Latest / What’s Up, Wake / Politely Breaking Down AI with Jen McFarland

Transcript

Politely Breaking Down AI with Jen McFarland · What’s Up, Wake

• 28 - What's Up Wake - Jen McFarland === Melissa: [00:00:00] For many of us, a motto in life is, fake it till you make it. But now with the AI revolution upon us, I'm wondering is the fake part of fake it till you make it? Taking things too far? We are fortunate to have many talented writers who contribute to our publications, and I'm sitting across from one of them today. Jen McFarland is the writer [00:01:00] of our magazine column called Geek Peak, A monthly Chat on Tech Tools and Biz for Individuals and Teams. She is the founder of Merit Digital and a contributing writer to WRAL. Tech Wire. . She's also a member of the Town of Carey, information Service Advisory Board. Her mission is to help triangle organizations plan, implement, and improve their digital solutions. In our May issue, Jen wrote about artificial intelligence and my mind was officially blown aI is technology that enables computers and machines to simulate human learning, comprehension, problem solving, decision-making, creativity and autonomy. So pretty much everything in our brains, not only is most of the information way over my head, it's also intimidating and quite frankly, a little scary. Here to break AI down for us, hopefully in very simple [00:02:00] elementary terms. Please welcome our tech guru, Jen McFarland. Hello, Jen. Hi, Melissa. Thank you for having me. Thank you. You know, honestly it was a little bit hard even coming up for the questions because I was like, do I use AI for these questions? How do I use a, I don't even know it. So these are all my questions and I think you're gonna be able to tell that. Jen McFarland: No, when Melissa: in Jen McFarland: doubt use ai. That's one of my sayings. So you, I don't even know how to use it. Yeah, my Melissa: husband was like, you're not on chat GPT, and I was like, no, I'm not. I guess I'm gonna have to look into that. And I am also gonna add that there are two smart tech people in the room right now. One is not me, the other one is Joe, our producer buttons. So yes, I call him buttons. So Joe, please feel free to speak up at any point in this discussion. Buttons: Great. Yeah, I love talking about ai. Melissa: Good, good. I can't even hardly say the word artificial intelligence. I have to say it very slowly. Okay, Jen, can you please first [00:03:00] introduce yourself and tell us how you got to be so knowledgeable about technology? Jen McFarland: Ooh. That's a big question. So I started working in it at the computer help desk at NC State when I was a, an undergrad. So I've Oh, nice. I've been at it for a long time. Go pack. Okay. Yeah, go pack. I kind of, fell into web work and in 2022, I started my business, which a little bit later that year, chat, GPT hit and it. It's not that I wasn't interested in AI before that, but certainly the way that AI was used prior to chat, GBT was very much where it was integrated in the tools where we were talking to Alexa, where we're using Google Maps, those are our traditional Google search. AI is in all of those things, but we tend not to think about it as much, and I have not moved beyond those things. So you gotta give yourself more credit. There's, you're using AI more than you think. But, but when Chat GPT came along, that sort of, it was really. Put in our faces, right? There's been a lot of articles and conversation about it. It's very, you know, built into business and, [00:04:00] and new product development and things like that. Now, there's a lot of focus on anything that's being developed working with AI or using ai. It's, it's, you know, it's the big word. It's the big name in tech right now. So, and, and it probably will be for a while. So that was something that pretty quickly after my business had started was something that. I felt was, was really important to be talking about. And to be clear, there's lots and lots of people, especially people who are small business owners who are extremely intimidated by, by AI and, and rightfully so. There's a lot to it. There's plenty to be nervous about. There's a lot of unknowns. Getting used to figuring out how to put it into your day-to-day life is it's not always obvious. So, you know, it's something that needs to be worked through. And I think especially early on, there was a tendency to believe that people could figure it out for themselves. And I think we're doing, we cannot, we're doing, I think we're doing a disservice to a lot of people to assume that, so. Mm-hmm. Yeah, this, I'm always really happy to come and talk about ai, especially for the folks that are a little bit more anxious about it. Like, let's, let's work out how it's gonna work for you. [00:05:00] And, and that's why I, you know, kind of got into this as part of this journey. So my business does sort of technical operations generally, but of course, AI does fit into that in quite a few places and more and more all the time. Melissa: One thing that I mention about you is that you are on the Town of Care Information Service advisory board. What is that? Jen McFarland: Yeah. Thank you for asking that. So the town of Cary is really great at communicating with its constituents and its citizens. And one of the ways that they do that is through their advisory boards, and they have a bunch of different advisory boards which any person who lives in Cary can apply to join. So I encourage you to go out and investigate this some more. There's a. Sustainability advisory board. There's transportation, there's a bunch of other things I'm going to get wrong if I keep listing. So, but one of them is information services and that works with, there's members from the town of Cary It who can come and talk to us in communications offices who come and talk about how they're using, you know, technology and, and different tools to help improve services for the town. And we get sort of an inside look at [00:06:00] what's being considered. We at our last meeting did a tour of the downtown Curry Park, which has a lot of technology behind it. I've actually written about that for Curry Magazine as well. So I encourage you to go back and look through that. 'cause there's some. Really fascinating ways. The town of Carey is ahead of a lot of other municipalities in this area of really making use of tech and then also making the data that comes from the technology that they're using widely available so that constituents and, and even people outside the town of Carey can use it for, for anything that they're trying to implement. Melissa: Yeah. Joe and I actually discussed the town of Cary downtown C Park and how, how they're so smart with Tech. Joe, you can. Speak up because we, you were telling me, I asked you the other day, well, how do they know how many people go to the park? And you said, oh, there's a room. And this room gets all this information about, you know, who, who walks into the park, what kind of dog they have, how old they are, how many kids they have. I mean, it kind of freaked me out a little bit. It's also very cool because I'm, I'm picturing, you know. [00:07:00] What kind of room here, Joe? Like, very cold and sterile with like one guy kind of like, yeah, picture, like wall something Buttons: from 24, you know. Oh, the show? Yeah. Yeah. I'm going straight to the Disney Melissa: animated. So what were you saying about that, Joe? I was just, I Buttons: was just remember that's what Joanna's pointed out on, on the episode when we heard her and about the park. She was mentioning how they calculate and they have cameras and all that good stuff around the park. And I also know it because, you know, I've heard from people in the. The town that, that, that technology room is pretty cool. I think I've seen a couple pictures of it on social media, Melissa: so we, we need to see the room where it happens. Like Hamilton. I wanna be in the room where it happen. Okay. I'm gonna, I'm gonna name drop a few names, of course, people I've never met before. The first is Bill Gates. Mm-hmm. He has called ai, quote free intelligence, but we all know that nothing in life is free. So talk to me about how AI has already changed our day-to-day lives, maybe even in ways that we're not [00:08:00] knowing, you know, not cog Cognizantly knowing and what life might look like even in one year from now. 'cause you mentioned that you started your business in 2022, that was not very long ago. And you, you know, that's kind of when you said the beginning of all the AI stuff really kind of snowballed. Jen McFarland: Yeah. So, yeah, I'm, I'm three years in it was three years last week. So technically again, a lot of what I do is not ai, even though I think that it's really useful and I want more businesses to lean into that, especially small businesses. 'cause I think they can get a lot of bang for their buck. There's still a lot of places where we are not using AI as as much as we could be. The places where you tend to see it the most. Are again, the places where you're not noticing it. It's very subtle. It's, you know, it's in your Spotify recommendations for your next song. It's in, you know, what, what different routes you should be using with Google Melissa: It's, or the Jen McFarland: Facebook Melissa: ads. Like if I, if I'm talking to my husband about something and the next thing I know there's an ad for [00:09:00] it on Facebook Jen McFarland: there. Yeah. There's a lot of places where, any place where it really. Data has been collected. It's, it's been working for a while that these companies that have this large amount of data is, is using a and I more, more and more to help sort through it because it, it's really difficult otherwise to get that kind of, to work through that. And that's kind of another reason why it's not something I think that. Smaller businesses tend to think about is we don't always have tons of data at our fingertips that we want to work through, right? But there, that does exist more than we realize. So I think we're gonna continue to see places where ai, using AI is going to become more and more. I wouldn't even say common. It's just gonna occur to us more. Mm-hmm. So it's gonna become a little bit more commonplace. It's gonna become the thing that you lean back on that you hadn't thought about before. I just, this past month in my column, I was writing about tech for summer, and one of the recommended recommendations I came up with [00:10:00] was. Work, you know, get your family together around a computer and ask the AI to help you come up with your own game. Come up with the mechanics, come up with a theme or a story, or, you know, if you're someone who does Dungeons and Dragons or something, come up with a d and d adventure that you can play at your table. So, work through some of the different you know, options that AI is giving you in terms of new ideas or new processes or, you know, different things like that. So yeah, I think we'll continue to see more and more AI in the tools that we're using. And then eventually more and more it will occur to us to ask AI tools to help us solve this problem or think of new ideas or you know, use, use it for different purposes. Melissa: I've heard a lot of people. Maybe even including myself that have kind of said, whoa, whoa, whoa. Is there a way that we can press the brakes on this whole AI thing so we can kind of understand it a little bit more? It sounds to me like [00:11:00] it, it's already happening there. It's kind of too late to to pause in any way. So we just kind of have to learn to accept a new normal, right. Break it to me gently. Jen McFarland: Yeah. There's definitely not there's no putting the cap back in the bag. Yeah, yeah. The models that we have have all been trained, and as you may have heard, they've oftentimes been trained on data that they really weren't supposed to be trained on. Melissa: Mm. Jen McFarland: And there's, there's not a good way there. You can't pull that information back out of the ai if, if you put some kind of personal information out there. And it was used to train, it's there. There's no, again, reaching in, finding that, that piece of code and taking it back out. Mm. You can, yeah. So, so that's gonna be something that people are gonna have to get used to with respect to. They, they're gonna wanna think twice about the information that they allow the AI to find. To begin with. Right. So, Melissa: okay. Jen McFarland: There's a, there's a few different things that you can do to make it so that AI tools hopefully won't scrape your website or won't [00:12:00] use the resources that you've put out there if they need to be private. But most of the ways of doing that are gonna involve, you know, keeping that private, either with password protection or, or some other kind of, I feel like I've lost the thread on this question. Melissa: Well, no, I mean, honestly, my, my brain, it's all jumbled up in my brain anyway, like a big ball of twine. But you are saying things like. That, that are actually putting me at ease because I don't have a website. I don't have a business online. I don't have anything that somebody can technically steal the, the data. What about your voice? Ooh, oh. If anybody wants it. You say that now, but Yeah, exactly. We don't know where this is going. Yeah. And I think that's what is scary about it. Mm-hmm. I could go in your voice. Oh. There's, there's Jen McFarland: enough of, at this point I can tell that's been recorded that they could make a bot that could use your voice. Well, Melissa: I was telling Joe last week I heard about this radio station in Australia that very popular radio [00:13:00] station and, you know, normally radio personalities like our friend Erica DeLong on G 1 0 5, she'll be like out in the community doing people meet and greets and stuff. Well, this, this lady in Australia. Nobody ever saw her. So finally they had to admit, oh, she's ai, she's not even real. So I, I can see that people might, I don't know if anybody's gonna wanna steal my voice, but it, it's, that type of thing is definitely already happening. Yep. Buttons: I just asked Chad, GBT to sum up your personality based on your Facebook profile. Melissa: Mine, yeah. Oh. What's it saying about me, Joe? I don't think I spelled your name right. Buttons: Okay. You're community focused and inviting. Mm-hmm. You're curious and engaging. You're approachable and positive. So of your show is upbeat and friendly. Okay. If you give people a good sense of insider access to their cities. Melissa: Well, I like that. Ai. I'll take it. I'll take it. I'll be friends with him. Go ahead and friend him for me on Facebook. Okay. I'm gonna name, drop another name, mark Cuban. You know, all my buddies, bill Gates, mark Cuban, mark Zuckerberg, [00:14:00] mark Cuban says, learn AI or become a dinosaur within three years. Is that true? And if so, what kind of dinosaur should I be? But seriously, how do we even go about learning a, a normal person like me? Yeah. How do we even do this? Jen McFarland: I think that there's, there's two questions to that. One is learning about it for business, which ideally if your company or your business needs you to know something about ai, they're going to support you in the process of getting that training or getting that information. Melissa: Mm-hmm. Jen McFarland: Regardless, we, we already talked a little bit about how there does tend to be, this seems to be sort of a belief that everybody can do it, they just need to play with it. And there may be some people that are already really confident tech users that are off and running and that's cool for them, but there are. Just as many if not more people who are not super confident tech users and want a little bit more support in this area. They want some confidence that they're [00:15:00] doing things right or you know, some sort of a structure to how to start to work with it. And that is all. Okay. That is all good. The other side besides business is really the personal side. And for that I would say I would you, you obviously, you can go ahead and get training for yourself as well, but I would lean more into your curiosity because that's where you're gonna have more success in playing with it. If you are. Using chat, GBT or some other whatever AI flavor you prefer to answer questions about your hobbies, to help you plan a trip that's personal to you, to help you learn a little bit more about a subject that's of interest to you, a historic whatever, biography. Those are the places where you're gonna be that much more likely to be intrigued by the tool or come up with the next question really easily. It's gonna be obvious to you. You're gonna be, you know, having a conversation with the AI and, and starting to understand. How talking to it works, you'll start to get a sense for, you know, how long are the responses? I, for me, I feel like [00:16:00] generative AI tools are really good at generating content, so I tend to feel like the responses are very long. Mm-hmm. So then maybe you do, you learn that you don't like it to give you that much. So maybe you're gonna say. Modify your prompt to say, give me a response to this in 500 words or less, you know? So you'll start to see, or give me a list of your best suggestions for this. So you're gonna start to direct it and get comfortable with asking it for the output that you really want. Melissa: So, speaking of JA Chat, GPT, my husband, like I said, was telling me that I need to be using. This is, it isn't an app, is I, I, guys, I'm telling you, please feel free to laugh at me. I don't know anything about this. And then he's, then my husband said, when I explained to him, he said, well, how do you write all these, you know, questions for your podcast and your inter introductions and everything? I said, well, I do old fashioned research and I write it. And he said, oh, you're working too hard. Jen McFarland: He first of all, I think [00:17:00] that's lovely, Melissa, and good job. I appreciate all the work that you're putting into, into this. Thank you very much, Jen. Yes. Yeah. But yeah, you're, you might be working a little too hard. Okay. You could certainly get some help as someone who you know, writes for a living. Mm-hmm. I do get a fair number of people who are like, are you just having chat GBT write your articles? Oh, yeah. And heck no. Mm-hmm. Like, there, no, no one should be doing that. Melissa: See, that's the type of thing I don't like about. About this, you know, that the fact that I could just say, Hey, chat, GPT, Mr Chat, GPT, can you write an introduction for me when I'm talking to Jen McFarland? You know, I want to be able to do the research Yeah. So that I know I'm going into this with, you know, some semblance of an idea about what I wanna talk about. Jen McFarland: Yeah. I feel the same way. You know, I'm. I got anytime I, you know, go to an event or talk to someone and I'm writing an article about them or, or about a subject that I've researched, it just does not feel like like it's me, like it's human. Like it represents the work that I wanna do if I'm having, I. And ai, write that for [00:18:00] me now. Yeah. It doesn't Melissa: take your same tone, right? Or your flavor. Yeah. Yeah. Jen McFarland: And, and there are some tools you can use to try to write it in your quote unquote tone and, and things like that. Mm-hmm. And, you know, more power to the people who wanna hone in on that. That's not where I wanna be. So, where I would use ai, where I would point to others to use AI is really as that support system, as that little robot on your shoulder, right? So, you know, for you. It. There's certainly no harm in saying, here's some questions that I have that I've come up with. Like, so do the part that you're already doing and then take the questions that you have and put them in chat, GPT and say, what do you think of these questions? Are there, do these lead you to any other questions that might be useful in this conversation? And see what it gives you. And maybe it's great, maybe it prompts another idea for you. Maybe it's not that helpful, but you know, try it again. For me again, like when I have an article or where I've worked on something, I might drop that into chat, GPT or again, different flavor flavors of AI and say. You know, [00:19:00] read through this. Are there places where my flow is awkward, where the sentence structure is awkward, where I have grammar or, you know, other errors or something is unclear and it will give you recommendations like, you know, maybe you wanna spell out this acronym, or, you know, this doesn't necessarily make sense. Or a lot of times if I have a quote in there, it will try to correct the quote in some way. And I'm like, look guys, it's a quote. Leave me alone. So there's, there's places where it's going to you know, support you. And then, you know, like I said, like with a quote, if you, if you wanna take that advice or you say No, I wanna leave that quote. So stop. Melissa: It's kind of like the red underline on Microsoft Word. It's like giving you suggestions as you go along. Yeah. In my case, I, I spell everything wrong. So it, so it just fixes it as you go. It fixes it, yeah. Mm-hmm. [00:20:00] it sounds to me like chat, GPT is like an elevated Google. I mean, am I wrong in even saying that? No, because you speak about the robot on my shoulder. That's always been Google. Jen McFarland: Yeah. Okay. So now the robot on your shoulder should be something like perplexity where never heard of it. Yeah, that's, that is a search and AI search tool specifically. Mm-hmm. So, because you're right, for a lot of us, Google has been that robot on your shoulder. Mm-hmm. Where as soon as you are working on your computer and you have a question, you just type it in. To your search bar and you get your answer right. So Perplexity is an AI tool specific specifically for search. Arguably Google is also an AI tool specifically for search. But perplexity is going to be a little bit more targeted in how it gives you the answers. It's going to cite the [00:21:00] sources in the response. So you'll have a list of all the sources at the end, but they'll also be inline numbers to say. This piece of information I got from this page, this piece of information I got from this page, then you can follow. So citing as you go. Mm-hmm. Yep. Okay. Yep. They also have an extension, so you don't even have to go to perplexity to see this. You can just click on the extension and it's gonna give you a little popup window. You can type what you're looking for and it's gonna give you the answer so you can stop opening as many browser tabs, which for me it's, it's seriously, like in the hundreds, it's a problem. So the perplexity, little popup is a big savior for that. I wanna know where this was when Melissa: I was in middle school. Yeah. Because I would've totally cheated. Don't tell my kids that all the students, it's way too easy to cheat now, right? I mean, and like Wake County public schools I guess all the school systems have some sort of, I'm gonna call it a secretive robot, that the teachers can. Type place your paper into some sort of platform and they can pull up immediately [00:22:00] how many words were plagiarized, how much percentage of the paper was plagiarized? Is that gonna be still a thing that we're gonna be able to bust our kids with cheating? Jen McFarland: So the, the problem with a lot of those tools is that they can be inaccurate, and you're gonna be a lot worse off if you accuse a kid of cheating who Melissa: didn't cheat. That's very interesting that you say that because I, I know a kid not gonna call them out. But swears they did not cheat writing a paper and their teacher said, yeah, you did. Yeah, you did. And this child is like, I promise I did not. Yeah. Plagiarize this paper. Yeah. So that's interesting that you're saying that Jen McFarland: it gets into a really weird space really quickly. So. The, there's been, I mean, this is a whole podcast by itself. Yeah. I know plenty of people in, in academia who are very concerned about this topic. And the question, you know, kind of becomes, how, how do you work, how do you teach with chat GPT? Because you can't ignore it. It's, it, students [00:23:00] are going to use it for you know, for better or worse. It makes sense that they do in a lot of ways. So, so really the question becomes how do you integrate it into. Your assignments, how do you integrate it into, you know, the what, the, what they're learning? Because they are also gonna need, you know, no question. AI is not going away. We're not putting it back in the bottle. So how are we preparing students to go into the real world knowing that AI is in their pocket? I used to have Melissa: something called encyclopedias. And there was a whole series, a whole set on my shelf, and I had to go refer to the encyclopedia. Now, then we, I eventually moved on to Google, and it sounds like Google is the OG of ai, but now I've gotta learn a whole new, A whole new thing. Yeah. To give Google Jen McFarland: credit. I mean, Google search is heavily dependent on ai. That's mm-hmm. That, and so a lot of. The chat, GPT and the generative AI functionality that we have today came from Google's work [00:24:00] to, to create the search functionality that exists. So if you go back through, they have some really fascinating papers and they came up with the transformer architecture. That was a big piece of the chat GBT functionality that that came out in 2022 and has evolved into. Chat, GPT. Mm-hmm. Which we have to be clear a tendency to say chat GPT for all models, but of course that's just open ais and just the most well known, but there's plenty of other ones. Melissa: Okay. Jen McFarland: So that's just one of many Melissa: brands. Yes. Yes. For lack of better word, many flavors. Yes. Yeah. Last name drop, I promise. Mark Zuckerberg, he has definitely been in the news a lot when it comes to ai. And he has said that AI can replace human relationships. So when we start getting into this sphere, that's when I think people get really scared. You know? I think it says more about Mark Zuckerberg's relationships, perhaps. Yeah, perhaps it does. And he also said that he's recruiting a new team at Meta to build a super intelligence. Yeah. [00:25:00] So what does all this mean? It, I mean, is he gonna create a army of robots or what are we talking about here? I cannot see into the future. Jen McFarland: If I could, I would be much wealthier and more successful Yeah. Than I am. I don't know. It, it's, I will say it's concerning to know that anyone, let alone someone with a means of Mark Zuckerberg is interested in this being a goal. Yeah. It's very scary. Yeah. There's quite a few people out there with a lot of money and a lot of interest in ai. Mm-hmm. That is enough to explain why plenty of people are worried about Skynet. So that was the AI from the Terminator series in case you need a reminder. Mm-hmm. So that, that went self-aware and destroyed the world and made the Terminator, i, I don't, new fear unlocked. There we go. Oh, I'm sorry. I just assumed that was already a fear. Melissa: Yeah, there's, oh, I think with the Terminator, like we, we just thought that was like, [00:26:00] oh, that's never gonna happen. That's just, that's so farfetched. None of this is ever gonna happen. Mm-hmm. You know, but here we are, 2025, we're having to look at at Terminator Arnold Schwarzenegger's turning into robots. So. We've entered this new dimension. Jen McFarland: Yeah, I, I mean, there's plenty to be concerned about there, especially, because like you say there, this, this is the kind of thing where it's technology that moves faster than any of the regulations around it. You know, you only need to look at social media to realize that, you know, regulations are still trying to catch up with the impacts of social media. Certainly, it's gonna take a while for that to, to take place for ai and who knows what state we'll be in by the time we get to that point. So, and the way Melissa: that government moves so slowly. Yeah. I mean, by the time ai snowballs, government's like gonna be. 10 years in the past, you know? Yes. It's gonna be too late and Jen McFarland: it will be a whole new set of problems. Yeah, yeah, yeah. So there's, there's certainly some concern around that. I think what that leads us to is that a lot of us need to be responsible for, you know, the things that [00:27:00] we choose to use and. To that end, you know, there's a limit to how much control you and I have. It's one of the things that I pay attention to is, you know, what AI tools am I using specifically? Am I paying for, where am I putting my money? And then, you know, where I do have the opportunity to sort of, you know, educate people on the different options that are out there. It's worthwhile to educate yourself on the companies behind the tools. Are they being run by? You know, a tech bro who wants to see AI take over the world? Yeah. Or are they being run by a company that wants to make this, you know, something that's open source and is more responsible? What guardrails are they putting in place for themselves since we're lacking guardrails being put in place? Yeah, at a federal level or what motivations they have. Yeah. Right. What are their motivations? Are they considering the environmental impact of the ais that they're using? Because as we know, a lot of this tech can be extremely heavy handed in terms of water and energy usage. So are they looking for solutions that are going to Im improve [00:28:00] that, if not right away then down the line. So for example, there's quite a few different companies that are trying to bring more energy online through solutions like nuclear that can be much more. Environmentally friendly. So looking, you know, you, you kinda have to do your homework a little bit more, but wherever you can make those choices to be more responsible, to use the more responsible models, hopefully that's a step in the right direction. What Melissa: advice can you give to a tech Unsavvy person like myself? Especially a tech Unsavvy mom with three kids who is a little freaked out by all the AI on the horizon. I think I've made that clear. So what do you say to the average person? I was gonna say the average Joe, but Joe is not average. Are you Joe? Buttons: Buttons. Melissa: So what is Joe Always exceptional. He is exceptional. So what would you say to someone like me? Yeah, Jen McFarland: I, I mean, I have, I have a couple kids as well. [00:29:00] For me, it's a little easier 'cause they're a little bit younger. So I'm, I am hoping that they've got the AI and academia thing figured out by the time my kids get to a point where they need to worry about that. Because, but I'm sure there will be a new challenge to live through at that point. I would say where possible, be cautious, right? Just like you would with social media or any other new technology. If you're not sure, then don't grant unlimited access. Don't put all your information in there. Once that is out there, it's out there. So it's, you know, with, with respect to privacy and security, you want to keep your information to yourself as much as possible. Where you can use it is, like I was saying, in those places where you have curiosity, where you want to try out new things, but you're not being super personal, you know, you're not putting any passwords in there. You're not, you know, you're not doing anything that's gonna endanger yourself or your children. Then try them out. Get, get comfortable with those things. You know, play around with them a little and see where they deliver [00:30:00] and where they fall short and, and share that. I, I do think, like I put in that article, it's a good idea to play with some of these things together as a family and see what comes out of it and where everybody is impressed. And then, you know, the other piece of that is just continuing to, be responsible about the tools that you're using as, as a family or, or as a as a business. Melissa: Well, speaking of your article, your May column for Carry magazine was titled Be Kind To Your AI and. The, a big part of the article is talking about how if your prompts are polite, you get a better outcome showing a connection between the behavior of AI and human social etiquette. I guess that's a long-winded question of why does it even matter how we talk to ai? I, I, the only thing I can think of that that I have used is I catch myself being very. Nice [00:31:00] to Alexa, you know, I don't talk to her. Good. Great. And it, I guess, you know, I say please and then thank you. So is that kind of what you're saying about AI as well? Jen McFarland: Yeah, and it is really not that surprising if you think about it. The generative AI models we're trained on lots and lots and lots of human content and lots and lots and lots of human content has you know, human conversations that tend to have. The kinds of words and, you know, interactions that we have, and most of our words and interactions hopefully are polite. So please, and thank you, and that comes up a lot and what we are seeing from these results, and, and there's quite a few studies on this, is essentially that when you use those terms, you are more likely to get a better outcome from the ai basically saying, you know, that if, if this is mapping to what it sees from all the data that it pulled in, then. Basically it's saying that all the human interactions, that it has tracked over time where people were more polite mm-hmm. Where they said, please, and thank [00:32:00] you, they got better outcomes, which is not really a surprise. Yeah. It's, it's, if anything, it's a great sort of reinforcement of, you know, human culture. It's, it's great that we are being polite and that that that is having the expected response, which is that, you know, people are more likely to wanna work with you and support you. So Buttons: I I ask Chad GPT, like if it gives better answers, and but I ask it to respond as Arnold Schwartzenegger, as if you were on a podcast. And this is what it says. Jen McFarland: You're gonna do this in a, in an accent, right Joe? I need to hear your Austrian accent. All Buttons: right. Ready? Okay, so it says, listen to me now, believe me later. All right. I won't read the rest of it, but you guys can imagine. Arnold. No, Melissa: please do. Buttons: He says, I don't feel your politeness. I'm not your sweet little grandma's chatbot. Okay? I don't get my feelings hurt, but, but if you are polite, I, I'm slipping back in early. You usually get better in. Usually give better instructions and say things like, please explain this like I'm five, and boom, I know exactly what you need. No guesswork, no nonsense. Just pure efficient output, like a Wellow machine. But if you scream, what is [00:33:00] DI? That was, I don't even know what that was. I mean, come on. Even I don't respond well to that. But I'm built like a tank, so yeah, don't be polite for me. Do it for you. You want better answers, then talk to the bot, like a civilized human being. Otherwise you'll get dominated. Melissa: So take it from Arnold. The, the, if you're polite Chay, BT Arnold. Yes. Yes. I don't think Arnold wants to take Arnold responsibility for that response. So yeah, I guess if you, if you. Enter something and you're just very blunt and not not polite. You're not gonna get quite as detailed of what you're looking for. Jen McFarland: Yeah. The other piece in there that, that I think I talked about in the article is if you are, positive and, and polite, but also like kind and supportive. Mm-hmm. So one of the prompts that was shown to do the best in some mathematical testing was take a deep breath and think through this step by step. So. Obviously your AI, [00:34:00] hopefully is not taking any deep breaths, right? Mm-hmm. Not yet, but, but it is, you know, that, or, or saying things like you are really good at and then, you know, giving it some context for the solution that you have. Okay. Those are things that apparently, again, like based on the training in the real world, saying those kinds of things. Is a way of supporting the, the person, or, you know, in this case, the tool that is responding to you and you're more likely to get a response that tries harder, that gives more output or better output. And you know, that's kind of shocking, but also logical based on how this came about. So it's, it's just a good sort of tool. When I, when I talk about, aI prompting. We talk about the different things that you need to include in your prompt, excuse me. And one of those things that I always include is motivation. So like, again, the pleases, the thank you. Mm-hmm. The, this is why this is important to you. It's is ultimately more likely to get some success, [00:35:00] Melissa: well, I guess including with your prompt things like, Hey, chat, GPT, you're really good at blank. That's showing the, the system. That that's the kind of response you want. Mm-hmm. Context. That's also context, what they're good at. Yes. Mm-hmm. Giving context. Yeah. Yep. And I would like for them to rename chat, GPT, because I just don't even like saying it. It's hard for me to say. It's a tongue twister, so I'd like for them to name it like Alexa and let's, make it very easy. Buttons: We'll ask Sam over there at ai, open ai. Oh yeah. He's name Melissa: dropping now. Buttons: I think also the well, we're just saying cool. First names. The, I think, Melissa: although I feel sorry for anybody named Alexa, but go ahead. Buttons: Yeah. I, do you ever get that when you're watching TV or something and somebody says the name Alexa tv. Tv, yes. And then your Alexa starts trying to answer their question. Not great. Poor everybody named Alexa before that was a thing. But I was just saying, just jokingly thinking, you know, hey, and just be polite because if the robot overlords do rise up, you know, I want to be on the nice list. Like this guy was always polite. That [00:36:00] is very true. Yeah. They're not Melissa: gonna come after buttons, right? 'cause Jen McFarland: buttons was always very nice. So, so when the AI comes and bangs on your door, you know that, you know, at least saying please has a chance of saving your life. Okay. Good to know. Hey, if that's the only Melissa: reason, then I'll take it. Yeah. So Jen. Tell us where everybody can find you and will you give me your personal cell phone number in case? I'm just kidding, but I do wanna know where everybody can find you. Jen McFarland: Yeah, sure. So, please be reading care Magazine 'cause I do, I like how you said, please, Jen. Yeah, see it's, it is already ingrained. I do the monthly article on Geek Peak. So we talk about different tech things, usually with some kind of context around Carrie. So it's nice to get a little local focus on that. I also occasionally, as Melissa mentioned, write for WRL Tech Wire, usually covering events. And then my company is Merit Digital and so we do technical operations support for small businesses and teams, and we also do training, so. We do teach classes [00:37:00] on ai, sort of, primers on getting started, developing prompts, and then we also have a new class out that's focused on working with you to, to really get AI working for your business. So again, a lot of people seem to be intimidated by it. This class is meant to sit down and look at all the different models that are out there, figure out what the right one is for your business, figure out what you need to be budgeting, things like that in order to sort of get started properly. So how Melissa: do people find you for merit? Is that M-A-R-I-T? Yes. Okay. M-A-R-I-T digital.com. Perfect. I would think a lot of businesses need, need a merit in their life. Jen McFarland: I, I do. Again, I really think that small businesses in particular can you know, find a lot of help from AI tools. So trying to get these, these tools actually to the point of really using. Get, get these businesses to the point of really using the tools and using them in the best way possible. Yeah, the most efficient way possible. The way that, you know, helps their business the most within their bottom line. [00:38:00] That's where I'm hoping to help serve. Yeah, Melissa: maybe I will plug into the, the chat. GBT. See, I can hardly even say it. Maybe I'll plug it into my, like, my introduction I used today and see how they would've changed it. And come in next time with a a chat GPT version. Nah, you're, I don't like that idea either. Jen McFarland: You're thinking of ways to use it because, because honestly, that's the first step is, is kind of training your brain. Mm-hmm. To think about what you could be doing and how to make life easier. Mm-hmm. I'm all about that though. Melissa: I was, Buttons: I was only laughing 'cause I already did that and had it rewrite your intro for so, oh no. And it's not as good as what you said. Melissa: Oh, thank you buttons. That's very kind of you. There you go. Well, thank you. Sorry, GPT if you're listening. Oh yes. You better apologize. You better apologize. Thank you both so much, Jen and Joe for enlightening us today about ai. Oh, I always happy to talk about it. I don't quite as scared. I do have a million more questions, but I'm not quite as scared, so that's great. Oh, always happy Jen McFarland: to chat. Ai. [00:39:00] Melissa: Thank you. Mm-hmm.