Leah Warwick: Hi, everyone. I’m Leah Warwick, and you’re listening to “The Admin Edge.” AI has come up quite a bit on this podcast, and AI training is core to our programming here at the American Society of Administrative Professionals, whether at our conferences or through our online courses. Our guest is Neil Malick, an expert on this subject and an ASAP trainer. So please enjoy this conversation between Neil and Katie Hendrickson at our event, EA Ignite.

Katie Hendrickson: Hi, I’m Katie Hendrickson, Executive Assistant to the President and Chief Operating Officer of AIT Worldwide Logistics, and a member of the ASAP Advisory Board. My guest today is Neil Malek, President of Knack Training, and a trainer at this event, EA Ignite. Welcome to the podcast, Neil.

Neil Malek: Thanks so much for having me. It’s been a lot of fun so far.

00:00:57

Katie Hendrickson: It has. I had the opportunity to sit in on your conversation this morning about Microsoft 365 Copilot, and as somebody who’s a new user, I learned so much.

Neil Malek: It’s a brand-new world for all of us at this point. And the technology is changing literally every second, so even the words that are coming out of my mouth will be out-of-date by the time tomorrow rolls around, you know? It’s so much fun.

Katie Hendrickson: It definitely is. Well, speaking about it being so much fun, can you tell us a little bit about your background and what made you get interested in AI?

Neil Malek: Yeah, absolutely. I come from a technology training background. I have taught Microsoft-centric classes for basically my entire adult life, and so I’m always on top of whatever it is that Microsoft is releasing at any given moment. So when Microsoft Copilot came out, I said, “Well, I guess this is the new thing that I’ve got to be on top of.” And of course, before that happened, everybody in the world got caught up with ChatGPT being announced, and all of the interesting things that seem to be possible once you get into using AI. It’s been eye-opening.

00:02:08

Katie Hendrickson: Absolutely. And I think there’s a lot of opportunities for people to jump in, but for our audience, which is all administrative professionals, what are some of the key benefits that they can expect when utilizing AI?

Neil Malek: AI as a general term — a lot of people, they saw ChatGPT and they kind of pigeon-holed what AI was. AI is a chatbot, I guess, that I can talk to. That’s not wrong. That’s absolutely something that AI is good at. But AI is just a general term for anything that a computer is trying to basically impersonate a human at doing, and so there are thousands and thousands of places where different flavors of AI could be helpful. So we all see things like, okay, you can go to ChatGPT and ask it questions, try to plan things out. For example, my wife is planning our trip to Maine using just a conversation with ChatGPT, and it’s good at retrieving interesting and useful internet information.

00:03:18

But it can also pick up on patterns that are in our everyday lives, in our calendar and things like that, and try to help us identify problems that are in those spaces, and it can also identify patterns in the way that we write messages and help us to write better messages, or maybe even impersonate us. So there are a lot of places where administrative professionals, as they’re trying to wrap their arms around this huge quantity of information that’s in the world, stay on top of it and present useful ideas to people, offloading some of that challenge to a PC just makes sense because, at the end of the day, all of our brains have some amount of limitation to them. [chuckles]

00:04:00

Katie Hendrickson: I can attest to that, definitely. There are some days you go into work and you’re just like, “I don’t even know where to start,” and it’s really helpful to have that partner with you that is, again, impersonating human thinking. So for someone that’s new to AI — maybe they are just starting to use it, maybe they’ve never used it before, [or] maybe they’ve just gotten Microsoft 365 Copilot in their organization — what are the very first steps that you recommend that they take, and what areas of their work should they focus on when utilizing AI?

Neil Malek: So I think the first steps, especially if we’re talking about those chat-based AI tools, the first thing you want to do is — there are a bunch of them. You can Google it and come up with AI chatbots, like ChatGPT and Claude and Copilot, and I would open up each of them and start to ask it questions about the world, about scenarios they find themselves in, and start to see how giving a bit of text information to one of these chatbots generates something back, right? So you could get a feel for that.

00:05:12

And then, interestingly, probably one of the easiest things to do to teach yourself how to do better is then to turn around and ask the AI, “How would you frame this prompt to get a more thorough or more useful answer?” And it will actually answer in a way that you could just copy and paste and then reuse it, and you get better at it, and you go, “Oh, okay.”

I think probably the number one thing folks can do is they can realize that in most situations AI is like this incredibly capable intern, with a brilliant brain, tons and tons of useful knowledge, but they’re an intern. They don’t know you. They don’t know your boss. They don’t know what’s going on inside the business that well. And so a lot of times when you are dealing with somebody like that, you realize, “Oh, I have to be a lot more specific. I have to give people a lot more framing so that they make better choices and give you back better results,” and it’s the same thing with these tools, right?

00:06:23

We want to learn how to interact with them and not get frustrated. It’s just like having an intern or a kid — I’m thinking about my kids, right? The first time I tell my kids to do things, they may or may not be successful at doing the thing I asked them to. A lot of times, that’s on me because they’re just a kid. They don’t know what I’m looking for if I’m not really specific and clear about that. So that’s the best thing you can do, right? Get in there, start communicating with it, and figure out what it is that will get you good results back. Because sometimes it is really about learning to communicate more effectively with the computer, interestingly enough.

00:07:07

Katie Hendrickson: Absolutely. And one of the things that I have found really beneficial for me is to actually, in my prompt, ask it to interview me one question at a time until it feels like it has enough information to give me the output that I’m asking for. That has actually, in turn, taught me a lot about critical thinking and decision-making, as I realize, “Oh, I didn’t even think of that question. Let me remember that for next time I’m doing this in the real world with a human in front of me.”

So we’ve been talking a lot about generative AI, but there’s also a lot of other versions of AI that specifically help with automation. So, from your perspective, how can administrative professionals leverage AI for these automated tasks, and which tasks or processes do you see as the most beneficial to be automated?

00:08:01

Neil Malek: So I think the best thing that you can do for yourself in this moment in time is to realize that your effort, your raw hours in the day, are going to become less and less valuable because the computer can do it faster, and the computer is not going to break a sweat. It’s not going to require coffee or anything like that. And so you need to frame what you do in terms of decision-making, framing things successfully, figuring out what a really good result would look like, and then turn to the automation/the artificial intelligence, and start to think about things that it can do that are the most rigorous and laborious things that you normally have to do.

00:09:01

I’ll give you a simple example. A couple of weeks ago, I was working with my salesperson and her sales assistant, and we had been working on trying to find different groups who might be interested in what we do. Then I turned around and I went to the Google Gemini site and I prompted it by saying I wanted this very thorough scraping of the internet for groups who fit this criteria. And it took about 20 minutes, so that’s not what you normally think of with like ChatGPT or Copilot or things like that, where it’ll respond to you in a few seconds, at most. No, this was a rigorous, intensive process, where it actually listed out all of the decisions it was making, sort of checking in with me about whether it should go a different direction, or not.

00:09:55

By the end of it, it was able to identify thousands of results that would’ve taken a sales assistant months to accumulate. That’s not to say the sales assistant — [that] her job is challenged in that, but she needs to realize, “Oh, it’s not in the actual grinding through all of these websites that the value that I am bringing to the table is going to be found,” because even if it does take 20 minutes, the thousands and thousands of results speak for themselves.

So there are a lot of these tools on the internet. There’s a website called N8N, and what you do is you created these automated agents and you say to it, okay, let’s say every morning or every time I talk to you, your job is to accomplish X, Y and Z. And so, from that, you can really prompt it with things like, “If I tell you who is going to be at the meeting, identify their LinkedIn profiles,” and things of that nature.

00:11:03

So whether it’s through a third-party system or N8N, or it’s through an internal system for the Microsoft 365 Copilot, you can then create these agents that are scheduled to go up and do the research, do that deep, grinding research that would normally — let’s be honest. If we’re doing it, we’re exhausted at the end of that day, right?

Katie Hendrickson:  Yes.

Neil Malek: We don’t want to have to do that, and so it’s a great opportunity to offload that, to see what kind of value that can generate, and then we take the efforts that the AI has brought back to us and do something really special with it.

Katie Hendrickson: Yeah. And you’re talking about a project that normally would take months has now taken 20 minutes, and it gives administrative professionals an extra step up from others. It’s like, now I have all of this free time. What can I do with it to level up, to be more impactful for my team, or my executive, for my organization?

00:12:09

Neil Malek: Absolutely. It’s interesting. I saw an article recently that pointed out that, at the C-suite level, we need to identify whether the returns — the productivity returns of AI — are actually giving you more productivity, or if your people are using it to drink more coffee, right? So it’s really important to say, “Okay, I would normally have this incredibly dense challenge that would normally take me multiple days out of a given month. Now I don’t have that challenge.” And so what I’m doing now is I’m turning that into something that has more value, and I’m showing that I’m using it for something, as opposed to being able to take too much of a breather and just sort of swallow up that productivity and just go, “Oh, good. I didn’t have to do that work.”

00:13:04

It’s something that is in the ether right now to sort of pinpoint whether it’s real productivity benefits, or it’s simply exchanging it for downtime.

Katie Hendrickson: Yeah, that’s really important to keep in mind, especially as a lot of administrative professionals are getting some pushback on utilizing AI. I know there’s a few folks that I’ve spoken to here at EA Ignite who have told me, “I work in an organization that they are 100% anti-AI.” So what are the common challenges that you see when some folks are trying to use AI, whether they have access at work or not, and how can we, as administrative professionals, overcome those challenges?

00:13:49

Neil Malek: So I do think during one of our sessions today I pointed out that really one of the big problems that people run into is they get into this mindset of: I’ve asked the AI to do something on my behalf,” and it comes back and it says, “I did it.” And if you just let that go and you stamp it and you get it out the door, all too often people are going to be able to pick up on the fact that there wasn’t a human touch anywhere in there. The specific details and things like that, you didn’t take that time to go, “Okay, now it’s time for the human layer of this operation.” That’s, I think, the biggest thing is that you want to make sure that it is this perfect blend of the computer doing what the computer is great at, and the human doing what the human is great at.

00:14:53

You’re always going to make mistakes, right? If it’s your job to grind through 10,000 things a day, some percentage of those things are going to be mistakes because you’re a human being. But on the flipside, you actually have interpersonal capabilities. You have these abilities to say, “Oh, wait a second. This person needs it this way. They have a specific flavor to the way they like to do things.” And so taking the time to combine those two things is really where the magic is.

When an organization is pushing back against AI, there are probably dozens and dozens of reasons for that. One simple one is, well, if we can’t be sure that what the AI generates isn’t sort of cribbed from somebody else’s work, then we could actually get in trouble, right? And you have to realize that. If it’s anything to do with the creative process, if it’s anything to do with something that down the road you would see as being copyrighted or trademarked or in any way really tied to your organization, you have to be careful about those things, because these AIs have been trained on basically everything that’s in the internet, and so there are a lot of copyrighted materials in those results that could be returned.

00:16:19

So you want to be careful of those things, clearly. And sometimes the business is okay with just being able to say, “Listen, we can’t trust the AI to do this.” But when it comes down to something simple, like, okay, there are calendars, there are all of these messages, and no one person can actually manage all of this information. It’s now gotten to the point where it’s beyond human capability. I think it’s sabotage to force somebody to take on that volume of work and to not say, “You know what? It doesn’t make any sense for this to be this one person’s job when the AI could handle the part that it’s good at, and then this person who’s freed up to do the most human work possible, they’re able to bring their own touch to all of this.”

00:17:11

So in those situations, clearly, there are going to be IT considerations and cybersecurity considerations and sorts of different things. So you want to make sure that there is a grounding of what that pushback is all about, [that] there’s a conversation that happens, and also that everybody realizes the first time you interacted with ChatGPT, that was the worst version of ChatGPT that you will ever see for the rest of your life. From now on, ChatGPT will be getting better and better with every passing moment. And so the idea that you might say, “Oh, AI is not good at this…” Not yet.

00:17:53

Katie Hendrickson: It’s going to get there, though. It’s going to get there. One of my taglines is: AI is a tool; it’s not a replacement. So when I have folks that I work with that just give a straight AI output, I’m like, “Hey, I can tell. That is AI. You can’t allow it to simply replace you as a human being, because we do have a special touch.” And especially as administrative professionals, we have the opportunity to have this scope of vision to an entire organization in a very unique way. We can see things from the operations side, to the sales side, to the IT side, and present all of that to our executive. It’s incredibly powerful, but we do ourselves a disservice if we try and utilize AI to replace ourselves, instead of utilizing it as a tool.

00:18:49

Neil Malek: I could not agree more. I actually find myself — the biggest use that I have for — for example, I use Claude a lot. I use Claude when I’m brainstorming classes and content and anything where I’m being creative, but, interestingly, I almost never use what Claude actually returns to me. What I’m actually doing is I’m saying, “Okay, I have this challenge in front of me. I have this scenario. I have this type of content I need to make. And if I were to just sit here in my own brain and try to imagine a perfect situation, it make take me hours. I might never succeed at it.”

But then I go to Claude and I say, “Hey, give me five things that might solve this problem.” And it generates some stuff back for me, and I go, “Oh, well, this one isn’t good, but there’s something there.” And then I go back to Claude and I say, “Take this and revise it, and come up with five different ways that this could go.” And then, again, there’s one in there that’s not perfect, but it’s in the right direction.

00:19:57

So I think what the real opportunity is, is in allowing the AI to go back and forth with you to get you into gear and to get you going down that path. To give you another example, you’re faced with a challenge with hundreds of people coming to a meeting and organizing everybody’s concerns and everything else, and for a moment you kind of rock back on your heels and go, “Wow, there’s a lot here. This is such an undertaking. I don’t even know where to start.”

You go to AI and you say, “Hey, what would you do here?” And AI does not rock back on its heels. That’s the nice thing. AI never goes, “Oh, geez, that seems like a lot. I don’t know. I don’t think it’s time.” It just goes, “Okay, I guess this is what you want,” and it just keeps going and going and going and going.

00:20:50

At some point, you go, “Ah, yeah, I’m in the zone now. I know what direction I want to go. It’s time for me to really put in the work,” and you do something really great with it. So, yeah, I think there’s plenty of opportunities where you don’t use AI for the end product almost at all, but just as something to get you rolling, to get you moving in the direction that you want to, to be as successful as possible.

Katie Hendrickson: That’s so good, and I know that’s a way that I’ve used it as well. In fact, I just used AI last week to help me decide which MBA program that I had been accepted into, to actually go into that program. I had it ask me questions one at a time and figured out, okay, based on what I’m looking for, this is the best option forward. It really helped me learn in even more ways, like, “Oh, that’s a good question. I need to remember how to do that in the future.”

00:21:46

Neil Malek: That’s fantastic. And congratulations.

Katie Hendrickson: Thank you.

Neil Malek: That’s awesome.

Katie Hendrickson: So as we’re looking forward — as you said at the beginning of our conversation, things are changing daily within these different AI programs, so how do you see the role of AI evolving in the administrative industry over the next few years?

00:22:07

Neil Malek: It’s really interesting. I think that if you look at the news, everybody, what I want you to pay attention to is what the predictions around AGI are, which is your artificial general intelligence. A lot of places that you go look will tell you that AGI is going to be out there sometime in the next five to ten years. When that happens, they’re basically saying at that point, there’s an AI that is actually a human — not actually a human, but is fully capable of looking at many dimensions to something, making many different choices, solving many different types of problems, etc.

00:22:48

Right now, for the next few years, AIs are tools. They plug in and they help us solve a very specific problem. For example, I use an AI service right now to take long videos and to identify clips out of those videos. What could be better than [when] I don’t have to sit through a 27-minute video and figure out which three 30-second clips are going to be the highlights. The AI goes through. It scrubs through it and says, “We think these are going to engage with people.” Fantastic. That is a fantastic tool. It does one job really, really well. And the same thing goes for Claude when I ask it to brainstorm with me, or a Copilot agent when I ask it questions about Microsoft Excel. It solves a problem. Over and over and over again over the next five years or so, we’re going to see exactly that: these systems that solve real problems that are more specific to a given job use case.

00:23:53

When we get to that place with AGI, that’ll be a different, interesting conversation. Then we’ll be able to say that we legitimately have an AI service that could be an employee unto itself, which is an interesting space to be in. That’s where I think you have to ask yourself the question: What is it that an AI employee would be great at? But over the next five years or so, I think finding those tools that solve the real problems for you, that’s where you want to be.

Katie Hendrickson: That’s fantastic. Thank you so much for sharing your perspective on that. We actually have a listener to wrote in, and they have a question. They wrote: “Can you talk about the benefits and quality of paid versus unpaid versions of AI software?”

00:24:44

Neil Malek: So the first level here, if you go online to services like Claude, ChatGPT, these sorts of things, sometimes the paid version will simply get you access to more of it. For instance, I go into Claude and it says, “Hey, just FYI: If you keep asking me questions, we’re going to run into a top limit here. But if you go to the paid version, there’s no top limit to the number of questions you can ask.”

Similarly, the video services — there’s a limit to how many minutes you can chop up with this AI video service. Then, additionally, it gives you access to better and better models. An AI model might be able to answer your question more effectively. And as they keep improving it, they’re going to keep the best models behind the paywall so that you’ll want to pay for that. Also, sometimes you’re paying for it to remember you. So if it can remember everything about you and your job and the way that you write and things like that, it can answer your questions better.

00:25:49

I’ll be more specific, though, here with the tool I’m most engaged with on an everyday basis, which is Microsoft 365 Copilot. In that one, it is a night and day difference because if you just have the unpaid version of Copilot, you basically have the same thing as ChatGPT, which is fine. It’s a great tool, but it’s not particularly exciting. But when you pay for Microsoft 365 Copilot, you all of a sudden have an AI service that is actually plugged into your Outlook, your Teams, your SharePoint. It’s able to answer questions about files you have access to and everything else under the sun. And so the unpaid version of Copilot is totally unexciting to me. It’s just a chatbot. You can bounce ideas off of it, but not much else. But the paid version, you can say, “Hey, what’s going on with Drew?” And it can respond with meetings and emails and files and Teams messages where you and Drew have been engaging with one another, and that is really a very interesting space.

00:27:00

So definitely practice with the unpaid products almost across the board. Sometimes you’ll realize that you want the upgraded version, with more time or a better engine behind it, or where it remembers you, but especially if you’ve got the Microsoft apps, everybody, it’s a no-brainer. You either pay for it or you just go and use all the other AI tools, because there’s nothing particularly special about the unpaid Copilot.

Katie Hendrickson: That is incredibly helpful. That you so much again for joining us, Neil. Where can our listeners find you online?

Neil Malek: So my business’ name is Knack Training (with a silent K). I rue the day that I named it that. So it’s knacktraining.com. You can find me on LinkedIn. I’m the only person with name. It’s N-E-I-L M-A-L-E-K, Neil Malek. I’m easy to find on LinkedIn, on Twitter, and all the other locations.

Katie Hendrickson: Awesome. Thank you so much.

00:27:58

Neil Malek: My pleasure. Thanks so much for having me.

[music playing]

Leah Warwick: Thank you for listening to “The Admin Edge,” produced by the American Society of Administrative Professionals, original music and audio editing by Warwick Productions, with audio and video production by 5Tool Productions. If you liked this podcast, please leave us a nice review, five stars, and subscribe. If you’d like to submit a listener question, you may do so on our website at ASAPorg.com/podcast.