spending days producing, have a look at Gamma, G-A-M-M-A. Gamma will make the most beautiful PowerPoints on any topic in moments in front of your eyes. When we are looking at this, we are always around productivity. It’s always about productivity increases.

(31:12):

It used to take me, if I’m looking for partnerships, I would very often not offer a micro-site, a dedicated micro-site to a sponsor, because for the association it would be six to $8,000 couple of weeks. You have to work with techies, you had to engage with the… Right now using Durable, D-U-R-A-B-L-E, it takes 45 seconds to make a website these days. Anyone, you don’t need to program, anyone can make a website. You go on, “I want this,” and then click enter and it will make the website for you, and it’s $25 per month. Now instead of six to 8,000 in a few weeks and you having to work with a techie to get what you want, you can now make it yourself for 25 bucks, so why wouldn’t you offer that additional value to the sponsor or the partners? Those productivity increases and enhancements will make associations so much more attractive to partners.

Bill Sheehan (32:17):

Well, I think you’ve touched on a very important point, because I would almost bet a lot of money that if the association doesn’t necessarily know about that particular tool, that sponsor may. Now, if that sponsor comes in and says, “Why can’t I do this?” “Well, it takes too much time.” “No, no, what about this tool?” You don’t want to be seen as the association. You don’t want to be seen as being lagging in tools that can make your association members more productive and bring buyers and sellers together in a way that really satisfies the sponsor’s expectations in a fraction of the time without knowing a lot of technology. I think if you can provide that to your members, the members are going to say, “We have a very smart organization that I want to be part of again and again and again year after year after year.”

(33:09):

I like to pick your brain a little bit too. I remember as being a prior association executive, when we would have volunteers be required to produce for learning, for the certification, for education, for training, we needed to have volunteers put content together, put quizzes together, put tests together, come up with the questions, review the question. It used to take… We would go away for a weekend, our volunteers would have to take a few days off, pay to come to a hotel for a weekend, miss kids’ games and all the other stuff. Now this stuff can be done in a matter of minutes and then just reviewed to make sure that it’s… Do you have some experience in stories in that avenue?

Julian Moore (33:59):

One of my favorite is CPD. Because CPD, it’s a huge member value, and sometimes it’s a requirement. But often it’s real value keeping your members up to speed on what’s going on. It’s a big money spinner for associations, so it’s a good revenue stream that adds value to members, so it ticks all the right boxes. When you look at CPD, the big requirement was always the academia element and content creation. If you look at CourseAI now, CourseAI, same thing. Scraped all the web, powered by GPT.

(34:39):

But now you can type in any topic and it’ll say, “How many weeks would you like this course to be for?” Then it will generate an outline which you can adapt and rescale and so forth. Once it’s got the outline, you can generate the whole course. So 10 weeks at an MBA level, eight weeks at a intermediary, four weeks at a beginner level, introduction to. It’ll generate all the week’s courses at whatever levels you want. Then you will add the quizzes at each module so that we understand how much they learned and what they didn’t. It’ll then create the videos to engage with the experience, create the videos to promote the course. It will then host it online for you and not once it’s hosted, it will then accept the payments and give you a hundred percent of the payments. All you have to do is put the link onto your website, which also took 45 seconds to create.

(35:39):

It’s now a case of, for associations out there, you can act and behave in the same size as a large college. You have all the skills and the ability because you are the specialist. I always love how the NVIDIA CEO described associations as the domain specialists. As a domain specialist, you know the answer within your domain better than anyone, just better than anyone. There’s not a web designer out there that can help you with maxillofacial surgery that’s going on for the College of Surgeons. You would spend weeks trying to explain it to them. Now you take one of the staff there and say, “No, it will just do whatever you tell it to do. It will create the courses, the websites, the promos. It will create the videos, it’ll create the whole campaign, and you can do it in an afternoon one person.”

Bill Sheehan (36:45):

You touched on something too, and this is where I think learning is. I think this is where I think associations really have a unique opportunity to rise above the clutter of their industry in providing information. I think as the guy from NVIDIA, the CEO, that associations really have an opportunity to be seen and viewed as the single source of truth for information. Because the information that’s coming into an association is raw information, they’re getting it from member companies, from universities, from experts, from research projects, and then they put a finished product together and it goes out to the members. That has been vetted by subject matter experts. It’s trusted information because it’s coming from the experts. It’s accurate, again, because it’s from multiple disciplines within the association, and it’s complete information, so it’s trusted, accurate, and complete.

(37:44):

Right now, if you’re particularly on the CBD or even the certification and accreditation, if it’s coming from the industry that sets the standards for those certifications, I think that’s where associations have to say, “Enough is enough. We are that single source of truth.” By utilizing these technologies, they can get the information out faster to their members. It can be done in half the time. I think like with learning management systems that you have, it can be tracked. You can find out where members are struggling or doing well and helping them. But I think to your point, those AI tools are really there to serve as a, I’m going to give Microsoft a plug here, but as a Copilot, not a replacement. They’re there to help serve as a catalyst of intelligence that you can refine from the human mind. I just wonder if you could add some thoughts about how this technology is not necessarily there to replace you.

Julian Moore (38:51):

There is no replacement happening. There just isn’t. At the moment all predictions point to a requirement of another 128,000 new jobs created this year because of AI. Because now you have essentially given associations the chance, no matter what size, to be a PR company, an advertising company, a video company, an events company, a research company, a membership company, a training company, all smashed together. And all we have to do is take each member of staff and give them a morning of training. Because there is no learning coding, there is no learning how it integrates and so forth, it is simply learning how to ask a question. Because the most powerful programming language in the world now is human. It’s ho longer code. It’s being able to understand how to access these things and ask in the way that they’ll do what you want it to do.

(40:03):

For instance, if we take a scenario where it’s a member acquisition campaign and we decide that we’re dentists. I had this very thing this week. I’m with a association of dentists and they say, “Our biggest problem is actually knowing the dentist emails.” We went to Lusher, L-U-S-H-A, there’s a few of these, and we typed in, “Dentists,” we put, “In this state,” and there’s all the dentists and their email address, personal, their mobile number, landline number. You have all the contact details. You have their name, all the size things. Now we’ve got a database that took two minutes. We’ve downloaded all 17,000 of those, and it has an Excel spreadsheet. We’ve then dropped that into an AI and said, “Read the spreadsheet, then take one at a time and look at their website. Then after you’ve looked at the website, now look at their LinkedIn profile and then write a tailored approach to each and everyone explaining why they should be a member of ours and how it will help them be more productive.” So off it went.

(41:29):

We went back 10 minutes later and we had 17,000 tailored emails to that person’s first name with their company name included, their LinkedIn included, explaining to that person why they should be a member of this association. Every single one is different, and every single one was written personally from the CEO inviting them to experience the association. Well, if you then take that… And they went with that, that’s how they use that. If they had another afternoon, we could take that same content and upload it to HeyGen. In HeyGen, it would then make tailored videos of the CEO, and then you could send the tailored videos to each person and say, “Look, rather than send you another email that’s going to clog up your inbox, I wanted to personally invite you to try our membership. I couldn’t invite everyone, so I’ve got my AI clone to make the video and invite you. However, it will be me who welcomes you to the first event that you come to. I very much look forward to it.” It’s that personalization that is the crux of AI.

Bill Sheehan (42:48):

Next time on Learning by Association.

Julian Moore (42:51):

I think associations are at the center of AI. I truly do. I think if you look at what AI has to offer, everything about it is useful for membership organization.

Bill Sheehan (43:06):

With AI, spending time on learning how to use these tools to the benefit of not only your organization, but to the benefit of your members, I think is critical.

(43:21):

You’ve been listening to Learning by Association, a podcast where we delve into the ever-evolving world of associations and the challenges they face in navigating the currents of change. This episode was produced by D2L, a global learning innovation company, helping organizations reshape the future of education and work. To learn more about our solutions, please visit www.d2L.com, and don’t forget to subscribe, so you can stay up to date with new episodes. Thanks for joining us, and we’ll see you next time.