undefined cover
undefined cover
AI Revolution or Invasion? Charles Ferguson on the Future We Can’t Ignore cover
AI Revolution or Invasion? Charles Ferguson on the Future We Can’t Ignore cover
The Not Old - Better Show

AI Revolution or Invasion? Charles Ferguson on the Future We Can’t Ignore

AI Revolution or Invasion? Charles Ferguson on the Future We Can’t Ignore

32min |18/11/2024
Play
undefined cover
undefined cover
AI Revolution or Invasion? Charles Ferguson on the Future We Can’t Ignore cover
AI Revolution or Invasion? Charles Ferguson on the Future We Can’t Ignore cover
The Not Old - Better Show

AI Revolution or Invasion? Charles Ferguson on the Future We Can’t Ignore

AI Revolution or Invasion? Charles Ferguson on the Future We Can’t Ignore

32min |18/11/2024
Play

Description

Welcome to The Not Old – Better Show, Technology Interview Series on radio and podcast, where we explore ideas that matter to those who are 50 and older, encouraging discovery, adventure, and learning. 

Today’s episode promises to be one of our most insightful yet, as we delve into a topic shaping our world at lightning speed: artificial intelligence. 


Our guest is Charles Ferguson, the co-founder of PromptPerfect, a pioneering company in AI innovation that is making waves with its cutting-edge tools designed to enhance how we interact with AI. Charles has been at the forefront of developing AI solutions that are not only powerful but also accessible, especially for small businesses and individuals who might just be stepping into the AI landscape.


In this conversation, we’re going to cover a lot of ground—from what AI actually is and how it can be useful, to whether we should be wary of its rapid rise. Charles will offer insights on how tools like PromptPerfect work, why he believes AI is a vital tool for growth, and how you, our listeners, can embrace AI in your daily lives, whether for personal projects or small business ventures.

With AI becoming more prevalent in everything from healthcare to marketing, this episode is for everyone who’s curious, cautious, or just eager to understand where AI is headed and how it can be a force for good—especially for those who might feel left behind in the digital age. So, sit back, enjoy the conversation, and let’s learn together.


Join me in welcoming our guest today, co founder of PromptPerfect, Charles Ferguson

My thanks to all, Charles Ferguson, Sam Heningerand our wonderful audience on radio and podcast.  Be well, be safe and let’s talk about better.  The Not Old Better Show.


Hosted by Ausha. See ausha.co/privacy-policy for more information.

Transcription

  • Speaker #0

    Welcome to the Not Old Better Show, the show covering all things health, wellness, culture, and more. The show for all of us who aren't old, we're better. Each week, we'll interview superstars, experts, and ordinary people doing extraordinary things, all related to this wonderful experience of getting better, not older. Now, here's your host, the award-winning Paul Vogelzang. Welcome to the Not Old Better Show Technology. interview series on radio and podcast where we explore ideas that matter to those who are 50 and older, encouraging discovery, adventure, and learning. I'm Paul Vogelsang, and today's episode promises to be one of our most insightful yet as we delve into a topic shaping our world. It is everywhere right now, and it's doing this at lightning speed. I'm talking, of course, of artificial intelligence, AI. Our guest... is Charles Ferguson, the co-founder of Prompt Perfect, a pioneering company in AI innovation that is making waves with its cutting-edge tools designed to enhance how we interact with AI, not just for businesses, but for individuals, for small businesses, really for everyone. And you're going to hear Charles talk about what it's been like to be at the forefront of developing AI solutions that are not only powerful, but also accessible to all of us, especially for small businesses and individuals who might just be stepping into the AI landscape. In this conversation with Charles Ferguson, we are going to cover a lot of ground from what AI actually is, how it can be useful, to whether we should be wary of its rapid rise. Charles Ferguson will offer some insights today on how tools like PromptPerfect work, why he believes AI is a vital tool for growth as a society and how you, our listeners, can embrace AI in your daily lives, whether for personal projects or small business ventures. You're going to love this interview with AI becoming more prevalent in everything from healthcare to marketing. This episode is for everyone who's curious, cautious, or just eager to understand where AI is headed and how it can be a force for good, especially for those who might feel left behind in the digital age. So sit back, enjoy our conversation with Charles Ferguson. Let's learn together. Join me in welcoming our guest today, co-founder of Prompt Perfect, Charles Ferguson. Charles Ferguson, welcome to the program.

  • Speaker #1

    Thanks, Paul. I appreciate it. We're very, very appreciative of the opportunity to chat.

  • Speaker #0

    Well, thank you. I am excited about that too. I think first and foremost, happy Halloween. We're talking on Halloween. And so I want to make sure to extend those wishes. My wife is going to come home after being away for a few days, so I don't think we're going to have any trick-or-treaters. How about you guys? Are you planning for some?

  • Speaker #1

    No, I think of this more of the eve of my girlfriend's birthday, more so than Halloween, to be candid. But there's certainly a lot of people walking around in costume outside of New York City right now.

  • Speaker #0

    Yep, yep. Well, again, welcome and congratulations on all you are doing with Prompt Perfect. We're going to talk about that and AI, the subject of artificial intelligence today. I think that's on everybody's minds. And I want to get into that with you. But again, first and foremost, congrats on this wonderful tool. I use it. I'm going to talk about it a little bit personally, but I love what you guys are doing.

  • Speaker #1

    Yes, we've been very, very fortunate with the timing. It was sort of built out of necessity for ourselves, as we'll go into, I'm sure. And the reception has been incredible. It's just so much fun to be able to work with people all over the world, over 130 countries represented in our customer base of hundreds of thousands of people. It's been used over a million and a half times. And so we're glad that anybody finds it useful, but we also use it daily. And it's been a long 18 months of a lot of iteration and it's gotten a lot better. And we're excited also to be launching a new product this week, very similar. It's omnipresent. It isn't confined to ChatGPT. It's within your web browser, within Google Chrome.

  • Speaker #0

    I'm a user. I love the tool. It has been so helpful for me. This is purely an editorial interview. I can talk and wax. poetically on this service that you offer and this fantastic means for me to do work as a small business person. But I think it's important to point out that I think our audience is going to be starting a few steps back in terms of their understanding of what artificial intelligence represents. And so I wonder if you could tell us about the inspiration for starting Prompt Perfect. I know a little bit about the story, but it is this amazing tool. Did you start to see some gaps in some of the development that was going on around artificial intelligence?

  • Speaker #1

    Yes, I need to make sure first and foremost, it's extremely clear that my co-founder is the ideator here, my closest friends from college. We lived together for a long time and we spent about 10 years working on other things completely separately, but always remain really, really close. When ChatGPT 3 and then 3.5 had come out and there was a front end introduced and that was in 2023. We were playing with the tool a little bit, but it was really my co-founder who was trying to assist himself with this new tool to make his job easier. He's a copywriter at the time, and he found a lot of inspiration with providing examples of copy that he loved, and then getting iterations of things that were more applicable to him based on context he was providing. And what he was really teaching himself is prompt engineering before reading any papers on it. So just understanding the correlation between the input you provide it as being directly correlated to the quality of the output you receive back, which is very different than Google search or internet browsing. They're very, very different. And so the skill set is different. And that's something that he just intuited himself almost two years ago now. And he built it out of hackathon over a weekend. He had been given that opportunity. And then subsequently, ChatGPT introduced this beta feature plugins, and it was featured as one of the top eight plugins. And that was before we really had good analytics or any sort of understanding of where this might go. And overnight, within three weeks, there were thousands and thousands of users. And it was very interesting to see. We just sort of took it as it came. And he had asked me to be his co-founder around that time. And we had a lot of bookkeeping to do. And that's neither of our core competencies, but I figured it out. And we've come a long way in that time. It really came from a necessity of trying to figure out how do we use these tools effectively and realizing this is not the same as our past 15, 20 years of experience working and searching on the internet. This is a very new skill set. It's all about understanding the context of the question you're trying to ask. providing as much context and detail as possible. And that's going to provide better responses always. It really came from us trying to get the most out of the platform, getting the most out of ChatGPT at the time, without having to spend too much time writing out the details. And so what Prompt Perfect does is it takes the sentence or the prompt that you provided, and it enhances it with more context and more detail. And it shows you that more detailed prompt and then it'll show you the response that that more detailed and that more context provides to you. So it enhances prompts for people. We assumed people cared about how to learn how to prompt engineer, but what we've learned from launching eight products is they're much more interested in having that activity done for them more so than learning how to do it. I think it's a little bit easier if it just sort of does it for you. And maybe it'll show you its work, but ultimately teaching somebody to prompt engineer isn't nearly as much fun or what we found is useful economically for people. They just want it done for themselves. So that's what we do is we just provide a service that enhances your prompts for you automatically.

  • Speaker #0

    Brilliant. Again, congratulations to you and your co-founder, because I'm always amazed at the development of businesses and how they really get started. And I want to drill down for just a moment and make sure that our audience understands that Prompt Perfect serves as a tool, as you describe it, is really a companion to ChatGPT. And in order to really maximize the sheer efforts that ChatGPT can bring. prompt perfect allows you to enter a string of words that are related but you don't necessarily have to share the sheer volume of material that you might within chat gpt because the technology within prompt perfect supplements your own work to give you this tool that really generates this fantastic result that you're getting from the artificial intelligence yes

  • Speaker #1

    without a doubt i think if you're using prompt perfect but you're very well versed in prompt engineering, you might not see that significant of an enhancement in your prompt. So if you provide it with five paragraphs and 300 words, and it's extremely detailed, we're going to be able to automatically enhance that maybe slightly or change some syntax. And we really try to optimize for the model. And so we're trying to think about how the model is going to receive that. So we're iterating on that. So it could see some improvements. But the time when we see the most significant improvement is in two, three, four, five lines, stream of consciousness. with misspellings, whatever it might be, just people being people. And what we are going to be able to do there is enhance that and show you how we've enhanced it, most importantly to me, so that you can then ultimately employ that strategy yourself. But the idea of the time savings of not having to figure out how to word these things or provide that context, we've seen that our customers find it very, very valuable.

  • Speaker #0

    Yeah. And as I say, I'm a customer. I'm a small businessman. I am just a tiny, tiny little business. It's really me. And I've since expanded. my business a little bit so that I work with an editor. I'm working with another individual who helps kind of keep me on track in terms of schedule. I really look at Prompt Perfect as another level of staffing to my business. It helps me that much. It helps me refine questions. It helps me with writing passages for the program. It helps me with social media posts. It really does so much to supplement me. Yeah. And so selfishly, I just think this tool is all about empowering people. Was that kind of your thinking in terms of empowering individuals and small businesses? Because I will tell you again, that's what it's doing for me.

  • Speaker #1

    That's fabulous. We love to hear that. I feel I'm gushing. I'm not sure how to explain it more than that. But Eric and myself, we both immediately following undergrad, we graduated in 2013. I'd done the Peace Corps for a couple of years. He was in AmeriCorps as a teacher. And so we're both very service oriented. Well, I don't profess to say that Eric had thought of this because he was doing it out of a greater good. It really was a self necessity. And we're a two person business as well. And we use it constantly. And we have a lot of feedback from that. We've spoken to thousands. High hundreds at the very least. I haven't aggregated all of the numbers in terms of the user feedback and the way that we correspond with folks, but I wouldn't be terribly surprised if it's over a thousand after 18 months. The feedback is resoundingly positive. We did, actually, for four months very concertedly double-click into trying to work with small businesses specifically. I had hundreds of conversations with different small business owners, mostly small businesses, and then we ultimately built some bespoke products for several of those small businesses. It's a common thread. AI is something that we internally call them interns. They're not somebody you're going to hand a task over to entirely. It's not going to get it over the finish line necessarily or create hundreds of blog posts for you without any sort of thought. There should be, and I think there will be for the foreseeable future, a human in the loop, so to speak. And also just for quality control and making sure that it's in your voice and your tone and things are done accurately. But it is a massive time saver for all sorts of small businesses. We're very fortunate. And we've also, one of the things is working with small businesses was a surprise, but also we've been very surprised, or at least I was, I should say, by the international adoption. So I'd said earlier, 130 some odd countries, 133, 134 countries. So much of what people are using Prompt Perfect for is English as a second language, but not necessarily as a tutor as much as just making their own verbiage more clear, a little bit more concise, different things like that. We find a lot of feedback from our customers around that as well. So helping them level up their own writing in English, which is really, really great. And then also in Greek, we worked with a Greek marketer who we figured out how to make the Greek as good as the English, depending on the time at which you ask it to translate into Greek. So corresponding in English and then asking it to translate into Greek at the end of the conversation, so on and so forth. There are these little hacks that you kind of have to work through together. But I think that language and natural language processing is unbelievably exciting. We've been very, very lucky to be at the intersection of that.

  • Speaker #0

    I will say anything that is done to improve communication this day and age is, I think, a boon and a bonus for all of us. So hats off to you for all of that work, too. Well, give us a very basic level of understanding about artificial intelligence. We hear the term an awful lot. It's bandied about in some very positive ways, and we'll talk about that. It's bandied about in terms of its potential for harm, perhaps. shifting the economy one direction or another. For our audience of those who may not be as current on some of this technology, give us that fundamental understanding of what artificial intelligence really is all about.

  • Speaker #1

    Sure, of course, AI or artificial intelligence as a technology, it's not brand new with the advent of chat GPT, things like recommendations on Netflix or recommendations on Spotify or on YouTube, those algorithms. are a form of artificial intelligence, even in Microsoft Word or Google Sheets or Grammarly, which sort of lives above all of those things. That ability for it to distill grammatical errors and spelling errors, those are more the grammatical side, but Grammarly has been around probably not since I was in college, but for some time now, I know that it's very, very popular. But in terms of the reason that people get, at least myself, get concerned is this term or this notion of artificial general intelligence, which is a completely autonomous agent that can think on its own and has the ability to think on its own. to make its own decisions and enact in its own best interest, which makes me immediately think of The Matrix, one of my favorite movies of all time, just farming people for their energy. Obviously, those dystopian ideals will sell movie tickets, but I think AGI is still very far away from where we are now. Of course, we do have AI that has for years beaten humans at chess or Go or any of these complex games. But what you'll notice is that those AI instantiations, one of them that can beat you in chess, they don't use that same model to beat you in Go. what they have to do is they start from square one again. And that model is unable to distill new knowledge or to deduce how to play a new game. It's not using cognition in the way that humans do. Even with my cousins, I can teach them how to play a card game because they have an understanding of how a deck of cards works. And that's not the way that these models really work. So it's all about the training data. And so we do see a lot of excitement over, let's just give it more data. That's the answer to this. We need to give it more data. And people are really focused on large language models. There's a lot more out there. There are a lot more ideas than just large language models. And I think that AGI as an idea, which I think personally is very exciting, although I do love science fiction and do find myself reading as much science fiction as I can make time for. I do think that that is a very far away concern personally. In terms of what artificial intelligence is, it's really very simple. It's just the ability for a machine. to think in a way that we recognize as human. So if you think about it like that, there are all these instances, whether it's navigations and using ways on your phone telling you where to turn as an instance of that. And then obviously, there are these Alexa and Siri, which we will leave a little bit to be desired, but are certainly an instantiation of that AI. But I think the reason that people are so excited about it is because of ChatGPT and tools like that, which are chatbots empowered by AI. So I would call them AI chatbots. But it's a very complex and complicated question. I think the chatbots are really what people think of with AI now, though. And I don't think that YouTube recommendations or Netflix recommendations, even though my mom uses them, I don't think that she would really think about that as AI. Although it technically is, it certainly is.

  • Speaker #0

    Our guest today is Charles Ferguson. Charles Ferguson co-founded Prompt Perfect. We'll have links in our show notes today for more information about Prompt Perfect, more information about Charles Ferguson and the work that the firm is doing with Prompt Perfect. I know you're introducing some... new products. I saw, for example, today that my prompt perfect was updated just recently. You guys are very busy doing all of this work. I love one of your quotes that you say, the internet is the most important invention in recent history because of the proliferation of information it enables. Should we be afraid of some of this information as it gets aggregated? There's almost a natural fear of some of the unknown, but knowing this, I don't know that fear is the right way to proceed. What's your reaction?

  • Speaker #1

    It's a great question. I do think that the democratization of access to information is an absolute net positive, and it'd be very difficult for me to concede that somehow. But I do think that there is very reasonable concern, and it's important to be concerned, and it's important to be curious enough to ask questions about these things and to be critical. I absolutely do not think of it as a utopia where everything is going to be perfect. The internet is absolutely... an open place. And that's why when you go and you just type into ChatGPT, which is trained on the internet, and the internet is inherently average, but just by virtue of the fact that there are such a large group of people contributing to it. And just in general, that's sort of the definition of what average is. So what you're going to do is when you type in a simple query, you're going to get a relatively simple response or an average response. Whereas if you tell it or prompt it in such a way and let it know, hey, you're an expert, you're a senior level programmer, keep that in mind in your response. The response very significantly becomes higher quality very quickly. So I think that when it comes to the concern around AI and that information, that misinformation, I think it's very, very reasonable. But I tend to have more of a growth mindset than a scarcity mindset, which sometimes gets me in a little bit of trouble. But I think back to my co-founder is very, very talented and his girlfriend as well at Improv and they're always saying yes and as a sort of a mantra. And I think that the ability to sort of pick things up and be intellectually curious about them and to not think of them as necessarily inherently negative. I think that the internet was founded by a group of people, which is a very positive thing. It's the centralization of ideation that I think is a really, really scary ideal. And I think that the internet is sort of the antithesis of that. And I think that AI has the potential to be an extension of that. But the internet is certainly not being replaced by AI. I think it's just being augmented by it, which is, I think, a positive thing.

  • Speaker #0

    Yeah, I tend to look at this very positively, too. I wonder if you'll tell us a little bit about what's coming out of Prompt Perfect, the group. And you touched on the new product and how it's going to be browser-based, correct me, but tell us a little bit about what's coming from Prompt Perfect.

  • Speaker #1

    So it's been 14 months of being live within ChatGPT. And so what we've done is we have sort of moved with them and ridden that whale. We're sort of one of those small fish under the large whale. And we've been sort of adapting as they adapt. And they went from plugins to GPTs. And so we graduated into a GPT, which then became a feature that is available to everybody. So everybody can use Prompt Perfect within the ChatGPT marketplace now, which is fabulous. We saw a huge uptick in usage as a result of that. But ultimately, not everybody uses ChatGPT. So we've really narrowed ourselves to being just in that one marketplace. And really, just to take a step back, what Prompt Perfect does is it just perfects prompts, it just enhances prompts. And what we've done is we've taken a lot of conversations with customers in the last year, And we've built a new product that lives within your browser instead of within ChatGPT. So anybody who uses Google Chrome might be familiar with Chrome extensions, whether it's an antivirus software or something like that. What now PromptPerfect does is it sits as a layer on top of all of these different chatbots. So whether it's Gemini from Google, or Anthropix, LLM POE, or any one of these different examples, I'm sorry, Claude, but nonetheless, any of these chatbots now PromptPerfect will sit on top of and you can perfect Perfect prompts in the same way that you can in chat GPT and prompt perfect, but also you can receive feedback. We'll give you feedback on your prompt if you're interested in learning. Something that some people want to know how to do it better themselves and others just want to hit that perfect button and get it done for them. But we give the option to get that feedback. And then the third button, which is very, very new for us is the ability to save prompts. We've heard time and time and time again from customers that they want to be able to save prompts. So whether that's in a notepad that they have on their desktop that they're constantly copying and pasting from, now that will live within their browser. And not only will it live within their browser, but when they go from ChatGPT to Claude, it follows them there. Those saved prompts will always be there within that browser for them. So we're really trying to expand the offering from PromptPerfect and making it more omnipresent. We just launched a couple of days ago. It's been very, very exciting and people really seem to like it.

  • Speaker #0

    We are going to put links so their audience can find out more about Prompt Perfect, this new product that Charles Ferguson is referring to and talking about. You guys are just doing some really amazing things. I think saving those prompts from a selfish standpoint, that's huge. Charles, I've gotten to know you a little bit over the last several months as I've used the product. You've been very available and I'm very honest with my audience about my interest and use. Thanks. of Front Perfect. Why don't you tell us a little bit more about what some other businesses might be using the tool for and in particular AI, because I think sparking some of that information for all of us might create some additional enthusiasm. I'm very enthusiastic about the product, but I think others can be too, if they kind of learn how it might apply to them.

  • Speaker #1

    I have six, seven years of experience selling to businesses. This is the first time I'd ever sold to just people. So, you know, it's a fast app that's available. to just consumers globally. So it's a very different approach. But we're able to receive feedback from folks. But ultimately, these are people with prompt engineering is new to them, and we get a lot of feedback from them. But ultimately, for some time over the summer, we spent about four months and we did a very concerted business to business play. We worked with businesses to see if they wanted services, we did some consulting, and we built some bespoke bots for people instead of them just using Prompt Perfect, which is very general purpose. We built them very specific bots. We worked with a few different teams in very different capacity. And like I'd said, I offered 15 minutes to anybody who wanted it, which was amazing over the summer. So I spoke to hundreds of people. It was very interesting.

  • Speaker #0

    I was one of those.

  • Speaker #1

    Yes.

  • Speaker #0

    The first. Yeah. Oh, well,

  • Speaker #1

    yeah. I was the first, literally the first. The diversity of applications is amazing. But ultimately, some of the things that we saw are most, so many of these conversations, the ideas are amazing, but it's impractical to build them. somebody in France was working on building something that was going to be an assistant for her grandmother to be able to understand her options in terms of community building in her neighborhood or in her local community. So I want to play tennis, but it's very specific because she doesn't want to play with a young 22-year-old boy. She wants to play with people her age. She wants to be able to play in more of a comfortable setting. And there really isn't a good tool or use case. There isn't really a great marketplace for those sorts of facts. And she was really looking to aggregate that. It's not something that we could build. The infrastructure for that sort of information just doesn't exist. So we weren't able to deliver on that. What we really saw is that the more clearly defined the problem is, the easier it is for us to implement it and to solve for that. And what that really looks like is, for example, we work with a franchise that does insect... I don't want to give it away, but they do insect application sprays on properties in the United States. And what they want to be able to do, what they need to be able to do is they take an address that comes through an inbound form, and a customer support agent then goes to Zillow or Redfin through Google and tries to find the square footage and then uses a pricing matrix with six different columns to figure out based on the size of the lot, how much the cost should be, and then also depending on what the upsell could be or whether or not it's adjacent to the water is a factor in the pricing. And what we were able to do is since that all is really just one big equation, went from that individual spending about 20 minutes typing in these things into Google to just using a GPT that we built them, a chatbot that is trained very specifically on that pricing matrix, has access to Zillow through an API. And what we're able to then do is you just type in the address. That's really all it wants is the address. And then we on the backend have not only provided it with that information from Zillow and the pricing matrix, but also then the email copy and the phone copy for this person to follow up with them. So it takes that. 20-minute process, distills it into about 90 seconds or less, depending on how long the copy is. And then that allows for that person to immediately follow up and feel a little bit less aimless is what the founder of that business said explained to us, is that there's a little bit of indecision. And it can be difficult to come up with these conversations off the top of your head, but by having a script here with all of the details already inputted, not as variables, but as the input, was extremely helpful. So things like that. Also, we found a lot of value in reporting. So for me personally, as I've learned how to do bookkeeping and work with our accountants. I've had a lot of success in our weekly reporting. which is always the same. My co-founder and I want to know the same metrics every week. It's not really that variable. And so since we know what we're looking for, we're able to just build an equation around it, so to speak. So I know that we want to know daily active users, monthly active users, the conversion rate from users, and at what point they're converting, things like that. But they're always the same questions. So if you're able to just articulate those questions and then help bridge the gap in getting them answered by providing the input that it requires each week. So the variable is the input, which is variable week over week. the questions remain the same. So you build this GPT. If you pay for ChatGPT+, which is $19 a month in the United States, I believe, building GPTs is something that comes with that. So we've built internally over 100 of these for various use cases. But the more clearly defined you can distill your problem and think about the process itself and what it is that needs to happen, the easier it is then to make that process much, much faster, because it's already clearly defined and the variable itself is what you're inputting. So I'm not sure if that's useful to the audience. I hope that it is. I hope it's not too abstract.

  • Speaker #0

    No, I think that's very helpful. I think these ways that we can winnow down what it is that we're doing from real broad, almost top of the funnel down into kind of a narrow, more narrow understanding of what it is that we really can do to make ourselves efficient. I mean, that's how I look at this in terms of my business. I so appreciate your time, Charles Ferguson. I really, I just have one broad question to end with. And it has to do with our upcoming election. We're really smack dab in the middle of that. Here we are on Halloween. The general election is coming up quickly. How do you feel about regulations with respect to AI? Because it's moving so fast. Your own business has grown so quickly and AI is just developing at a pace that it's moving so quickly that I think it's hard to regulate it. But maybe there are some needs. You hear sometimes the term guard rails. I don't know if that makes sense. really good sense, but how do you view this? And what's your perspective as an industry leader around AI?

  • Speaker #1

    I'm flattered to think that we're industry leaders. I appreciate that. But I think that for myself, the person I most relate to, this is something that people do talk a lot about. It's very interesting. And I think that as we had said earlier, I think it's easy to have a little bit of a scared mindset, especially when you see, I remember watching Star Wars as a boy. My dad watched Star Wars as a boy with C-3PO, who's an autonomous agent to themselves. So these are not new ideas and these are not new things to be afraid of.

  • Speaker #0

    When it comes to regulation, I was doing some research about the last few days ahead of this, what has been instantiated, there was an executive order, I remember that way before that executive order had been put into place, they had solicited feedback, they had asked and my co founder and I went back and forth. And we had thoughts on it and shared those thoughts. And I'm sure that there are hundreds of thousands of people who did the same there must have been just because it was a public request. And so the idea that it's an open discussion, I think is an extremely important thing. you do see some headlines and it's interesting to see them. There's a little bit of around Sam Altman, the CEO of OpenAI. Sometimes some of the things he says makes me afraid. And I wonder why. There's somebody who I really admire who seems to be somewhat of an independent thinker in this regard. His name is Mike Knoop. And he is the co-founder of Zapier, which is an amazing automation tool that's been around for over 10 years, which I've used extensively for all sorts of roles in the past. I think that we still use it internally at Prompt Perfect. I think that my co-founder uses it for some of the email marketing automations that we do. But Mike Knoop co-founded something called the ARC Prize in concert with Francois. I don't know his last name off the top of my head. I would butcher it. He's from France. But nonetheless, Mike Knoop thinks the only way to set good policy is really from looking at what can be done currently based on empirical evidence. It's more of a scientific pragmatic perspective, which I appreciate personally, as you can tell. And we should make decisions based on that empirical evidence, because we run the risk of cutting off progress too early if we're doing it based on theoretical situations that are impossible at the current state. Maybe that comes off as naive. I think that my instinct to steel man it is maybe I'm being naive and I'm not privy to all of the information that's available. But the way that we got to here with the large language models and transistors that enable that large language model and the neural networks beneath that, that was all done in the open source scientific movement through Google. and through DeepMind, through Google, all these different organizations are publishing work. Francois worked at Google when he published his paper in 2019. And so I think that the idea that we should cut these things off from public and from real experts to be able to discuss them and make them regulated based on very well-funded, self-interested to a certain extent, not because that's a negative thing as much as it's just a function of the way that the economy works. And I don't want to pass judgment on that, but I do think that in the same way that the internet was commissioned initially, by the government and then asked for it to be given back to them from these educators who were instantiating in Central and Southern California. And they had said, no, it's not something that we should centralize through this entity that requires trust. We should make this more open to people. People should be able to host their own server. I think that that's a beautiful notion. Although it might seem, again, I hate to come off as naive, but I do think that it's important to continuously evaluate where things currently stand, what is actually happening, and regulate based on that and not necessarily be reactive, but. the way that you get ahead of things is keeping experts in the room and not just listening to one or two who are the loudest or maybe the most well-funded. And I hate to cause waves or seem like I'm being contrarian. It's not my intention, but I really like the way that Mike New particulates this notion.

  • Speaker #1

    Thanks for sharing that. And no, not naive in the least, very informed and very helpful. It's so great to get this perspective, Charles Ferguson. We'd love to have you back too. This is just moving so quickly. Maybe you'll come back and... talk to us again because I know our audience is going to be really interested in this subject and as it develops certainly it's moving so quickly but please selfishly come back and talk to us a little bit about that because it's great to talk to you good luck with the rest of all of this this rollout we will have links so that our audience can find out more about Charles Ferguson and Prompt Perfect such a wonderful product and one that I just want to recommend to our audience to check out because I think it offers some vital support for somebody who is interested in this subject and wants to learn more. But Charles Ferguson, thank you for all of your work. My best to you. Have a great Halloween and certainly we'll be in touch with you.

  • Speaker #0

    Appreciate it. Thanks for the support and thanks for all you're doing.

  • Speaker #1

    My thanks, of course, to Charles Ferguson for joining us today. You'll find links in our notes for more information about Charles Ferguson, all of his details, including information about his company, Prompt Perfect. My thanks, of course, to our executive editor, Sam Henniger, for all of his work. My thanks always to our audience on radio and podcast. Please be well, be safe, and let's talk about better. The Not Old Better Show on radio and podcast. Thanks, everybody. We will see you next week.

  • Speaker #2

    Thanks for joining us this week on The Not Old Better Show. To find out more about all of today's stories or to view our extensive back catalog of previous shows. simply visit notold-better.com. Join us again next time as we deep dive into some of the most fascinating real-life stories from across the world, all focused on this wonderful experience of getting better, not just older. Let's talk about better. The Not Old Better Show.

  • Speaker #1

    I won't find a thing. Please check out our website for this episode and all episodes at notold-better.com or subscribe to the podcast on Apple Podcasts. And be sure to check out your local radio stations to find out more about The Not Old Better Show on podcast and radio. You can find us all over social media. Our Twitter feed is notoldbetter and we're on Instagram at notoldbetter2. The Not Old Better Show is a production of NOBS Studios. I'm Paul Vogelsang and I hope you'll join me again next time to talk about... Better. The Not Old Better Show. Thanks, everybody. We'll see you next week.

Description

Welcome to The Not Old – Better Show, Technology Interview Series on radio and podcast, where we explore ideas that matter to those who are 50 and older, encouraging discovery, adventure, and learning. 

Today’s episode promises to be one of our most insightful yet, as we delve into a topic shaping our world at lightning speed: artificial intelligence. 


Our guest is Charles Ferguson, the co-founder of PromptPerfect, a pioneering company in AI innovation that is making waves with its cutting-edge tools designed to enhance how we interact with AI. Charles has been at the forefront of developing AI solutions that are not only powerful but also accessible, especially for small businesses and individuals who might just be stepping into the AI landscape.


In this conversation, we’re going to cover a lot of ground—from what AI actually is and how it can be useful, to whether we should be wary of its rapid rise. Charles will offer insights on how tools like PromptPerfect work, why he believes AI is a vital tool for growth, and how you, our listeners, can embrace AI in your daily lives, whether for personal projects or small business ventures.

With AI becoming more prevalent in everything from healthcare to marketing, this episode is for everyone who’s curious, cautious, or just eager to understand where AI is headed and how it can be a force for good—especially for those who might feel left behind in the digital age. So, sit back, enjoy the conversation, and let’s learn together.


Join me in welcoming our guest today, co founder of PromptPerfect, Charles Ferguson

My thanks to all, Charles Ferguson, Sam Heningerand our wonderful audience on radio and podcast.  Be well, be safe and let’s talk about better.  The Not Old Better Show.


Hosted by Ausha. See ausha.co/privacy-policy for more information.

Transcription

  • Speaker #0

    Welcome to the Not Old Better Show, the show covering all things health, wellness, culture, and more. The show for all of us who aren't old, we're better. Each week, we'll interview superstars, experts, and ordinary people doing extraordinary things, all related to this wonderful experience of getting better, not older. Now, here's your host, the award-winning Paul Vogelzang. Welcome to the Not Old Better Show Technology. interview series on radio and podcast where we explore ideas that matter to those who are 50 and older, encouraging discovery, adventure, and learning. I'm Paul Vogelsang, and today's episode promises to be one of our most insightful yet as we delve into a topic shaping our world. It is everywhere right now, and it's doing this at lightning speed. I'm talking, of course, of artificial intelligence, AI. Our guest... is Charles Ferguson, the co-founder of Prompt Perfect, a pioneering company in AI innovation that is making waves with its cutting-edge tools designed to enhance how we interact with AI, not just for businesses, but for individuals, for small businesses, really for everyone. And you're going to hear Charles talk about what it's been like to be at the forefront of developing AI solutions that are not only powerful, but also accessible to all of us, especially for small businesses and individuals who might just be stepping into the AI landscape. In this conversation with Charles Ferguson, we are going to cover a lot of ground from what AI actually is, how it can be useful, to whether we should be wary of its rapid rise. Charles Ferguson will offer some insights today on how tools like PromptPerfect work, why he believes AI is a vital tool for growth as a society and how you, our listeners, can embrace AI in your daily lives, whether for personal projects or small business ventures. You're going to love this interview with AI becoming more prevalent in everything from healthcare to marketing. This episode is for everyone who's curious, cautious, or just eager to understand where AI is headed and how it can be a force for good, especially for those who might feel left behind in the digital age. So sit back, enjoy our conversation with Charles Ferguson. Let's learn together. Join me in welcoming our guest today, co-founder of Prompt Perfect, Charles Ferguson. Charles Ferguson, welcome to the program.

  • Speaker #1

    Thanks, Paul. I appreciate it. We're very, very appreciative of the opportunity to chat.

  • Speaker #0

    Well, thank you. I am excited about that too. I think first and foremost, happy Halloween. We're talking on Halloween. And so I want to make sure to extend those wishes. My wife is going to come home after being away for a few days, so I don't think we're going to have any trick-or-treaters. How about you guys? Are you planning for some?

  • Speaker #1

    No, I think of this more of the eve of my girlfriend's birthday, more so than Halloween, to be candid. But there's certainly a lot of people walking around in costume outside of New York City right now.

  • Speaker #0

    Yep, yep. Well, again, welcome and congratulations on all you are doing with Prompt Perfect. We're going to talk about that and AI, the subject of artificial intelligence today. I think that's on everybody's minds. And I want to get into that with you. But again, first and foremost, congrats on this wonderful tool. I use it. I'm going to talk about it a little bit personally, but I love what you guys are doing.

  • Speaker #1

    Yes, we've been very, very fortunate with the timing. It was sort of built out of necessity for ourselves, as we'll go into, I'm sure. And the reception has been incredible. It's just so much fun to be able to work with people all over the world, over 130 countries represented in our customer base of hundreds of thousands of people. It's been used over a million and a half times. And so we're glad that anybody finds it useful, but we also use it daily. And it's been a long 18 months of a lot of iteration and it's gotten a lot better. And we're excited also to be launching a new product this week, very similar. It's omnipresent. It isn't confined to ChatGPT. It's within your web browser, within Google Chrome.

  • Speaker #0

    I'm a user. I love the tool. It has been so helpful for me. This is purely an editorial interview. I can talk and wax. poetically on this service that you offer and this fantastic means for me to do work as a small business person. But I think it's important to point out that I think our audience is going to be starting a few steps back in terms of their understanding of what artificial intelligence represents. And so I wonder if you could tell us about the inspiration for starting Prompt Perfect. I know a little bit about the story, but it is this amazing tool. Did you start to see some gaps in some of the development that was going on around artificial intelligence?

  • Speaker #1

    Yes, I need to make sure first and foremost, it's extremely clear that my co-founder is the ideator here, my closest friends from college. We lived together for a long time and we spent about 10 years working on other things completely separately, but always remain really, really close. When ChatGPT 3 and then 3.5 had come out and there was a front end introduced and that was in 2023. We were playing with the tool a little bit, but it was really my co-founder who was trying to assist himself with this new tool to make his job easier. He's a copywriter at the time, and he found a lot of inspiration with providing examples of copy that he loved, and then getting iterations of things that were more applicable to him based on context he was providing. And what he was really teaching himself is prompt engineering before reading any papers on it. So just understanding the correlation between the input you provide it as being directly correlated to the quality of the output you receive back, which is very different than Google search or internet browsing. They're very, very different. And so the skill set is different. And that's something that he just intuited himself almost two years ago now. And he built it out of hackathon over a weekend. He had been given that opportunity. And then subsequently, ChatGPT introduced this beta feature plugins, and it was featured as one of the top eight plugins. And that was before we really had good analytics or any sort of understanding of where this might go. And overnight, within three weeks, there were thousands and thousands of users. And it was very interesting to see. We just sort of took it as it came. And he had asked me to be his co-founder around that time. And we had a lot of bookkeeping to do. And that's neither of our core competencies, but I figured it out. And we've come a long way in that time. It really came from a necessity of trying to figure out how do we use these tools effectively and realizing this is not the same as our past 15, 20 years of experience working and searching on the internet. This is a very new skill set. It's all about understanding the context of the question you're trying to ask. providing as much context and detail as possible. And that's going to provide better responses always. It really came from us trying to get the most out of the platform, getting the most out of ChatGPT at the time, without having to spend too much time writing out the details. And so what Prompt Perfect does is it takes the sentence or the prompt that you provided, and it enhances it with more context and more detail. And it shows you that more detailed prompt and then it'll show you the response that that more detailed and that more context provides to you. So it enhances prompts for people. We assumed people cared about how to learn how to prompt engineer, but what we've learned from launching eight products is they're much more interested in having that activity done for them more so than learning how to do it. I think it's a little bit easier if it just sort of does it for you. And maybe it'll show you its work, but ultimately teaching somebody to prompt engineer isn't nearly as much fun or what we found is useful economically for people. They just want it done for themselves. So that's what we do is we just provide a service that enhances your prompts for you automatically.

  • Speaker #0

    Brilliant. Again, congratulations to you and your co-founder, because I'm always amazed at the development of businesses and how they really get started. And I want to drill down for just a moment and make sure that our audience understands that Prompt Perfect serves as a tool, as you describe it, is really a companion to ChatGPT. And in order to really maximize the sheer efforts that ChatGPT can bring. prompt perfect allows you to enter a string of words that are related but you don't necessarily have to share the sheer volume of material that you might within chat gpt because the technology within prompt perfect supplements your own work to give you this tool that really generates this fantastic result that you're getting from the artificial intelligence yes

  • Speaker #1

    without a doubt i think if you're using prompt perfect but you're very well versed in prompt engineering, you might not see that significant of an enhancement in your prompt. So if you provide it with five paragraphs and 300 words, and it's extremely detailed, we're going to be able to automatically enhance that maybe slightly or change some syntax. And we really try to optimize for the model. And so we're trying to think about how the model is going to receive that. So we're iterating on that. So it could see some improvements. But the time when we see the most significant improvement is in two, three, four, five lines, stream of consciousness. with misspellings, whatever it might be, just people being people. And what we are going to be able to do there is enhance that and show you how we've enhanced it, most importantly to me, so that you can then ultimately employ that strategy yourself. But the idea of the time savings of not having to figure out how to word these things or provide that context, we've seen that our customers find it very, very valuable.

  • Speaker #0

    Yeah. And as I say, I'm a customer. I'm a small businessman. I am just a tiny, tiny little business. It's really me. And I've since expanded. my business a little bit so that I work with an editor. I'm working with another individual who helps kind of keep me on track in terms of schedule. I really look at Prompt Perfect as another level of staffing to my business. It helps me that much. It helps me refine questions. It helps me with writing passages for the program. It helps me with social media posts. It really does so much to supplement me. Yeah. And so selfishly, I just think this tool is all about empowering people. Was that kind of your thinking in terms of empowering individuals and small businesses? Because I will tell you again, that's what it's doing for me.

  • Speaker #1

    That's fabulous. We love to hear that. I feel I'm gushing. I'm not sure how to explain it more than that. But Eric and myself, we both immediately following undergrad, we graduated in 2013. I'd done the Peace Corps for a couple of years. He was in AmeriCorps as a teacher. And so we're both very service oriented. Well, I don't profess to say that Eric had thought of this because he was doing it out of a greater good. It really was a self necessity. And we're a two person business as well. And we use it constantly. And we have a lot of feedback from that. We've spoken to thousands. High hundreds at the very least. I haven't aggregated all of the numbers in terms of the user feedback and the way that we correspond with folks, but I wouldn't be terribly surprised if it's over a thousand after 18 months. The feedback is resoundingly positive. We did, actually, for four months very concertedly double-click into trying to work with small businesses specifically. I had hundreds of conversations with different small business owners, mostly small businesses, and then we ultimately built some bespoke products for several of those small businesses. It's a common thread. AI is something that we internally call them interns. They're not somebody you're going to hand a task over to entirely. It's not going to get it over the finish line necessarily or create hundreds of blog posts for you without any sort of thought. There should be, and I think there will be for the foreseeable future, a human in the loop, so to speak. And also just for quality control and making sure that it's in your voice and your tone and things are done accurately. But it is a massive time saver for all sorts of small businesses. We're very fortunate. And we've also, one of the things is working with small businesses was a surprise, but also we've been very surprised, or at least I was, I should say, by the international adoption. So I'd said earlier, 130 some odd countries, 133, 134 countries. So much of what people are using Prompt Perfect for is English as a second language, but not necessarily as a tutor as much as just making their own verbiage more clear, a little bit more concise, different things like that. We find a lot of feedback from our customers around that as well. So helping them level up their own writing in English, which is really, really great. And then also in Greek, we worked with a Greek marketer who we figured out how to make the Greek as good as the English, depending on the time at which you ask it to translate into Greek. So corresponding in English and then asking it to translate into Greek at the end of the conversation, so on and so forth. There are these little hacks that you kind of have to work through together. But I think that language and natural language processing is unbelievably exciting. We've been very, very lucky to be at the intersection of that.

  • Speaker #0

    I will say anything that is done to improve communication this day and age is, I think, a boon and a bonus for all of us. So hats off to you for all of that work, too. Well, give us a very basic level of understanding about artificial intelligence. We hear the term an awful lot. It's bandied about in some very positive ways, and we'll talk about that. It's bandied about in terms of its potential for harm, perhaps. shifting the economy one direction or another. For our audience of those who may not be as current on some of this technology, give us that fundamental understanding of what artificial intelligence really is all about.

  • Speaker #1

    Sure, of course, AI or artificial intelligence as a technology, it's not brand new with the advent of chat GPT, things like recommendations on Netflix or recommendations on Spotify or on YouTube, those algorithms. are a form of artificial intelligence, even in Microsoft Word or Google Sheets or Grammarly, which sort of lives above all of those things. That ability for it to distill grammatical errors and spelling errors, those are more the grammatical side, but Grammarly has been around probably not since I was in college, but for some time now, I know that it's very, very popular. But in terms of the reason that people get, at least myself, get concerned is this term or this notion of artificial general intelligence, which is a completely autonomous agent that can think on its own and has the ability to think on its own. to make its own decisions and enact in its own best interest, which makes me immediately think of The Matrix, one of my favorite movies of all time, just farming people for their energy. Obviously, those dystopian ideals will sell movie tickets, but I think AGI is still very far away from where we are now. Of course, we do have AI that has for years beaten humans at chess or Go or any of these complex games. But what you'll notice is that those AI instantiations, one of them that can beat you in chess, they don't use that same model to beat you in Go. what they have to do is they start from square one again. And that model is unable to distill new knowledge or to deduce how to play a new game. It's not using cognition in the way that humans do. Even with my cousins, I can teach them how to play a card game because they have an understanding of how a deck of cards works. And that's not the way that these models really work. So it's all about the training data. And so we do see a lot of excitement over, let's just give it more data. That's the answer to this. We need to give it more data. And people are really focused on large language models. There's a lot more out there. There are a lot more ideas than just large language models. And I think that AGI as an idea, which I think personally is very exciting, although I do love science fiction and do find myself reading as much science fiction as I can make time for. I do think that that is a very far away concern personally. In terms of what artificial intelligence is, it's really very simple. It's just the ability for a machine. to think in a way that we recognize as human. So if you think about it like that, there are all these instances, whether it's navigations and using ways on your phone telling you where to turn as an instance of that. And then obviously, there are these Alexa and Siri, which we will leave a little bit to be desired, but are certainly an instantiation of that AI. But I think the reason that people are so excited about it is because of ChatGPT and tools like that, which are chatbots empowered by AI. So I would call them AI chatbots. But it's a very complex and complicated question. I think the chatbots are really what people think of with AI now, though. And I don't think that YouTube recommendations or Netflix recommendations, even though my mom uses them, I don't think that she would really think about that as AI. Although it technically is, it certainly is.

  • Speaker #0

    Our guest today is Charles Ferguson. Charles Ferguson co-founded Prompt Perfect. We'll have links in our show notes today for more information about Prompt Perfect, more information about Charles Ferguson and the work that the firm is doing with Prompt Perfect. I know you're introducing some... new products. I saw, for example, today that my prompt perfect was updated just recently. You guys are very busy doing all of this work. I love one of your quotes that you say, the internet is the most important invention in recent history because of the proliferation of information it enables. Should we be afraid of some of this information as it gets aggregated? There's almost a natural fear of some of the unknown, but knowing this, I don't know that fear is the right way to proceed. What's your reaction?

  • Speaker #1

    It's a great question. I do think that the democratization of access to information is an absolute net positive, and it'd be very difficult for me to concede that somehow. But I do think that there is very reasonable concern, and it's important to be concerned, and it's important to be curious enough to ask questions about these things and to be critical. I absolutely do not think of it as a utopia where everything is going to be perfect. The internet is absolutely... an open place. And that's why when you go and you just type into ChatGPT, which is trained on the internet, and the internet is inherently average, but just by virtue of the fact that there are such a large group of people contributing to it. And just in general, that's sort of the definition of what average is. So what you're going to do is when you type in a simple query, you're going to get a relatively simple response or an average response. Whereas if you tell it or prompt it in such a way and let it know, hey, you're an expert, you're a senior level programmer, keep that in mind in your response. The response very significantly becomes higher quality very quickly. So I think that when it comes to the concern around AI and that information, that misinformation, I think it's very, very reasonable. But I tend to have more of a growth mindset than a scarcity mindset, which sometimes gets me in a little bit of trouble. But I think back to my co-founder is very, very talented and his girlfriend as well at Improv and they're always saying yes and as a sort of a mantra. And I think that the ability to sort of pick things up and be intellectually curious about them and to not think of them as necessarily inherently negative. I think that the internet was founded by a group of people, which is a very positive thing. It's the centralization of ideation that I think is a really, really scary ideal. And I think that the internet is sort of the antithesis of that. And I think that AI has the potential to be an extension of that. But the internet is certainly not being replaced by AI. I think it's just being augmented by it, which is, I think, a positive thing.

  • Speaker #0

    Yeah, I tend to look at this very positively, too. I wonder if you'll tell us a little bit about what's coming out of Prompt Perfect, the group. And you touched on the new product and how it's going to be browser-based, correct me, but tell us a little bit about what's coming from Prompt Perfect.

  • Speaker #1

    So it's been 14 months of being live within ChatGPT. And so what we've done is we have sort of moved with them and ridden that whale. We're sort of one of those small fish under the large whale. And we've been sort of adapting as they adapt. And they went from plugins to GPTs. And so we graduated into a GPT, which then became a feature that is available to everybody. So everybody can use Prompt Perfect within the ChatGPT marketplace now, which is fabulous. We saw a huge uptick in usage as a result of that. But ultimately, not everybody uses ChatGPT. So we've really narrowed ourselves to being just in that one marketplace. And really, just to take a step back, what Prompt Perfect does is it just perfects prompts, it just enhances prompts. And what we've done is we've taken a lot of conversations with customers in the last year, And we've built a new product that lives within your browser instead of within ChatGPT. So anybody who uses Google Chrome might be familiar with Chrome extensions, whether it's an antivirus software or something like that. What now PromptPerfect does is it sits as a layer on top of all of these different chatbots. So whether it's Gemini from Google, or Anthropix, LLM POE, or any one of these different examples, I'm sorry, Claude, but nonetheless, any of these chatbots now PromptPerfect will sit on top of and you can perfect Perfect prompts in the same way that you can in chat GPT and prompt perfect, but also you can receive feedback. We'll give you feedback on your prompt if you're interested in learning. Something that some people want to know how to do it better themselves and others just want to hit that perfect button and get it done for them. But we give the option to get that feedback. And then the third button, which is very, very new for us is the ability to save prompts. We've heard time and time and time again from customers that they want to be able to save prompts. So whether that's in a notepad that they have on their desktop that they're constantly copying and pasting from, now that will live within their browser. And not only will it live within their browser, but when they go from ChatGPT to Claude, it follows them there. Those saved prompts will always be there within that browser for them. So we're really trying to expand the offering from PromptPerfect and making it more omnipresent. We just launched a couple of days ago. It's been very, very exciting and people really seem to like it.

  • Speaker #0

    We are going to put links so their audience can find out more about Prompt Perfect, this new product that Charles Ferguson is referring to and talking about. You guys are just doing some really amazing things. I think saving those prompts from a selfish standpoint, that's huge. Charles, I've gotten to know you a little bit over the last several months as I've used the product. You've been very available and I'm very honest with my audience about my interest and use. Thanks. of Front Perfect. Why don't you tell us a little bit more about what some other businesses might be using the tool for and in particular AI, because I think sparking some of that information for all of us might create some additional enthusiasm. I'm very enthusiastic about the product, but I think others can be too, if they kind of learn how it might apply to them.

  • Speaker #1

    I have six, seven years of experience selling to businesses. This is the first time I'd ever sold to just people. So, you know, it's a fast app that's available. to just consumers globally. So it's a very different approach. But we're able to receive feedback from folks. But ultimately, these are people with prompt engineering is new to them, and we get a lot of feedback from them. But ultimately, for some time over the summer, we spent about four months and we did a very concerted business to business play. We worked with businesses to see if they wanted services, we did some consulting, and we built some bespoke bots for people instead of them just using Prompt Perfect, which is very general purpose. We built them very specific bots. We worked with a few different teams in very different capacity. And like I'd said, I offered 15 minutes to anybody who wanted it, which was amazing over the summer. So I spoke to hundreds of people. It was very interesting.

  • Speaker #0

    I was one of those.

  • Speaker #1

    Yes.

  • Speaker #0

    The first. Yeah. Oh, well,

  • Speaker #1

    yeah. I was the first, literally the first. The diversity of applications is amazing. But ultimately, some of the things that we saw are most, so many of these conversations, the ideas are amazing, but it's impractical to build them. somebody in France was working on building something that was going to be an assistant for her grandmother to be able to understand her options in terms of community building in her neighborhood or in her local community. So I want to play tennis, but it's very specific because she doesn't want to play with a young 22-year-old boy. She wants to play with people her age. She wants to be able to play in more of a comfortable setting. And there really isn't a good tool or use case. There isn't really a great marketplace for those sorts of facts. And she was really looking to aggregate that. It's not something that we could build. The infrastructure for that sort of information just doesn't exist. So we weren't able to deliver on that. What we really saw is that the more clearly defined the problem is, the easier it is for us to implement it and to solve for that. And what that really looks like is, for example, we work with a franchise that does insect... I don't want to give it away, but they do insect application sprays on properties in the United States. And what they want to be able to do, what they need to be able to do is they take an address that comes through an inbound form, and a customer support agent then goes to Zillow or Redfin through Google and tries to find the square footage and then uses a pricing matrix with six different columns to figure out based on the size of the lot, how much the cost should be, and then also depending on what the upsell could be or whether or not it's adjacent to the water is a factor in the pricing. And what we were able to do is since that all is really just one big equation, went from that individual spending about 20 minutes typing in these things into Google to just using a GPT that we built them, a chatbot that is trained very specifically on that pricing matrix, has access to Zillow through an API. And what we're able to then do is you just type in the address. That's really all it wants is the address. And then we on the backend have not only provided it with that information from Zillow and the pricing matrix, but also then the email copy and the phone copy for this person to follow up with them. So it takes that. 20-minute process, distills it into about 90 seconds or less, depending on how long the copy is. And then that allows for that person to immediately follow up and feel a little bit less aimless is what the founder of that business said explained to us, is that there's a little bit of indecision. And it can be difficult to come up with these conversations off the top of your head, but by having a script here with all of the details already inputted, not as variables, but as the input, was extremely helpful. So things like that. Also, we found a lot of value in reporting. So for me personally, as I've learned how to do bookkeeping and work with our accountants. I've had a lot of success in our weekly reporting. which is always the same. My co-founder and I want to know the same metrics every week. It's not really that variable. And so since we know what we're looking for, we're able to just build an equation around it, so to speak. So I know that we want to know daily active users, monthly active users, the conversion rate from users, and at what point they're converting, things like that. But they're always the same questions. So if you're able to just articulate those questions and then help bridge the gap in getting them answered by providing the input that it requires each week. So the variable is the input, which is variable week over week. the questions remain the same. So you build this GPT. If you pay for ChatGPT+, which is $19 a month in the United States, I believe, building GPTs is something that comes with that. So we've built internally over 100 of these for various use cases. But the more clearly defined you can distill your problem and think about the process itself and what it is that needs to happen, the easier it is then to make that process much, much faster, because it's already clearly defined and the variable itself is what you're inputting. So I'm not sure if that's useful to the audience. I hope that it is. I hope it's not too abstract.

  • Speaker #0

    No, I think that's very helpful. I think these ways that we can winnow down what it is that we're doing from real broad, almost top of the funnel down into kind of a narrow, more narrow understanding of what it is that we really can do to make ourselves efficient. I mean, that's how I look at this in terms of my business. I so appreciate your time, Charles Ferguson. I really, I just have one broad question to end with. And it has to do with our upcoming election. We're really smack dab in the middle of that. Here we are on Halloween. The general election is coming up quickly. How do you feel about regulations with respect to AI? Because it's moving so fast. Your own business has grown so quickly and AI is just developing at a pace that it's moving so quickly that I think it's hard to regulate it. But maybe there are some needs. You hear sometimes the term guard rails. I don't know if that makes sense. really good sense, but how do you view this? And what's your perspective as an industry leader around AI?

  • Speaker #1

    I'm flattered to think that we're industry leaders. I appreciate that. But I think that for myself, the person I most relate to, this is something that people do talk a lot about. It's very interesting. And I think that as we had said earlier, I think it's easy to have a little bit of a scared mindset, especially when you see, I remember watching Star Wars as a boy. My dad watched Star Wars as a boy with C-3PO, who's an autonomous agent to themselves. So these are not new ideas and these are not new things to be afraid of.

  • Speaker #0

    When it comes to regulation, I was doing some research about the last few days ahead of this, what has been instantiated, there was an executive order, I remember that way before that executive order had been put into place, they had solicited feedback, they had asked and my co founder and I went back and forth. And we had thoughts on it and shared those thoughts. And I'm sure that there are hundreds of thousands of people who did the same there must have been just because it was a public request. And so the idea that it's an open discussion, I think is an extremely important thing. you do see some headlines and it's interesting to see them. There's a little bit of around Sam Altman, the CEO of OpenAI. Sometimes some of the things he says makes me afraid. And I wonder why. There's somebody who I really admire who seems to be somewhat of an independent thinker in this regard. His name is Mike Knoop. And he is the co-founder of Zapier, which is an amazing automation tool that's been around for over 10 years, which I've used extensively for all sorts of roles in the past. I think that we still use it internally at Prompt Perfect. I think that my co-founder uses it for some of the email marketing automations that we do. But Mike Knoop co-founded something called the ARC Prize in concert with Francois. I don't know his last name off the top of my head. I would butcher it. He's from France. But nonetheless, Mike Knoop thinks the only way to set good policy is really from looking at what can be done currently based on empirical evidence. It's more of a scientific pragmatic perspective, which I appreciate personally, as you can tell. And we should make decisions based on that empirical evidence, because we run the risk of cutting off progress too early if we're doing it based on theoretical situations that are impossible at the current state. Maybe that comes off as naive. I think that my instinct to steel man it is maybe I'm being naive and I'm not privy to all of the information that's available. But the way that we got to here with the large language models and transistors that enable that large language model and the neural networks beneath that, that was all done in the open source scientific movement through Google. and through DeepMind, through Google, all these different organizations are publishing work. Francois worked at Google when he published his paper in 2019. And so I think that the idea that we should cut these things off from public and from real experts to be able to discuss them and make them regulated based on very well-funded, self-interested to a certain extent, not because that's a negative thing as much as it's just a function of the way that the economy works. And I don't want to pass judgment on that, but I do think that in the same way that the internet was commissioned initially, by the government and then asked for it to be given back to them from these educators who were instantiating in Central and Southern California. And they had said, no, it's not something that we should centralize through this entity that requires trust. We should make this more open to people. People should be able to host their own server. I think that that's a beautiful notion. Although it might seem, again, I hate to come off as naive, but I do think that it's important to continuously evaluate where things currently stand, what is actually happening, and regulate based on that and not necessarily be reactive, but. the way that you get ahead of things is keeping experts in the room and not just listening to one or two who are the loudest or maybe the most well-funded. And I hate to cause waves or seem like I'm being contrarian. It's not my intention, but I really like the way that Mike New particulates this notion.

  • Speaker #1

    Thanks for sharing that. And no, not naive in the least, very informed and very helpful. It's so great to get this perspective, Charles Ferguson. We'd love to have you back too. This is just moving so quickly. Maybe you'll come back and... talk to us again because I know our audience is going to be really interested in this subject and as it develops certainly it's moving so quickly but please selfishly come back and talk to us a little bit about that because it's great to talk to you good luck with the rest of all of this this rollout we will have links so that our audience can find out more about Charles Ferguson and Prompt Perfect such a wonderful product and one that I just want to recommend to our audience to check out because I think it offers some vital support for somebody who is interested in this subject and wants to learn more. But Charles Ferguson, thank you for all of your work. My best to you. Have a great Halloween and certainly we'll be in touch with you.

  • Speaker #0

    Appreciate it. Thanks for the support and thanks for all you're doing.

  • Speaker #1

    My thanks, of course, to Charles Ferguson for joining us today. You'll find links in our notes for more information about Charles Ferguson, all of his details, including information about his company, Prompt Perfect. My thanks, of course, to our executive editor, Sam Henniger, for all of his work. My thanks always to our audience on radio and podcast. Please be well, be safe, and let's talk about better. The Not Old Better Show on radio and podcast. Thanks, everybody. We will see you next week.

  • Speaker #2

    Thanks for joining us this week on The Not Old Better Show. To find out more about all of today's stories or to view our extensive back catalog of previous shows. simply visit notold-better.com. Join us again next time as we deep dive into some of the most fascinating real-life stories from across the world, all focused on this wonderful experience of getting better, not just older. Let's talk about better. The Not Old Better Show.

  • Speaker #1

    I won't find a thing. Please check out our website for this episode and all episodes at notold-better.com or subscribe to the podcast on Apple Podcasts. And be sure to check out your local radio stations to find out more about The Not Old Better Show on podcast and radio. You can find us all over social media. Our Twitter feed is notoldbetter and we're on Instagram at notoldbetter2. The Not Old Better Show is a production of NOBS Studios. I'm Paul Vogelsang and I hope you'll join me again next time to talk about... Better. The Not Old Better Show. Thanks, everybody. We'll see you next week.

Share

Embed

You may also like

Description

Welcome to The Not Old – Better Show, Technology Interview Series on radio and podcast, where we explore ideas that matter to those who are 50 and older, encouraging discovery, adventure, and learning. 

Today’s episode promises to be one of our most insightful yet, as we delve into a topic shaping our world at lightning speed: artificial intelligence. 


Our guest is Charles Ferguson, the co-founder of PromptPerfect, a pioneering company in AI innovation that is making waves with its cutting-edge tools designed to enhance how we interact with AI. Charles has been at the forefront of developing AI solutions that are not only powerful but also accessible, especially for small businesses and individuals who might just be stepping into the AI landscape.


In this conversation, we’re going to cover a lot of ground—from what AI actually is and how it can be useful, to whether we should be wary of its rapid rise. Charles will offer insights on how tools like PromptPerfect work, why he believes AI is a vital tool for growth, and how you, our listeners, can embrace AI in your daily lives, whether for personal projects or small business ventures.

With AI becoming more prevalent in everything from healthcare to marketing, this episode is for everyone who’s curious, cautious, or just eager to understand where AI is headed and how it can be a force for good—especially for those who might feel left behind in the digital age. So, sit back, enjoy the conversation, and let’s learn together.


Join me in welcoming our guest today, co founder of PromptPerfect, Charles Ferguson

My thanks to all, Charles Ferguson, Sam Heningerand our wonderful audience on radio and podcast.  Be well, be safe and let’s talk about better.  The Not Old Better Show.


Hosted by Ausha. See ausha.co/privacy-policy for more information.

Transcription

  • Speaker #0

    Welcome to the Not Old Better Show, the show covering all things health, wellness, culture, and more. The show for all of us who aren't old, we're better. Each week, we'll interview superstars, experts, and ordinary people doing extraordinary things, all related to this wonderful experience of getting better, not older. Now, here's your host, the award-winning Paul Vogelzang. Welcome to the Not Old Better Show Technology. interview series on radio and podcast where we explore ideas that matter to those who are 50 and older, encouraging discovery, adventure, and learning. I'm Paul Vogelsang, and today's episode promises to be one of our most insightful yet as we delve into a topic shaping our world. It is everywhere right now, and it's doing this at lightning speed. I'm talking, of course, of artificial intelligence, AI. Our guest... is Charles Ferguson, the co-founder of Prompt Perfect, a pioneering company in AI innovation that is making waves with its cutting-edge tools designed to enhance how we interact with AI, not just for businesses, but for individuals, for small businesses, really for everyone. And you're going to hear Charles talk about what it's been like to be at the forefront of developing AI solutions that are not only powerful, but also accessible to all of us, especially for small businesses and individuals who might just be stepping into the AI landscape. In this conversation with Charles Ferguson, we are going to cover a lot of ground from what AI actually is, how it can be useful, to whether we should be wary of its rapid rise. Charles Ferguson will offer some insights today on how tools like PromptPerfect work, why he believes AI is a vital tool for growth as a society and how you, our listeners, can embrace AI in your daily lives, whether for personal projects or small business ventures. You're going to love this interview with AI becoming more prevalent in everything from healthcare to marketing. This episode is for everyone who's curious, cautious, or just eager to understand where AI is headed and how it can be a force for good, especially for those who might feel left behind in the digital age. So sit back, enjoy our conversation with Charles Ferguson. Let's learn together. Join me in welcoming our guest today, co-founder of Prompt Perfect, Charles Ferguson. Charles Ferguson, welcome to the program.

  • Speaker #1

    Thanks, Paul. I appreciate it. We're very, very appreciative of the opportunity to chat.

  • Speaker #0

    Well, thank you. I am excited about that too. I think first and foremost, happy Halloween. We're talking on Halloween. And so I want to make sure to extend those wishes. My wife is going to come home after being away for a few days, so I don't think we're going to have any trick-or-treaters. How about you guys? Are you planning for some?

  • Speaker #1

    No, I think of this more of the eve of my girlfriend's birthday, more so than Halloween, to be candid. But there's certainly a lot of people walking around in costume outside of New York City right now.

  • Speaker #0

    Yep, yep. Well, again, welcome and congratulations on all you are doing with Prompt Perfect. We're going to talk about that and AI, the subject of artificial intelligence today. I think that's on everybody's minds. And I want to get into that with you. But again, first and foremost, congrats on this wonderful tool. I use it. I'm going to talk about it a little bit personally, but I love what you guys are doing.

  • Speaker #1

    Yes, we've been very, very fortunate with the timing. It was sort of built out of necessity for ourselves, as we'll go into, I'm sure. And the reception has been incredible. It's just so much fun to be able to work with people all over the world, over 130 countries represented in our customer base of hundreds of thousands of people. It's been used over a million and a half times. And so we're glad that anybody finds it useful, but we also use it daily. And it's been a long 18 months of a lot of iteration and it's gotten a lot better. And we're excited also to be launching a new product this week, very similar. It's omnipresent. It isn't confined to ChatGPT. It's within your web browser, within Google Chrome.

  • Speaker #0

    I'm a user. I love the tool. It has been so helpful for me. This is purely an editorial interview. I can talk and wax. poetically on this service that you offer and this fantastic means for me to do work as a small business person. But I think it's important to point out that I think our audience is going to be starting a few steps back in terms of their understanding of what artificial intelligence represents. And so I wonder if you could tell us about the inspiration for starting Prompt Perfect. I know a little bit about the story, but it is this amazing tool. Did you start to see some gaps in some of the development that was going on around artificial intelligence?

  • Speaker #1

    Yes, I need to make sure first and foremost, it's extremely clear that my co-founder is the ideator here, my closest friends from college. We lived together for a long time and we spent about 10 years working on other things completely separately, but always remain really, really close. When ChatGPT 3 and then 3.5 had come out and there was a front end introduced and that was in 2023. We were playing with the tool a little bit, but it was really my co-founder who was trying to assist himself with this new tool to make his job easier. He's a copywriter at the time, and he found a lot of inspiration with providing examples of copy that he loved, and then getting iterations of things that were more applicable to him based on context he was providing. And what he was really teaching himself is prompt engineering before reading any papers on it. So just understanding the correlation between the input you provide it as being directly correlated to the quality of the output you receive back, which is very different than Google search or internet browsing. They're very, very different. And so the skill set is different. And that's something that he just intuited himself almost two years ago now. And he built it out of hackathon over a weekend. He had been given that opportunity. And then subsequently, ChatGPT introduced this beta feature plugins, and it was featured as one of the top eight plugins. And that was before we really had good analytics or any sort of understanding of where this might go. And overnight, within three weeks, there were thousands and thousands of users. And it was very interesting to see. We just sort of took it as it came. And he had asked me to be his co-founder around that time. And we had a lot of bookkeeping to do. And that's neither of our core competencies, but I figured it out. And we've come a long way in that time. It really came from a necessity of trying to figure out how do we use these tools effectively and realizing this is not the same as our past 15, 20 years of experience working and searching on the internet. This is a very new skill set. It's all about understanding the context of the question you're trying to ask. providing as much context and detail as possible. And that's going to provide better responses always. It really came from us trying to get the most out of the platform, getting the most out of ChatGPT at the time, without having to spend too much time writing out the details. And so what Prompt Perfect does is it takes the sentence or the prompt that you provided, and it enhances it with more context and more detail. And it shows you that more detailed prompt and then it'll show you the response that that more detailed and that more context provides to you. So it enhances prompts for people. We assumed people cared about how to learn how to prompt engineer, but what we've learned from launching eight products is they're much more interested in having that activity done for them more so than learning how to do it. I think it's a little bit easier if it just sort of does it for you. And maybe it'll show you its work, but ultimately teaching somebody to prompt engineer isn't nearly as much fun or what we found is useful economically for people. They just want it done for themselves. So that's what we do is we just provide a service that enhances your prompts for you automatically.

  • Speaker #0

    Brilliant. Again, congratulations to you and your co-founder, because I'm always amazed at the development of businesses and how they really get started. And I want to drill down for just a moment and make sure that our audience understands that Prompt Perfect serves as a tool, as you describe it, is really a companion to ChatGPT. And in order to really maximize the sheer efforts that ChatGPT can bring. prompt perfect allows you to enter a string of words that are related but you don't necessarily have to share the sheer volume of material that you might within chat gpt because the technology within prompt perfect supplements your own work to give you this tool that really generates this fantastic result that you're getting from the artificial intelligence yes

  • Speaker #1

    without a doubt i think if you're using prompt perfect but you're very well versed in prompt engineering, you might not see that significant of an enhancement in your prompt. So if you provide it with five paragraphs and 300 words, and it's extremely detailed, we're going to be able to automatically enhance that maybe slightly or change some syntax. And we really try to optimize for the model. And so we're trying to think about how the model is going to receive that. So we're iterating on that. So it could see some improvements. But the time when we see the most significant improvement is in two, three, four, five lines, stream of consciousness. with misspellings, whatever it might be, just people being people. And what we are going to be able to do there is enhance that and show you how we've enhanced it, most importantly to me, so that you can then ultimately employ that strategy yourself. But the idea of the time savings of not having to figure out how to word these things or provide that context, we've seen that our customers find it very, very valuable.

  • Speaker #0

    Yeah. And as I say, I'm a customer. I'm a small businessman. I am just a tiny, tiny little business. It's really me. And I've since expanded. my business a little bit so that I work with an editor. I'm working with another individual who helps kind of keep me on track in terms of schedule. I really look at Prompt Perfect as another level of staffing to my business. It helps me that much. It helps me refine questions. It helps me with writing passages for the program. It helps me with social media posts. It really does so much to supplement me. Yeah. And so selfishly, I just think this tool is all about empowering people. Was that kind of your thinking in terms of empowering individuals and small businesses? Because I will tell you again, that's what it's doing for me.

  • Speaker #1

    That's fabulous. We love to hear that. I feel I'm gushing. I'm not sure how to explain it more than that. But Eric and myself, we both immediately following undergrad, we graduated in 2013. I'd done the Peace Corps for a couple of years. He was in AmeriCorps as a teacher. And so we're both very service oriented. Well, I don't profess to say that Eric had thought of this because he was doing it out of a greater good. It really was a self necessity. And we're a two person business as well. And we use it constantly. And we have a lot of feedback from that. We've spoken to thousands. High hundreds at the very least. I haven't aggregated all of the numbers in terms of the user feedback and the way that we correspond with folks, but I wouldn't be terribly surprised if it's over a thousand after 18 months. The feedback is resoundingly positive. We did, actually, for four months very concertedly double-click into trying to work with small businesses specifically. I had hundreds of conversations with different small business owners, mostly small businesses, and then we ultimately built some bespoke products for several of those small businesses. It's a common thread. AI is something that we internally call them interns. They're not somebody you're going to hand a task over to entirely. It's not going to get it over the finish line necessarily or create hundreds of blog posts for you without any sort of thought. There should be, and I think there will be for the foreseeable future, a human in the loop, so to speak. And also just for quality control and making sure that it's in your voice and your tone and things are done accurately. But it is a massive time saver for all sorts of small businesses. We're very fortunate. And we've also, one of the things is working with small businesses was a surprise, but also we've been very surprised, or at least I was, I should say, by the international adoption. So I'd said earlier, 130 some odd countries, 133, 134 countries. So much of what people are using Prompt Perfect for is English as a second language, but not necessarily as a tutor as much as just making their own verbiage more clear, a little bit more concise, different things like that. We find a lot of feedback from our customers around that as well. So helping them level up their own writing in English, which is really, really great. And then also in Greek, we worked with a Greek marketer who we figured out how to make the Greek as good as the English, depending on the time at which you ask it to translate into Greek. So corresponding in English and then asking it to translate into Greek at the end of the conversation, so on and so forth. There are these little hacks that you kind of have to work through together. But I think that language and natural language processing is unbelievably exciting. We've been very, very lucky to be at the intersection of that.

  • Speaker #0

    I will say anything that is done to improve communication this day and age is, I think, a boon and a bonus for all of us. So hats off to you for all of that work, too. Well, give us a very basic level of understanding about artificial intelligence. We hear the term an awful lot. It's bandied about in some very positive ways, and we'll talk about that. It's bandied about in terms of its potential for harm, perhaps. shifting the economy one direction or another. For our audience of those who may not be as current on some of this technology, give us that fundamental understanding of what artificial intelligence really is all about.

  • Speaker #1

    Sure, of course, AI or artificial intelligence as a technology, it's not brand new with the advent of chat GPT, things like recommendations on Netflix or recommendations on Spotify or on YouTube, those algorithms. are a form of artificial intelligence, even in Microsoft Word or Google Sheets or Grammarly, which sort of lives above all of those things. That ability for it to distill grammatical errors and spelling errors, those are more the grammatical side, but Grammarly has been around probably not since I was in college, but for some time now, I know that it's very, very popular. But in terms of the reason that people get, at least myself, get concerned is this term or this notion of artificial general intelligence, which is a completely autonomous agent that can think on its own and has the ability to think on its own. to make its own decisions and enact in its own best interest, which makes me immediately think of The Matrix, one of my favorite movies of all time, just farming people for their energy. Obviously, those dystopian ideals will sell movie tickets, but I think AGI is still very far away from where we are now. Of course, we do have AI that has for years beaten humans at chess or Go or any of these complex games. But what you'll notice is that those AI instantiations, one of them that can beat you in chess, they don't use that same model to beat you in Go. what they have to do is they start from square one again. And that model is unable to distill new knowledge or to deduce how to play a new game. It's not using cognition in the way that humans do. Even with my cousins, I can teach them how to play a card game because they have an understanding of how a deck of cards works. And that's not the way that these models really work. So it's all about the training data. And so we do see a lot of excitement over, let's just give it more data. That's the answer to this. We need to give it more data. And people are really focused on large language models. There's a lot more out there. There are a lot more ideas than just large language models. And I think that AGI as an idea, which I think personally is very exciting, although I do love science fiction and do find myself reading as much science fiction as I can make time for. I do think that that is a very far away concern personally. In terms of what artificial intelligence is, it's really very simple. It's just the ability for a machine. to think in a way that we recognize as human. So if you think about it like that, there are all these instances, whether it's navigations and using ways on your phone telling you where to turn as an instance of that. And then obviously, there are these Alexa and Siri, which we will leave a little bit to be desired, but are certainly an instantiation of that AI. But I think the reason that people are so excited about it is because of ChatGPT and tools like that, which are chatbots empowered by AI. So I would call them AI chatbots. But it's a very complex and complicated question. I think the chatbots are really what people think of with AI now, though. And I don't think that YouTube recommendations or Netflix recommendations, even though my mom uses them, I don't think that she would really think about that as AI. Although it technically is, it certainly is.

  • Speaker #0

    Our guest today is Charles Ferguson. Charles Ferguson co-founded Prompt Perfect. We'll have links in our show notes today for more information about Prompt Perfect, more information about Charles Ferguson and the work that the firm is doing with Prompt Perfect. I know you're introducing some... new products. I saw, for example, today that my prompt perfect was updated just recently. You guys are very busy doing all of this work. I love one of your quotes that you say, the internet is the most important invention in recent history because of the proliferation of information it enables. Should we be afraid of some of this information as it gets aggregated? There's almost a natural fear of some of the unknown, but knowing this, I don't know that fear is the right way to proceed. What's your reaction?

  • Speaker #1

    It's a great question. I do think that the democratization of access to information is an absolute net positive, and it'd be very difficult for me to concede that somehow. But I do think that there is very reasonable concern, and it's important to be concerned, and it's important to be curious enough to ask questions about these things and to be critical. I absolutely do not think of it as a utopia where everything is going to be perfect. The internet is absolutely... an open place. And that's why when you go and you just type into ChatGPT, which is trained on the internet, and the internet is inherently average, but just by virtue of the fact that there are such a large group of people contributing to it. And just in general, that's sort of the definition of what average is. So what you're going to do is when you type in a simple query, you're going to get a relatively simple response or an average response. Whereas if you tell it or prompt it in such a way and let it know, hey, you're an expert, you're a senior level programmer, keep that in mind in your response. The response very significantly becomes higher quality very quickly. So I think that when it comes to the concern around AI and that information, that misinformation, I think it's very, very reasonable. But I tend to have more of a growth mindset than a scarcity mindset, which sometimes gets me in a little bit of trouble. But I think back to my co-founder is very, very talented and his girlfriend as well at Improv and they're always saying yes and as a sort of a mantra. And I think that the ability to sort of pick things up and be intellectually curious about them and to not think of them as necessarily inherently negative. I think that the internet was founded by a group of people, which is a very positive thing. It's the centralization of ideation that I think is a really, really scary ideal. And I think that the internet is sort of the antithesis of that. And I think that AI has the potential to be an extension of that. But the internet is certainly not being replaced by AI. I think it's just being augmented by it, which is, I think, a positive thing.

  • Speaker #0

    Yeah, I tend to look at this very positively, too. I wonder if you'll tell us a little bit about what's coming out of Prompt Perfect, the group. And you touched on the new product and how it's going to be browser-based, correct me, but tell us a little bit about what's coming from Prompt Perfect.

  • Speaker #1

    So it's been 14 months of being live within ChatGPT. And so what we've done is we have sort of moved with them and ridden that whale. We're sort of one of those small fish under the large whale. And we've been sort of adapting as they adapt. And they went from plugins to GPTs. And so we graduated into a GPT, which then became a feature that is available to everybody. So everybody can use Prompt Perfect within the ChatGPT marketplace now, which is fabulous. We saw a huge uptick in usage as a result of that. But ultimately, not everybody uses ChatGPT. So we've really narrowed ourselves to being just in that one marketplace. And really, just to take a step back, what Prompt Perfect does is it just perfects prompts, it just enhances prompts. And what we've done is we've taken a lot of conversations with customers in the last year, And we've built a new product that lives within your browser instead of within ChatGPT. So anybody who uses Google Chrome might be familiar with Chrome extensions, whether it's an antivirus software or something like that. What now PromptPerfect does is it sits as a layer on top of all of these different chatbots. So whether it's Gemini from Google, or Anthropix, LLM POE, or any one of these different examples, I'm sorry, Claude, but nonetheless, any of these chatbots now PromptPerfect will sit on top of and you can perfect Perfect prompts in the same way that you can in chat GPT and prompt perfect, but also you can receive feedback. We'll give you feedback on your prompt if you're interested in learning. Something that some people want to know how to do it better themselves and others just want to hit that perfect button and get it done for them. But we give the option to get that feedback. And then the third button, which is very, very new for us is the ability to save prompts. We've heard time and time and time again from customers that they want to be able to save prompts. So whether that's in a notepad that they have on their desktop that they're constantly copying and pasting from, now that will live within their browser. And not only will it live within their browser, but when they go from ChatGPT to Claude, it follows them there. Those saved prompts will always be there within that browser for them. So we're really trying to expand the offering from PromptPerfect and making it more omnipresent. We just launched a couple of days ago. It's been very, very exciting and people really seem to like it.

  • Speaker #0

    We are going to put links so their audience can find out more about Prompt Perfect, this new product that Charles Ferguson is referring to and talking about. You guys are just doing some really amazing things. I think saving those prompts from a selfish standpoint, that's huge. Charles, I've gotten to know you a little bit over the last several months as I've used the product. You've been very available and I'm very honest with my audience about my interest and use. Thanks. of Front Perfect. Why don't you tell us a little bit more about what some other businesses might be using the tool for and in particular AI, because I think sparking some of that information for all of us might create some additional enthusiasm. I'm very enthusiastic about the product, but I think others can be too, if they kind of learn how it might apply to them.

  • Speaker #1

    I have six, seven years of experience selling to businesses. This is the first time I'd ever sold to just people. So, you know, it's a fast app that's available. to just consumers globally. So it's a very different approach. But we're able to receive feedback from folks. But ultimately, these are people with prompt engineering is new to them, and we get a lot of feedback from them. But ultimately, for some time over the summer, we spent about four months and we did a very concerted business to business play. We worked with businesses to see if they wanted services, we did some consulting, and we built some bespoke bots for people instead of them just using Prompt Perfect, which is very general purpose. We built them very specific bots. We worked with a few different teams in very different capacity. And like I'd said, I offered 15 minutes to anybody who wanted it, which was amazing over the summer. So I spoke to hundreds of people. It was very interesting.

  • Speaker #0

    I was one of those.

  • Speaker #1

    Yes.

  • Speaker #0

    The first. Yeah. Oh, well,

  • Speaker #1

    yeah. I was the first, literally the first. The diversity of applications is amazing. But ultimately, some of the things that we saw are most, so many of these conversations, the ideas are amazing, but it's impractical to build them. somebody in France was working on building something that was going to be an assistant for her grandmother to be able to understand her options in terms of community building in her neighborhood or in her local community. So I want to play tennis, but it's very specific because she doesn't want to play with a young 22-year-old boy. She wants to play with people her age. She wants to be able to play in more of a comfortable setting. And there really isn't a good tool or use case. There isn't really a great marketplace for those sorts of facts. And she was really looking to aggregate that. It's not something that we could build. The infrastructure for that sort of information just doesn't exist. So we weren't able to deliver on that. What we really saw is that the more clearly defined the problem is, the easier it is for us to implement it and to solve for that. And what that really looks like is, for example, we work with a franchise that does insect... I don't want to give it away, but they do insect application sprays on properties in the United States. And what they want to be able to do, what they need to be able to do is they take an address that comes through an inbound form, and a customer support agent then goes to Zillow or Redfin through Google and tries to find the square footage and then uses a pricing matrix with six different columns to figure out based on the size of the lot, how much the cost should be, and then also depending on what the upsell could be or whether or not it's adjacent to the water is a factor in the pricing. And what we were able to do is since that all is really just one big equation, went from that individual spending about 20 minutes typing in these things into Google to just using a GPT that we built them, a chatbot that is trained very specifically on that pricing matrix, has access to Zillow through an API. And what we're able to then do is you just type in the address. That's really all it wants is the address. And then we on the backend have not only provided it with that information from Zillow and the pricing matrix, but also then the email copy and the phone copy for this person to follow up with them. So it takes that. 20-minute process, distills it into about 90 seconds or less, depending on how long the copy is. And then that allows for that person to immediately follow up and feel a little bit less aimless is what the founder of that business said explained to us, is that there's a little bit of indecision. And it can be difficult to come up with these conversations off the top of your head, but by having a script here with all of the details already inputted, not as variables, but as the input, was extremely helpful. So things like that. Also, we found a lot of value in reporting. So for me personally, as I've learned how to do bookkeeping and work with our accountants. I've had a lot of success in our weekly reporting. which is always the same. My co-founder and I want to know the same metrics every week. It's not really that variable. And so since we know what we're looking for, we're able to just build an equation around it, so to speak. So I know that we want to know daily active users, monthly active users, the conversion rate from users, and at what point they're converting, things like that. But they're always the same questions. So if you're able to just articulate those questions and then help bridge the gap in getting them answered by providing the input that it requires each week. So the variable is the input, which is variable week over week. the questions remain the same. So you build this GPT. If you pay for ChatGPT+, which is $19 a month in the United States, I believe, building GPTs is something that comes with that. So we've built internally over 100 of these for various use cases. But the more clearly defined you can distill your problem and think about the process itself and what it is that needs to happen, the easier it is then to make that process much, much faster, because it's already clearly defined and the variable itself is what you're inputting. So I'm not sure if that's useful to the audience. I hope that it is. I hope it's not too abstract.

  • Speaker #0

    No, I think that's very helpful. I think these ways that we can winnow down what it is that we're doing from real broad, almost top of the funnel down into kind of a narrow, more narrow understanding of what it is that we really can do to make ourselves efficient. I mean, that's how I look at this in terms of my business. I so appreciate your time, Charles Ferguson. I really, I just have one broad question to end with. And it has to do with our upcoming election. We're really smack dab in the middle of that. Here we are on Halloween. The general election is coming up quickly. How do you feel about regulations with respect to AI? Because it's moving so fast. Your own business has grown so quickly and AI is just developing at a pace that it's moving so quickly that I think it's hard to regulate it. But maybe there are some needs. You hear sometimes the term guard rails. I don't know if that makes sense. really good sense, but how do you view this? And what's your perspective as an industry leader around AI?

  • Speaker #1

    I'm flattered to think that we're industry leaders. I appreciate that. But I think that for myself, the person I most relate to, this is something that people do talk a lot about. It's very interesting. And I think that as we had said earlier, I think it's easy to have a little bit of a scared mindset, especially when you see, I remember watching Star Wars as a boy. My dad watched Star Wars as a boy with C-3PO, who's an autonomous agent to themselves. So these are not new ideas and these are not new things to be afraid of.

  • Speaker #0

    When it comes to regulation, I was doing some research about the last few days ahead of this, what has been instantiated, there was an executive order, I remember that way before that executive order had been put into place, they had solicited feedback, they had asked and my co founder and I went back and forth. And we had thoughts on it and shared those thoughts. And I'm sure that there are hundreds of thousands of people who did the same there must have been just because it was a public request. And so the idea that it's an open discussion, I think is an extremely important thing. you do see some headlines and it's interesting to see them. There's a little bit of around Sam Altman, the CEO of OpenAI. Sometimes some of the things he says makes me afraid. And I wonder why. There's somebody who I really admire who seems to be somewhat of an independent thinker in this regard. His name is Mike Knoop. And he is the co-founder of Zapier, which is an amazing automation tool that's been around for over 10 years, which I've used extensively for all sorts of roles in the past. I think that we still use it internally at Prompt Perfect. I think that my co-founder uses it for some of the email marketing automations that we do. But Mike Knoop co-founded something called the ARC Prize in concert with Francois. I don't know his last name off the top of my head. I would butcher it. He's from France. But nonetheless, Mike Knoop thinks the only way to set good policy is really from looking at what can be done currently based on empirical evidence. It's more of a scientific pragmatic perspective, which I appreciate personally, as you can tell. And we should make decisions based on that empirical evidence, because we run the risk of cutting off progress too early if we're doing it based on theoretical situations that are impossible at the current state. Maybe that comes off as naive. I think that my instinct to steel man it is maybe I'm being naive and I'm not privy to all of the information that's available. But the way that we got to here with the large language models and transistors that enable that large language model and the neural networks beneath that, that was all done in the open source scientific movement through Google. and through DeepMind, through Google, all these different organizations are publishing work. Francois worked at Google when he published his paper in 2019. And so I think that the idea that we should cut these things off from public and from real experts to be able to discuss them and make them regulated based on very well-funded, self-interested to a certain extent, not because that's a negative thing as much as it's just a function of the way that the economy works. And I don't want to pass judgment on that, but I do think that in the same way that the internet was commissioned initially, by the government and then asked for it to be given back to them from these educators who were instantiating in Central and Southern California. And they had said, no, it's not something that we should centralize through this entity that requires trust. We should make this more open to people. People should be able to host their own server. I think that that's a beautiful notion. Although it might seem, again, I hate to come off as naive, but I do think that it's important to continuously evaluate where things currently stand, what is actually happening, and regulate based on that and not necessarily be reactive, but. the way that you get ahead of things is keeping experts in the room and not just listening to one or two who are the loudest or maybe the most well-funded. And I hate to cause waves or seem like I'm being contrarian. It's not my intention, but I really like the way that Mike New particulates this notion.

  • Speaker #1

    Thanks for sharing that. And no, not naive in the least, very informed and very helpful. It's so great to get this perspective, Charles Ferguson. We'd love to have you back too. This is just moving so quickly. Maybe you'll come back and... talk to us again because I know our audience is going to be really interested in this subject and as it develops certainly it's moving so quickly but please selfishly come back and talk to us a little bit about that because it's great to talk to you good luck with the rest of all of this this rollout we will have links so that our audience can find out more about Charles Ferguson and Prompt Perfect such a wonderful product and one that I just want to recommend to our audience to check out because I think it offers some vital support for somebody who is interested in this subject and wants to learn more. But Charles Ferguson, thank you for all of your work. My best to you. Have a great Halloween and certainly we'll be in touch with you.

  • Speaker #0

    Appreciate it. Thanks for the support and thanks for all you're doing.

  • Speaker #1

    My thanks, of course, to Charles Ferguson for joining us today. You'll find links in our notes for more information about Charles Ferguson, all of his details, including information about his company, Prompt Perfect. My thanks, of course, to our executive editor, Sam Henniger, for all of his work. My thanks always to our audience on radio and podcast. Please be well, be safe, and let's talk about better. The Not Old Better Show on radio and podcast. Thanks, everybody. We will see you next week.

  • Speaker #2

    Thanks for joining us this week on The Not Old Better Show. To find out more about all of today's stories or to view our extensive back catalog of previous shows. simply visit notold-better.com. Join us again next time as we deep dive into some of the most fascinating real-life stories from across the world, all focused on this wonderful experience of getting better, not just older. Let's talk about better. The Not Old Better Show.

  • Speaker #1

    I won't find a thing. Please check out our website for this episode and all episodes at notold-better.com or subscribe to the podcast on Apple Podcasts. And be sure to check out your local radio stations to find out more about The Not Old Better Show on podcast and radio. You can find us all over social media. Our Twitter feed is notoldbetter and we're on Instagram at notoldbetter2. The Not Old Better Show is a production of NOBS Studios. I'm Paul Vogelsang and I hope you'll join me again next time to talk about... Better. The Not Old Better Show. Thanks, everybody. We'll see you next week.

Description

Welcome to The Not Old – Better Show, Technology Interview Series on radio and podcast, where we explore ideas that matter to those who are 50 and older, encouraging discovery, adventure, and learning. 

Today’s episode promises to be one of our most insightful yet, as we delve into a topic shaping our world at lightning speed: artificial intelligence. 


Our guest is Charles Ferguson, the co-founder of PromptPerfect, a pioneering company in AI innovation that is making waves with its cutting-edge tools designed to enhance how we interact with AI. Charles has been at the forefront of developing AI solutions that are not only powerful but also accessible, especially for small businesses and individuals who might just be stepping into the AI landscape.


In this conversation, we’re going to cover a lot of ground—from what AI actually is and how it can be useful, to whether we should be wary of its rapid rise. Charles will offer insights on how tools like PromptPerfect work, why he believes AI is a vital tool for growth, and how you, our listeners, can embrace AI in your daily lives, whether for personal projects or small business ventures.

With AI becoming more prevalent in everything from healthcare to marketing, this episode is for everyone who’s curious, cautious, or just eager to understand where AI is headed and how it can be a force for good—especially for those who might feel left behind in the digital age. So, sit back, enjoy the conversation, and let’s learn together.


Join me in welcoming our guest today, co founder of PromptPerfect, Charles Ferguson

My thanks to all, Charles Ferguson, Sam Heningerand our wonderful audience on radio and podcast.  Be well, be safe and let’s talk about better.  The Not Old Better Show.


Hosted by Ausha. See ausha.co/privacy-policy for more information.

Transcription

  • Speaker #0

    Welcome to the Not Old Better Show, the show covering all things health, wellness, culture, and more. The show for all of us who aren't old, we're better. Each week, we'll interview superstars, experts, and ordinary people doing extraordinary things, all related to this wonderful experience of getting better, not older. Now, here's your host, the award-winning Paul Vogelzang. Welcome to the Not Old Better Show Technology. interview series on radio and podcast where we explore ideas that matter to those who are 50 and older, encouraging discovery, adventure, and learning. I'm Paul Vogelsang, and today's episode promises to be one of our most insightful yet as we delve into a topic shaping our world. It is everywhere right now, and it's doing this at lightning speed. I'm talking, of course, of artificial intelligence, AI. Our guest... is Charles Ferguson, the co-founder of Prompt Perfect, a pioneering company in AI innovation that is making waves with its cutting-edge tools designed to enhance how we interact with AI, not just for businesses, but for individuals, for small businesses, really for everyone. And you're going to hear Charles talk about what it's been like to be at the forefront of developing AI solutions that are not only powerful, but also accessible to all of us, especially for small businesses and individuals who might just be stepping into the AI landscape. In this conversation with Charles Ferguson, we are going to cover a lot of ground from what AI actually is, how it can be useful, to whether we should be wary of its rapid rise. Charles Ferguson will offer some insights today on how tools like PromptPerfect work, why he believes AI is a vital tool for growth as a society and how you, our listeners, can embrace AI in your daily lives, whether for personal projects or small business ventures. You're going to love this interview with AI becoming more prevalent in everything from healthcare to marketing. This episode is for everyone who's curious, cautious, or just eager to understand where AI is headed and how it can be a force for good, especially for those who might feel left behind in the digital age. So sit back, enjoy our conversation with Charles Ferguson. Let's learn together. Join me in welcoming our guest today, co-founder of Prompt Perfect, Charles Ferguson. Charles Ferguson, welcome to the program.

  • Speaker #1

    Thanks, Paul. I appreciate it. We're very, very appreciative of the opportunity to chat.

  • Speaker #0

    Well, thank you. I am excited about that too. I think first and foremost, happy Halloween. We're talking on Halloween. And so I want to make sure to extend those wishes. My wife is going to come home after being away for a few days, so I don't think we're going to have any trick-or-treaters. How about you guys? Are you planning for some?

  • Speaker #1

    No, I think of this more of the eve of my girlfriend's birthday, more so than Halloween, to be candid. But there's certainly a lot of people walking around in costume outside of New York City right now.

  • Speaker #0

    Yep, yep. Well, again, welcome and congratulations on all you are doing with Prompt Perfect. We're going to talk about that and AI, the subject of artificial intelligence today. I think that's on everybody's minds. And I want to get into that with you. But again, first and foremost, congrats on this wonderful tool. I use it. I'm going to talk about it a little bit personally, but I love what you guys are doing.

  • Speaker #1

    Yes, we've been very, very fortunate with the timing. It was sort of built out of necessity for ourselves, as we'll go into, I'm sure. And the reception has been incredible. It's just so much fun to be able to work with people all over the world, over 130 countries represented in our customer base of hundreds of thousands of people. It's been used over a million and a half times. And so we're glad that anybody finds it useful, but we also use it daily. And it's been a long 18 months of a lot of iteration and it's gotten a lot better. And we're excited also to be launching a new product this week, very similar. It's omnipresent. It isn't confined to ChatGPT. It's within your web browser, within Google Chrome.

  • Speaker #0

    I'm a user. I love the tool. It has been so helpful for me. This is purely an editorial interview. I can talk and wax. poetically on this service that you offer and this fantastic means for me to do work as a small business person. But I think it's important to point out that I think our audience is going to be starting a few steps back in terms of their understanding of what artificial intelligence represents. And so I wonder if you could tell us about the inspiration for starting Prompt Perfect. I know a little bit about the story, but it is this amazing tool. Did you start to see some gaps in some of the development that was going on around artificial intelligence?

  • Speaker #1

    Yes, I need to make sure first and foremost, it's extremely clear that my co-founder is the ideator here, my closest friends from college. We lived together for a long time and we spent about 10 years working on other things completely separately, but always remain really, really close. When ChatGPT 3 and then 3.5 had come out and there was a front end introduced and that was in 2023. We were playing with the tool a little bit, but it was really my co-founder who was trying to assist himself with this new tool to make his job easier. He's a copywriter at the time, and he found a lot of inspiration with providing examples of copy that he loved, and then getting iterations of things that were more applicable to him based on context he was providing. And what he was really teaching himself is prompt engineering before reading any papers on it. So just understanding the correlation between the input you provide it as being directly correlated to the quality of the output you receive back, which is very different than Google search or internet browsing. They're very, very different. And so the skill set is different. And that's something that he just intuited himself almost two years ago now. And he built it out of hackathon over a weekend. He had been given that opportunity. And then subsequently, ChatGPT introduced this beta feature plugins, and it was featured as one of the top eight plugins. And that was before we really had good analytics or any sort of understanding of where this might go. And overnight, within three weeks, there were thousands and thousands of users. And it was very interesting to see. We just sort of took it as it came. And he had asked me to be his co-founder around that time. And we had a lot of bookkeeping to do. And that's neither of our core competencies, but I figured it out. And we've come a long way in that time. It really came from a necessity of trying to figure out how do we use these tools effectively and realizing this is not the same as our past 15, 20 years of experience working and searching on the internet. This is a very new skill set. It's all about understanding the context of the question you're trying to ask. providing as much context and detail as possible. And that's going to provide better responses always. It really came from us trying to get the most out of the platform, getting the most out of ChatGPT at the time, without having to spend too much time writing out the details. And so what Prompt Perfect does is it takes the sentence or the prompt that you provided, and it enhances it with more context and more detail. And it shows you that more detailed prompt and then it'll show you the response that that more detailed and that more context provides to you. So it enhances prompts for people. We assumed people cared about how to learn how to prompt engineer, but what we've learned from launching eight products is they're much more interested in having that activity done for them more so than learning how to do it. I think it's a little bit easier if it just sort of does it for you. And maybe it'll show you its work, but ultimately teaching somebody to prompt engineer isn't nearly as much fun or what we found is useful economically for people. They just want it done for themselves. So that's what we do is we just provide a service that enhances your prompts for you automatically.

  • Speaker #0

    Brilliant. Again, congratulations to you and your co-founder, because I'm always amazed at the development of businesses and how they really get started. And I want to drill down for just a moment and make sure that our audience understands that Prompt Perfect serves as a tool, as you describe it, is really a companion to ChatGPT. And in order to really maximize the sheer efforts that ChatGPT can bring. prompt perfect allows you to enter a string of words that are related but you don't necessarily have to share the sheer volume of material that you might within chat gpt because the technology within prompt perfect supplements your own work to give you this tool that really generates this fantastic result that you're getting from the artificial intelligence yes

  • Speaker #1

    without a doubt i think if you're using prompt perfect but you're very well versed in prompt engineering, you might not see that significant of an enhancement in your prompt. So if you provide it with five paragraphs and 300 words, and it's extremely detailed, we're going to be able to automatically enhance that maybe slightly or change some syntax. And we really try to optimize for the model. And so we're trying to think about how the model is going to receive that. So we're iterating on that. So it could see some improvements. But the time when we see the most significant improvement is in two, three, four, five lines, stream of consciousness. with misspellings, whatever it might be, just people being people. And what we are going to be able to do there is enhance that and show you how we've enhanced it, most importantly to me, so that you can then ultimately employ that strategy yourself. But the idea of the time savings of not having to figure out how to word these things or provide that context, we've seen that our customers find it very, very valuable.

  • Speaker #0

    Yeah. And as I say, I'm a customer. I'm a small businessman. I am just a tiny, tiny little business. It's really me. And I've since expanded. my business a little bit so that I work with an editor. I'm working with another individual who helps kind of keep me on track in terms of schedule. I really look at Prompt Perfect as another level of staffing to my business. It helps me that much. It helps me refine questions. It helps me with writing passages for the program. It helps me with social media posts. It really does so much to supplement me. Yeah. And so selfishly, I just think this tool is all about empowering people. Was that kind of your thinking in terms of empowering individuals and small businesses? Because I will tell you again, that's what it's doing for me.

  • Speaker #1

    That's fabulous. We love to hear that. I feel I'm gushing. I'm not sure how to explain it more than that. But Eric and myself, we both immediately following undergrad, we graduated in 2013. I'd done the Peace Corps for a couple of years. He was in AmeriCorps as a teacher. And so we're both very service oriented. Well, I don't profess to say that Eric had thought of this because he was doing it out of a greater good. It really was a self necessity. And we're a two person business as well. And we use it constantly. And we have a lot of feedback from that. We've spoken to thousands. High hundreds at the very least. I haven't aggregated all of the numbers in terms of the user feedback and the way that we correspond with folks, but I wouldn't be terribly surprised if it's over a thousand after 18 months. The feedback is resoundingly positive. We did, actually, for four months very concertedly double-click into trying to work with small businesses specifically. I had hundreds of conversations with different small business owners, mostly small businesses, and then we ultimately built some bespoke products for several of those small businesses. It's a common thread. AI is something that we internally call them interns. They're not somebody you're going to hand a task over to entirely. It's not going to get it over the finish line necessarily or create hundreds of blog posts for you without any sort of thought. There should be, and I think there will be for the foreseeable future, a human in the loop, so to speak. And also just for quality control and making sure that it's in your voice and your tone and things are done accurately. But it is a massive time saver for all sorts of small businesses. We're very fortunate. And we've also, one of the things is working with small businesses was a surprise, but also we've been very surprised, or at least I was, I should say, by the international adoption. So I'd said earlier, 130 some odd countries, 133, 134 countries. So much of what people are using Prompt Perfect for is English as a second language, but not necessarily as a tutor as much as just making their own verbiage more clear, a little bit more concise, different things like that. We find a lot of feedback from our customers around that as well. So helping them level up their own writing in English, which is really, really great. And then also in Greek, we worked with a Greek marketer who we figured out how to make the Greek as good as the English, depending on the time at which you ask it to translate into Greek. So corresponding in English and then asking it to translate into Greek at the end of the conversation, so on and so forth. There are these little hacks that you kind of have to work through together. But I think that language and natural language processing is unbelievably exciting. We've been very, very lucky to be at the intersection of that.

  • Speaker #0

    I will say anything that is done to improve communication this day and age is, I think, a boon and a bonus for all of us. So hats off to you for all of that work, too. Well, give us a very basic level of understanding about artificial intelligence. We hear the term an awful lot. It's bandied about in some very positive ways, and we'll talk about that. It's bandied about in terms of its potential for harm, perhaps. shifting the economy one direction or another. For our audience of those who may not be as current on some of this technology, give us that fundamental understanding of what artificial intelligence really is all about.

  • Speaker #1

    Sure, of course, AI or artificial intelligence as a technology, it's not brand new with the advent of chat GPT, things like recommendations on Netflix or recommendations on Spotify or on YouTube, those algorithms. are a form of artificial intelligence, even in Microsoft Word or Google Sheets or Grammarly, which sort of lives above all of those things. That ability for it to distill grammatical errors and spelling errors, those are more the grammatical side, but Grammarly has been around probably not since I was in college, but for some time now, I know that it's very, very popular. But in terms of the reason that people get, at least myself, get concerned is this term or this notion of artificial general intelligence, which is a completely autonomous agent that can think on its own and has the ability to think on its own. to make its own decisions and enact in its own best interest, which makes me immediately think of The Matrix, one of my favorite movies of all time, just farming people for their energy. Obviously, those dystopian ideals will sell movie tickets, but I think AGI is still very far away from where we are now. Of course, we do have AI that has for years beaten humans at chess or Go or any of these complex games. But what you'll notice is that those AI instantiations, one of them that can beat you in chess, they don't use that same model to beat you in Go. what they have to do is they start from square one again. And that model is unable to distill new knowledge or to deduce how to play a new game. It's not using cognition in the way that humans do. Even with my cousins, I can teach them how to play a card game because they have an understanding of how a deck of cards works. And that's not the way that these models really work. So it's all about the training data. And so we do see a lot of excitement over, let's just give it more data. That's the answer to this. We need to give it more data. And people are really focused on large language models. There's a lot more out there. There are a lot more ideas than just large language models. And I think that AGI as an idea, which I think personally is very exciting, although I do love science fiction and do find myself reading as much science fiction as I can make time for. I do think that that is a very far away concern personally. In terms of what artificial intelligence is, it's really very simple. It's just the ability for a machine. to think in a way that we recognize as human. So if you think about it like that, there are all these instances, whether it's navigations and using ways on your phone telling you where to turn as an instance of that. And then obviously, there are these Alexa and Siri, which we will leave a little bit to be desired, but are certainly an instantiation of that AI. But I think the reason that people are so excited about it is because of ChatGPT and tools like that, which are chatbots empowered by AI. So I would call them AI chatbots. But it's a very complex and complicated question. I think the chatbots are really what people think of with AI now, though. And I don't think that YouTube recommendations or Netflix recommendations, even though my mom uses them, I don't think that she would really think about that as AI. Although it technically is, it certainly is.

  • Speaker #0

    Our guest today is Charles Ferguson. Charles Ferguson co-founded Prompt Perfect. We'll have links in our show notes today for more information about Prompt Perfect, more information about Charles Ferguson and the work that the firm is doing with Prompt Perfect. I know you're introducing some... new products. I saw, for example, today that my prompt perfect was updated just recently. You guys are very busy doing all of this work. I love one of your quotes that you say, the internet is the most important invention in recent history because of the proliferation of information it enables. Should we be afraid of some of this information as it gets aggregated? There's almost a natural fear of some of the unknown, but knowing this, I don't know that fear is the right way to proceed. What's your reaction?

  • Speaker #1

    It's a great question. I do think that the democratization of access to information is an absolute net positive, and it'd be very difficult for me to concede that somehow. But I do think that there is very reasonable concern, and it's important to be concerned, and it's important to be curious enough to ask questions about these things and to be critical. I absolutely do not think of it as a utopia where everything is going to be perfect. The internet is absolutely... an open place. And that's why when you go and you just type into ChatGPT, which is trained on the internet, and the internet is inherently average, but just by virtue of the fact that there are such a large group of people contributing to it. And just in general, that's sort of the definition of what average is. So what you're going to do is when you type in a simple query, you're going to get a relatively simple response or an average response. Whereas if you tell it or prompt it in such a way and let it know, hey, you're an expert, you're a senior level programmer, keep that in mind in your response. The response very significantly becomes higher quality very quickly. So I think that when it comes to the concern around AI and that information, that misinformation, I think it's very, very reasonable. But I tend to have more of a growth mindset than a scarcity mindset, which sometimes gets me in a little bit of trouble. But I think back to my co-founder is very, very talented and his girlfriend as well at Improv and they're always saying yes and as a sort of a mantra. And I think that the ability to sort of pick things up and be intellectually curious about them and to not think of them as necessarily inherently negative. I think that the internet was founded by a group of people, which is a very positive thing. It's the centralization of ideation that I think is a really, really scary ideal. And I think that the internet is sort of the antithesis of that. And I think that AI has the potential to be an extension of that. But the internet is certainly not being replaced by AI. I think it's just being augmented by it, which is, I think, a positive thing.

  • Speaker #0

    Yeah, I tend to look at this very positively, too. I wonder if you'll tell us a little bit about what's coming out of Prompt Perfect, the group. And you touched on the new product and how it's going to be browser-based, correct me, but tell us a little bit about what's coming from Prompt Perfect.

  • Speaker #1

    So it's been 14 months of being live within ChatGPT. And so what we've done is we have sort of moved with them and ridden that whale. We're sort of one of those small fish under the large whale. And we've been sort of adapting as they adapt. And they went from plugins to GPTs. And so we graduated into a GPT, which then became a feature that is available to everybody. So everybody can use Prompt Perfect within the ChatGPT marketplace now, which is fabulous. We saw a huge uptick in usage as a result of that. But ultimately, not everybody uses ChatGPT. So we've really narrowed ourselves to being just in that one marketplace. And really, just to take a step back, what Prompt Perfect does is it just perfects prompts, it just enhances prompts. And what we've done is we've taken a lot of conversations with customers in the last year, And we've built a new product that lives within your browser instead of within ChatGPT. So anybody who uses Google Chrome might be familiar with Chrome extensions, whether it's an antivirus software or something like that. What now PromptPerfect does is it sits as a layer on top of all of these different chatbots. So whether it's Gemini from Google, or Anthropix, LLM POE, or any one of these different examples, I'm sorry, Claude, but nonetheless, any of these chatbots now PromptPerfect will sit on top of and you can perfect Perfect prompts in the same way that you can in chat GPT and prompt perfect, but also you can receive feedback. We'll give you feedback on your prompt if you're interested in learning. Something that some people want to know how to do it better themselves and others just want to hit that perfect button and get it done for them. But we give the option to get that feedback. And then the third button, which is very, very new for us is the ability to save prompts. We've heard time and time and time again from customers that they want to be able to save prompts. So whether that's in a notepad that they have on their desktop that they're constantly copying and pasting from, now that will live within their browser. And not only will it live within their browser, but when they go from ChatGPT to Claude, it follows them there. Those saved prompts will always be there within that browser for them. So we're really trying to expand the offering from PromptPerfect and making it more omnipresent. We just launched a couple of days ago. It's been very, very exciting and people really seem to like it.

  • Speaker #0

    We are going to put links so their audience can find out more about Prompt Perfect, this new product that Charles Ferguson is referring to and talking about. You guys are just doing some really amazing things. I think saving those prompts from a selfish standpoint, that's huge. Charles, I've gotten to know you a little bit over the last several months as I've used the product. You've been very available and I'm very honest with my audience about my interest and use. Thanks. of Front Perfect. Why don't you tell us a little bit more about what some other businesses might be using the tool for and in particular AI, because I think sparking some of that information for all of us might create some additional enthusiasm. I'm very enthusiastic about the product, but I think others can be too, if they kind of learn how it might apply to them.

  • Speaker #1

    I have six, seven years of experience selling to businesses. This is the first time I'd ever sold to just people. So, you know, it's a fast app that's available. to just consumers globally. So it's a very different approach. But we're able to receive feedback from folks. But ultimately, these are people with prompt engineering is new to them, and we get a lot of feedback from them. But ultimately, for some time over the summer, we spent about four months and we did a very concerted business to business play. We worked with businesses to see if they wanted services, we did some consulting, and we built some bespoke bots for people instead of them just using Prompt Perfect, which is very general purpose. We built them very specific bots. We worked with a few different teams in very different capacity. And like I'd said, I offered 15 minutes to anybody who wanted it, which was amazing over the summer. So I spoke to hundreds of people. It was very interesting.

  • Speaker #0

    I was one of those.

  • Speaker #1

    Yes.

  • Speaker #0

    The first. Yeah. Oh, well,

  • Speaker #1

    yeah. I was the first, literally the first. The diversity of applications is amazing. But ultimately, some of the things that we saw are most, so many of these conversations, the ideas are amazing, but it's impractical to build them. somebody in France was working on building something that was going to be an assistant for her grandmother to be able to understand her options in terms of community building in her neighborhood or in her local community. So I want to play tennis, but it's very specific because she doesn't want to play with a young 22-year-old boy. She wants to play with people her age. She wants to be able to play in more of a comfortable setting. And there really isn't a good tool or use case. There isn't really a great marketplace for those sorts of facts. And she was really looking to aggregate that. It's not something that we could build. The infrastructure for that sort of information just doesn't exist. So we weren't able to deliver on that. What we really saw is that the more clearly defined the problem is, the easier it is for us to implement it and to solve for that. And what that really looks like is, for example, we work with a franchise that does insect... I don't want to give it away, but they do insect application sprays on properties in the United States. And what they want to be able to do, what they need to be able to do is they take an address that comes through an inbound form, and a customer support agent then goes to Zillow or Redfin through Google and tries to find the square footage and then uses a pricing matrix with six different columns to figure out based on the size of the lot, how much the cost should be, and then also depending on what the upsell could be or whether or not it's adjacent to the water is a factor in the pricing. And what we were able to do is since that all is really just one big equation, went from that individual spending about 20 minutes typing in these things into Google to just using a GPT that we built them, a chatbot that is trained very specifically on that pricing matrix, has access to Zillow through an API. And what we're able to then do is you just type in the address. That's really all it wants is the address. And then we on the backend have not only provided it with that information from Zillow and the pricing matrix, but also then the email copy and the phone copy for this person to follow up with them. So it takes that. 20-minute process, distills it into about 90 seconds or less, depending on how long the copy is. And then that allows for that person to immediately follow up and feel a little bit less aimless is what the founder of that business said explained to us, is that there's a little bit of indecision. And it can be difficult to come up with these conversations off the top of your head, but by having a script here with all of the details already inputted, not as variables, but as the input, was extremely helpful. So things like that. Also, we found a lot of value in reporting. So for me personally, as I've learned how to do bookkeeping and work with our accountants. I've had a lot of success in our weekly reporting. which is always the same. My co-founder and I want to know the same metrics every week. It's not really that variable. And so since we know what we're looking for, we're able to just build an equation around it, so to speak. So I know that we want to know daily active users, monthly active users, the conversion rate from users, and at what point they're converting, things like that. But they're always the same questions. So if you're able to just articulate those questions and then help bridge the gap in getting them answered by providing the input that it requires each week. So the variable is the input, which is variable week over week. the questions remain the same. So you build this GPT. If you pay for ChatGPT+, which is $19 a month in the United States, I believe, building GPTs is something that comes with that. So we've built internally over 100 of these for various use cases. But the more clearly defined you can distill your problem and think about the process itself and what it is that needs to happen, the easier it is then to make that process much, much faster, because it's already clearly defined and the variable itself is what you're inputting. So I'm not sure if that's useful to the audience. I hope that it is. I hope it's not too abstract.

  • Speaker #0

    No, I think that's very helpful. I think these ways that we can winnow down what it is that we're doing from real broad, almost top of the funnel down into kind of a narrow, more narrow understanding of what it is that we really can do to make ourselves efficient. I mean, that's how I look at this in terms of my business. I so appreciate your time, Charles Ferguson. I really, I just have one broad question to end with. And it has to do with our upcoming election. We're really smack dab in the middle of that. Here we are on Halloween. The general election is coming up quickly. How do you feel about regulations with respect to AI? Because it's moving so fast. Your own business has grown so quickly and AI is just developing at a pace that it's moving so quickly that I think it's hard to regulate it. But maybe there are some needs. You hear sometimes the term guard rails. I don't know if that makes sense. really good sense, but how do you view this? And what's your perspective as an industry leader around AI?

  • Speaker #1

    I'm flattered to think that we're industry leaders. I appreciate that. But I think that for myself, the person I most relate to, this is something that people do talk a lot about. It's very interesting. And I think that as we had said earlier, I think it's easy to have a little bit of a scared mindset, especially when you see, I remember watching Star Wars as a boy. My dad watched Star Wars as a boy with C-3PO, who's an autonomous agent to themselves. So these are not new ideas and these are not new things to be afraid of.

  • Speaker #0

    When it comes to regulation, I was doing some research about the last few days ahead of this, what has been instantiated, there was an executive order, I remember that way before that executive order had been put into place, they had solicited feedback, they had asked and my co founder and I went back and forth. And we had thoughts on it and shared those thoughts. And I'm sure that there are hundreds of thousands of people who did the same there must have been just because it was a public request. And so the idea that it's an open discussion, I think is an extremely important thing. you do see some headlines and it's interesting to see them. There's a little bit of around Sam Altman, the CEO of OpenAI. Sometimes some of the things he says makes me afraid. And I wonder why. There's somebody who I really admire who seems to be somewhat of an independent thinker in this regard. His name is Mike Knoop. And he is the co-founder of Zapier, which is an amazing automation tool that's been around for over 10 years, which I've used extensively for all sorts of roles in the past. I think that we still use it internally at Prompt Perfect. I think that my co-founder uses it for some of the email marketing automations that we do. But Mike Knoop co-founded something called the ARC Prize in concert with Francois. I don't know his last name off the top of my head. I would butcher it. He's from France. But nonetheless, Mike Knoop thinks the only way to set good policy is really from looking at what can be done currently based on empirical evidence. It's more of a scientific pragmatic perspective, which I appreciate personally, as you can tell. And we should make decisions based on that empirical evidence, because we run the risk of cutting off progress too early if we're doing it based on theoretical situations that are impossible at the current state. Maybe that comes off as naive. I think that my instinct to steel man it is maybe I'm being naive and I'm not privy to all of the information that's available. But the way that we got to here with the large language models and transistors that enable that large language model and the neural networks beneath that, that was all done in the open source scientific movement through Google. and through DeepMind, through Google, all these different organizations are publishing work. Francois worked at Google when he published his paper in 2019. And so I think that the idea that we should cut these things off from public and from real experts to be able to discuss them and make them regulated based on very well-funded, self-interested to a certain extent, not because that's a negative thing as much as it's just a function of the way that the economy works. And I don't want to pass judgment on that, but I do think that in the same way that the internet was commissioned initially, by the government and then asked for it to be given back to them from these educators who were instantiating in Central and Southern California. And they had said, no, it's not something that we should centralize through this entity that requires trust. We should make this more open to people. People should be able to host their own server. I think that that's a beautiful notion. Although it might seem, again, I hate to come off as naive, but I do think that it's important to continuously evaluate where things currently stand, what is actually happening, and regulate based on that and not necessarily be reactive, but. the way that you get ahead of things is keeping experts in the room and not just listening to one or two who are the loudest or maybe the most well-funded. And I hate to cause waves or seem like I'm being contrarian. It's not my intention, but I really like the way that Mike New particulates this notion.

  • Speaker #1

    Thanks for sharing that. And no, not naive in the least, very informed and very helpful. It's so great to get this perspective, Charles Ferguson. We'd love to have you back too. This is just moving so quickly. Maybe you'll come back and... talk to us again because I know our audience is going to be really interested in this subject and as it develops certainly it's moving so quickly but please selfishly come back and talk to us a little bit about that because it's great to talk to you good luck with the rest of all of this this rollout we will have links so that our audience can find out more about Charles Ferguson and Prompt Perfect such a wonderful product and one that I just want to recommend to our audience to check out because I think it offers some vital support for somebody who is interested in this subject and wants to learn more. But Charles Ferguson, thank you for all of your work. My best to you. Have a great Halloween and certainly we'll be in touch with you.

  • Speaker #0

    Appreciate it. Thanks for the support and thanks for all you're doing.

  • Speaker #1

    My thanks, of course, to Charles Ferguson for joining us today. You'll find links in our notes for more information about Charles Ferguson, all of his details, including information about his company, Prompt Perfect. My thanks, of course, to our executive editor, Sam Henniger, for all of his work. My thanks always to our audience on radio and podcast. Please be well, be safe, and let's talk about better. The Not Old Better Show on radio and podcast. Thanks, everybody. We will see you next week.

  • Speaker #2

    Thanks for joining us this week on The Not Old Better Show. To find out more about all of today's stories or to view our extensive back catalog of previous shows. simply visit notold-better.com. Join us again next time as we deep dive into some of the most fascinating real-life stories from across the world, all focused on this wonderful experience of getting better, not just older. Let's talk about better. The Not Old Better Show.

  • Speaker #1

    I won't find a thing. Please check out our website for this episode and all episodes at notold-better.com or subscribe to the podcast on Apple Podcasts. And be sure to check out your local radio stations to find out more about The Not Old Better Show on podcast and radio. You can find us all over social media. Our Twitter feed is notoldbetter and we're on Instagram at notoldbetter2. The Not Old Better Show is a production of NOBS Studios. I'm Paul Vogelsang and I hope you'll join me again next time to talk about... Better. The Not Old Better Show. Thanks, everybody. We'll see you next week.

Share

Embed

You may also like