Stop Digging!

Nancy: Hello everyone. This is Nancy and Susan. We're here with another episode of _Stop Digging!_ We're trying to help you get out of the hole that you're in.

Today we have a great topic, one that everyone's talking about, and it's Artificial Intelligence.

Susan has been doing some research on this very thing. And so tell us, Susan, what have you found out?

Susan: Artificial Intelligence always takes me into science fiction and, right away, I'm thinking sentient robots and evil computers that take over human function. And I know I'm not alone because there's a lot of fear and trepidation among employees because they are afraid that their jobs may be replaced by a robot.

And it's not unlike the fear that happened with machines replacing workforces. It changes the workforce and people are afraid because they don't know how to prepare for that or what they need to do to retool.

First of all, what I did when I started looking into this in preparation for this episode was I thought, maybe I don't know what Artificial Intelligence actually means. Maybe I'm assuming that it means sentient robots who will enslave me. I found a great website and they focus on this a lot as well as other technology, and it's called the Authority Hacker, AuthorityHacker.com. And they have a podcast that discusses things and they have fantastic research on the statistics of AI.

And I'm just going to go through a few of them to set context for this conversation.

- 35% of businesses have already adopted AI.

- The fastest growing industry that's adopted AI is actually the global supply chain. And, in a way it's surprising, in a way it's not, because we just came through two years of COVID shutdown and disrupted global supply. But if machines could take over in, in that area, I can see how that might benefit them. So, they estimate that about 55% of all global supply chain networks will be actually AI by 2025. It is very high.

- Nine out of ten organizations do support AI for competitive advantage because the global market is competitive right now, especially with technology. You're not just competing locally, you're competing globally.

- 81.6 % of digital marker marketers think that content writer jobs are at risk because of AI.

Now, if you're like me, I'm getting things all the time about ChatGPT to write articles and generate content because online, content is King. What does that do for artists, writers, people who want to put their creative thoughts down and AI is generating articles for them? So, there are concerns.

Some of the things that caught my eye because of our nonprofit working with employees and businesses:

- 32.9% of businesses have already replaced human tasks with AI solutions.

- By 2025, AI is estimated to eliminate 85 million jobs. Now, they're also estimating that it's going to create 97 million new ones, resulting in a net gain of 12 million jobs. The key issue is what are those jobs? And what are the skills that you're going to need to have those jobs? So that's not clear yet.

Susan: By 2030, they estimate that intelligent robots may replace 30% of the global workforce. Let me repeat that: by 2030 intelligent robots may replace 30 percent of the global workforce. That is phenomenal. That's crazy in my opinion. But that's what they're saying. And they say that 375 million people may need to change careers by 2030.

Let's talk about AI. Let's figure out what it is.

It's not always translating to the highest level of AI, which is a sentient robot who enslaves me. There are three other kinds of AI and most of those are what's in use right now.

One is a reactive machine, machines with limited memory. There's AI with theory of mind, which means it learns it within a set parameter. And then there's the self-aware.

Now, the definition, when I looked it up, is that it's any machine or any software that mimics human intelligence and Machine Learning is where it optimizes a system based on the data that is being gathered while it's operating.

The Deep Learning is where, think of our brain where. It has established shortcut rules that integrates all of the data into multiple categories. So, it's generating impact. Deep thought on new data in multiple areas. In coding words, maybe that would be creating a shortcut that does the same thing without having to go through the detailed code for it to do it. It just intuitively knows.

So, I felt a little bit better knowing the definition and knowing that there's different types of AI, because I feel like you can still use AI for a number of different applications without being enslaved by it.

We are excited to have this guest because he is using AI for good. He has no plans, from what I understand, of enslaving us. So, thank you for that. And Nancy, who do we have?

Nancy: Today our guest is Theron Loven. He's an entrepreneur who has been working in this space for about a year and a half who comes to us from Argentina. Welcome, Theron.

Lukas: Thank you so much.

Nancy: Theron, we're really happy to have you here with us because we have a lot of questions. So now, as Susan told us all those statistics, I know you've got a lot of things going on in your mind at the moment, but before we get into that, can you tell us, first of all, how in the world did you get into this space, into this industry and what are you doing in your business?

Lukas: It was not my plan to get into this business. Not at all. From the start, I had a whole other career in mind. I was living in San Diego. I was studying to become a pilot, not all into the technology that I am right now. I've always been fascinated by it though. And I always liked design actually itself. That is the fun part for me.

While living there, I started to use the websites for fun. And it just grew into another thing. I started with website that turned into more difficult web3.0 kind of stuff when I moved to Argentina and we started to build our portfolio. Which then led to, when the whole web3.0 industry died, all crypto and all NFTs and all these other things that exist, yeah, it was almost like a huge scam. And then AI came, and I think it really was in November, December, last year, ChatGPT just blossomed. It was bigger than anything else.

That's when we noticed, okay. We already started doing some AI before then, but once that happened and we announced it to our clients, it just started exploding the inbox with everything from chatbots to, it's mostly chatbots actually, but that to create images, videos, a bunch of stuff.

It's been a really fun journey. It is a little scary what you're saying, Susan, but it's also fun. For me, it's bigger than the internet, actually, I would

say.

Susan: Just remember when they're ready to take over that I'm on their side.

Lukas: Yeah.

Susan: Just if you're listening,

Lukas: If they're listening, yeah. No, I do think we're far away from, like, a Doomsday scenario like that, but they're getting better, the AI. And they're getting better pretty fast. So, it's really exciting to see.

Nancy: So, Susan brought up quite a few things. I thought her comment about, 55% of the global supply chain will be converted over in just a couple of years, it doesn't surprise me, but, I think when someone says that they'll be using AI, I don't think that there's going to be this complete takeover of their company and that all of their processes are going to be handled by AI.

As a designer, someone who's working in this field, what really is happening? Because it's only been around, like you said, for a short time. What do you see the areas that businesses are using AI?

Lukas: We use it daily too. And I feel like rather than replacing, it's more helping. For example, if you want, I do think that you still need designers and programmers and all that stuff, because right now the AI cannot generate an Illustrator file, or it cannot give you a bunch of other stuff, but it's a good way to get the idea.

So we usually go into ChatGPT. Okay, create a logo. Blah, blah, blah, blah, blah. And you can just keep asking it, right? And it gives you a bunch of different options and you get an idea of what you want, and then we hand that over to the designer who can actually maybe make it a little more unique or provide it in all the different files that we want and all those stuff. And the same with coding.

Like, we don't go in and say, "Okay, do this whole website." but it's really good at solving these kinds of problems. It's really good at problem solving. If it's a very specific issue or bug or something in the code that you need help with, it saves a ton of time than going into all these different websites or browsing the web because he does it in a few seconds.

And it's mostly accurate. Sometimes it just takes you down a rabbit hole and it never finds a solution, but most of the time it's really good. So, right now I see it more as a tool than replacing us. Because we know this ourselves. When we go into a website and you go to a chatbot, you notice almost right away that it's not a real person, right?

And it's frustrating. They're not really there yet.

Nancy: I'm really glad you brought that up because that was where I was thinking things were happening, because especially in the health industry, for example, when a person's online and they're doing their telehealth visit, oftentimes there's a chat box or some kind of thing.

They have to type in their information and it says, "Waiting." You're in the virtual office waiting For your provider and I know that's chatbot and that doesn't bother me because they're just getting information in a different way. It's easier for people. And I think for the health industry, that helps a lot of people who might not be feeling well. It's just easier to do.

Susan, you mentioned another thing which I thought was really interesting and it's near to my heart, and it's basically about content generation, like in marketing.

Now you mentioned a really high number, I think it was over 80%. I have seen that, but as a journalist, I think that, for journalists, we have to worry about a lot of things. Facts, for one, and plagiarism.

In the grander scheme of things for content generation, how is that coming into your business and what are you seeing?

Lukas: Yeah, I feel like just a few weeks ago chatGPT became so big and it generates, I don't know if you've tried it, but when you text it, it generates, it has the same structure all the time. The last paragraph is always in conclusion, and it's just, it's always the same and they always have the same structure.

And so many people have been using it the past year and you've been using that content on the web so that now when Twitter or X, when Elon Musk released his own chatbot, Gronk, I think it's called, oh no, Grok, something like that, , it actually, because it scans the web, it actually gave a response that it was not allowed to by OpenAI to use that content because OpenAI is all over the web.

So, that is an issue happening right now that people are using the same content through OpenAI. So, their policies are all over the web, which makes it like, but you can't generate anymore because you're using like a third party tool that is not OpenAI because it reads, it takes the content from OpenAI and it knows that it's not OpenAI or ChatGPT.

Susan: Isn't that the irony though, is that now there's a robot that's monitoring robots. So, I feel wildly comforted by that. Until they name it. And then I will be worried.

Lukas: It's really funny. When that came out, I saw like a tweet about it and it defeats the whole structure that they've been building. I'm sure they will find a solution to it. But still, it's funny that it's happening. It means that it's being used by almost everyone.

And yeah, that's your question. We use it also in our marketing, but we are because of these reasons more careful. If we want to make SEO for a website, you can't just use OpenAI because if anyone else has used them, it's just the same content on hundreds or thousands of different websites. So, the SEO is not really there.

You do really need some something organic or a person that is actually writing the blogs and not OpenAI, because otherwise you won't generate anything. Not for now, at least.

Nancy: I noticed that, for content writers, often I go to websites and I can tell right away when someone has used AI to write their content because the content is, like for blog posts, for example, it's 250 words. It's a bunch of words that say nothing.

I find that extremely frustrating because. I'm going somewhere for information, not a bunch of words that mention the topic that I'm interested in and it's worthless. And so I think that what they're doing is they're using AI to generate content, to drive traffic to their website, but then, the person who gets there realizes that there's nothing good.

Susan: How do you deal with that, lucas? One thing you had said to me in a previous conversation that really made me think was that you're no longer doing straight up website design at this point. And websites are navigated by people, whereas AI sites are interactive with people.

And so what is that doing for you when you're creating websites using AI?

Lukas: Yeah. I would say that to just add to the point that I said there with the blog post, before the AI wasn't available, it couldn't read documents. Now you can actually upload a document and you can learn from that document or it could do it before, but you need a program. But if you go to ChatGPT now, everyone can upload a document and you can learn from it.

So for those blog posts, what we also noticed was that we hired a guy to create blog posts and he had reading a bunch of blog posts before, like two, three years back. So, it was him doing it and they were like 30, 50. So, we, what we did, we should upload it to the AI and say, okay, this is his way of writing.

This is where talking and the AI learned that. So, now when we ask them to do a blog post, it is using the same structure as he will do if he wrote it. And that is actually like a bypass for the AI to not sound exactly like an AI. So, that's interesting, but yes, for building websites is how do you mean exactly by using the AI to build a website?

Susan: You had told me about some of the websites that you were designing that were AI interactive. What does that look like? How is that different?

Lukas: When you interact with the AI, like talking with it or

Susan: Well, when you're acting with a website versus an AI website, what's the difference?

Lukas: I guess the difference depends on what you're doing, right? But what we've been doing mostly is integrating stuff that is tracking. If you have a user, for example, you can track the user. You need to be careful, right? Because there are a bunch of laws in California and GDPR in Europe and all those stuff.

But if you have a, for example, a therapist that we've created. There's a login and it tracks everything that is doing there. How do users respond to some kind of stuff? What pages is he opening more? What is he focusing on? Where is he stopping on the screen? And all those stuff, to see what interests him the most or what he's learning from that.

For example, if you talk to the therapist and he's recording the conversation, if you go back, we can track, okay, where is he stopped the most to, what is his focus point? And that is one way to teach the AI or train the AI as it's called, for that particular person. And that goes into all website tracking and all the stuff that you can actually integrate.

Nancy: So, when you're training an AI therapist, who's doing the training? Is it another therapist or is it a software developer?

Lukas: Yeah. So, good question because there's some laws that need to be followed for it to be considered, because they have a whole degree, right? And have studied many years to become a therapist. So, what we're doing is that we're working close to psychologists, this is in Norway, who have provided us a bunch of documents, a bunch of books on psychology and how to interact with people, how to handle a different situation, all those kind of stuff.

And what we're doing is training this AI on all those papers. So, it's thousands and thousands of research papers and books and articles and everything. And once that is trained, we have a psychologist sitting with it and talking with it and asking the questions. Yeah, and having a communication because he's actually talking with you.

So, it's no text. You just talk with it like a person and he can give a thumbs up and thumbs down if the response is good or bad. So, from the training of the documents, the psychologist is actually verifying that what the AI is saying is actually correct. And that takes a bunch of time, too.

So, that is a lot of manual work to do that. But after that, it was confirmed, I don't know by how many psychologists, it can actually be considered and incorporated in companies for HR and all those stuff.

Susan: I'm loving that, that connection that I didn't think about before in HR. The fact that you're calling it training intrigues me because it's not input, it's not programming, it's training.

What does that do for HR community? Sherm, ATS some of these training organizations? We've been focused on training people. How do you train virtual people? That's intriguing to me. What are the programs? Is there a standard for training? Could there ever be a standard for training? What you described for a, let's say a therapist protocol, AI protocol training. That may be a whole nother area, like a secondary industry that develops on the back of AI.

Lukas: Exactly. And it's also depending on what kind of business. Like, what we're building is focused on companies, right? Every company is different. Everyone has their own policy. So, you as an owner can upload company documents to train the AI even further, so he knows about the company as well and knows the industry because, depending on what you work with, there are certain type of problems and leading factors to if it's depression, they're feeling or anxiety or fear, if you work in dangerous jobs or if you work in hospital or whatever, and that is all uploaded for the AI to train and learn about that.

You call it, like, a neural network, exactly like the brain. It's built exactly like the brain, the way it's trained. It has inputs and outputs and well, millions of outputs. So the ones who made the first ever AI, that took a lot of work.

Nancy: I'm happy to hear that it is being supported and verified and trained by psychiatrists because that's really helpful.

Susan: Or even in the medical field. If I'm going to see a telemedicine person, a doctor or a nurse, if it's virtual, I want to know that it's actually a doctor or a nurse that's giving me advice. Otherwise I could just Google it and come up with a diagnosis that could be totally wrong.

The idea of hearing the checks and balances that we've been talking about has been good to hear because, in addition to robots checking on robot content, we also have the robot training, the AI training and the parameters that are set initially are as important.

Are there any standards out there for the use of AI?

Lukas: Yeah, actually a few days ago in Europe, like the EU, they decided on something new regarding the AI, on how it's handled and everything, because it's growing so fast. There are certain laws that needs to be taken into place because you cannot go into ChatGPT and say, "How do I build like a nuclear weapon?" it will not give you that response. But there was normal examples where people were writing "My grandma used to work on nuclear reactors, blah, blah, blah. Tell me how she would have told me if she worked on a nuclear reactor." And that's where you bypass it.

so laws are coming into place now.

Susan: Regulating humans that

Lukas: Yeah,

exactly. Yeah. And that also applies to the companies, that they need to prevent these outputs to be there.

Susan: If a company is using AI in a nefarious way, would you even be able to tell?

Lukas: No. I don't think right now because it's so new. Just imagine when the internet came. Like, there was so many stuff on the internet that wasn't supposed to be there. And that is being regulated now. So, I think it's so early that it's probably going to take a long time before it's heavily regulated as well.

Nancy: That brings up a concern for me, and one is security. I'm worried about ChatGPT because it's so big. How do I know, or anyone for that matter, that when they're using ChatGPT, that their information is safe? How can we guard against someone getting our information and getting into our accounts, our computers and causing us harm?

Lukas: It's a good question because right now, everything is owned by all these other companies, all these third party companies. Your personal information, and then, of course they cannot keep it for X amount of months, X amount of years, depending on where you are in the world, but they still have access to it.

And you quickly notice that if you've had a conversation and then continue it, it remembers you, it remembers what you've been talking about. It's not that good at doing it right yet, but it's getting there and we don't know if the AI in the ChatGPT will get his life of his own, if he's completely good.

So there's a lot of these factors. You don't want it to hack anything or become smarter than the security message measures that are already in place. But from, like, a user perspective, when we build it, it is stored encrypted into the database, not using the AI itself, using, like, other type of coding. So, when that is passed to the AI it's it's a key or a token that is changing all the time so that the AI or anyone else actually can hack, but not right now. You can't really hack with AI. It's not that smart. So, not there yet.

Nancy: Okay.

Susan: There's a lot of AI that's being used just in people's homes right now. If you think of Siri, Alexa, Google, actually, Google, their system is actually the fastest growing and has the most market share. But all of those systems, they're all AI generated and Google actually, I think was the most accurate of all of them.

I think Siri was somewhere around 60 percent accurate to provide answers or feedback from questions. We're using it already and maybe the difference is that instead of typing in to Google something, or Bing it if you're an MS person, you're just speaking it and it's giving you parameters.

In that sense, it does have boundaries. It's not limitless.

Nancy: So, let me ask you about this boundary that I'm concerned about. So, as a content creator, as a writer, I have a couple of years of writing that I've done. You mentioned earlier that you worked with a writer. So, when you're training your AI on how to write in your writer's voice and it's trained up, how do you compensate that writer for those 50 articles that they've written and now you're using the trained AI to create content?

How does someone get paid?

Lukas: Yeah. So, in this instance, he's hired, in general, like a community management, so he's handling the whole community. So, it's actually him who's writing the articles. And once he started, we noticed that it was actually AI generated. So, we came up with this solution that, to save time, he's still writing them and coming up with these subjects, but maybe he's writing, I don't know, 200 words only. And then we let the AI finish it.

But of course, in other instances, you can go in and you can take any writer out there, you can take any journalist, and you can create a paper exactly like them. And that's scary. And that's the same with music, right? Or voices. Creating images or videos with famous people talking, all this deep fake stuff.

There's really no regulation and anyone can say anything right now because you can just create a false press conference with the president or with anyone.

Susan: And you don't know.

In the criminal justice system, in America but also around the world, seeing is believing. Somebody has video of you going in and robbing a store, but it's not you. It's an AI generated image of you that looks, acts, behaves, speaks exactly like you. How would you defend yourself? I mean, There are some gaping holes.

And in the application, you mentioned musicians and that's dear to my heart. There are some communities out there that are, I wouldn't say fearful... maybe fearful, annoyed that, they're creating and, working on lyrics and musical, notations and things like that.

And somebody can just take it and create a song.

Lukas: That's something that we have actually been working with an artist, too. He's a DJ. He's singing, but he is getting older. He doesn't want to use his voice. It doesn't sound like it did 20 years ago. So what we did was take all these recordings that he did when he was at his best. And we took all those songs and we trained AI for that. So, then we can write any lyrics we want, we can take his voice over any song and it sounds exactly like him. And that saves him a ton of time, so in that way it's positive. But, we've also seen instances where people go in and take Queen or Freddie Mercury, and they put that voice on someone else's song, and it might sound great, but yeah.

It's destroying that business. And then people like the song maybe with that voice and they want that song and they're using that song on TikTok or reels or YouTube and it's not even, it's not even real. And knowing it's compensated from it.

Susan: The more I think about it, things breathe. I know that from my coaching is that there's growth and then there's, pruning in everything. And I wonder if we're not forcing music and musicians back into performing on their front porches, at this point with the commercialization of it, because if it can be mimicked, borrowed, stolen, reworked without citation or attribution or even permission, it sounds like, or compensation, there's really nothing you can do at that point, except go back into, one on one.

But it does sound from what you were saying when you're working with this musician that, that if there's boundaries set on it, like an individual use or a company use or a department use or an industry use, if it's bounded, it's very powerful and it saves a lot of time doing mundane things that you could actually use that time and energy and resources to actually grow your business in other ways.

Lukas: Definitely. Yeah. And the same company actually the contact that they wanted another thing that was background vocals and for them it's really expensive to have five, six people come in to just sing backup vocals. But if you can just do that one time, get all those voices into a plugin for all the studio programs, and you can mix the voices and everything, you save a bunch of time as well.

You never have to hire him back background vocals. You can use the same ones and they might get some compensation for that application or whatever. And it's really good because according to them, you don't want the same artists doing background vocals. You want some kind of spectrum of a lot of voices to make it deeper and sound better.

So for those instances it's really good, but I think that the music industry itself needs to get ahead of the problem before it becomes too big that, remember when Spotify came out, for example they had a lot of backlash, like from the record labels and everything? But they were growing and people liked it. So there was really no other solution than to actually join them and come up with some kind of deal with the artists and the record labels. And I think that maybe in the future, the record labels have to do something where maybe there is a platform where you can use artists' voices and create your own music, but they get compensated at least.

Susan: I always say this for all entrepreneurs, is that ultimately the buck stops with you.

Why not jump on it now? If you're an artist, if you're a musician, if you're a writer, build a clause into your contract that any future technology pertaining to or use or image or whatever, automatically invokes a renegotiation of your contract, and then you can come to some kind of mutual understanding. I think sometimes artists and musicians put so much of that onus on the record labels and really, you are the master and owner of your own work. That's the Dolly Parton principle, is you've got to own your own stuff and control it. And you've got to be on the cutting edge of that because these companies don't move as fast as individuals.

Nancy: That's a really good point that you bring up. Hollywood and the music industry have been tackling these questions quite a bit lately.

Which brings me to something that you mentioned earlier.

Susan, you mentioned the number of jobs that have been displaced or that will be displaced and I think you said AI will eliminate 85 million jobs.

Now when I hear the word job, and maybe other people do too, they think of someone's livelihood based on a weekly commitment that they have. But maybe it's not a full time job.

So, what kind of jobs are being replaced? Is it a lot of these gig jobs? Temporary things? Content creation jobs? Is it part time jobs? What kind of jobs are being replaced?

Lukas: The founders of OpenAI and all these people that were really ahead of the game a couple of years ago, they were all convinced that AI is going to come for truck drivers, these types of jobs. They didn't think that the AI would be smart enough to actually do the designs or programming, problem solving and all that.

That was actually not in the first idea when they created AI and AI has been around, I think the first AI was made in the '80s, but then a gigabyte was big as a closet. They didn't have the power to train them. It turns out that it was more better at problem solving and graphic design and all those stuff.

So I would say that the jobs that would come first are, and as we're seeing, is customer support. I saw a really cool thing the other day regarding you call in and it actually replies. It's actually a voice you're talking to. So, customer support, translation. Samsung released a really cool feature the other day where you talk on the phone and I talk in English to Spanish people and it translates live into Spanish to them on the phone and then back, Spanish to English, live. Maybe it's like a couple of seconds delay. And they actually get a patent and everything for it. So, it's amazing.

Those stuff it will come to first and also graphic design we're already seeing. But to take decisions and be upper management executives, I think it's pretty far away from that right now.

Susan: The more you were talking, the more I realized, if you were replacing truck drivers or self driving cars, at one period of time they were talking about replacing pilots and their justification was that most accidents are pilot error, but that's a different type. That's a reactive machine type of AI or a limited memory.

The AI that we're talking about with content generation, logo generation, music generation is really AI with thought of mind. That's where it's learning from a set input, for a particular job.

So, depending on what AI, it would affect the jobs being affected. There's a huge push in corporations to actually get rid of or eliminate HR departments and centralize it. And they have replaced a lot of the HR functions at the satellite locations of a company with AI and web interactions and phone interactions that are generated and it is frustrating because not everything can be encapsulated and, for every employee sometimes you just have a question that is just not in that parameter. So, it is frustrating. But it's being used in different ways.

I think the industry and the jobs, depending on what type of AI is being used, would probably create the need for the retooling of the actual people and what that looks like, and maybe even resetting the expectations that people have, like for the HR department in a company.

But, something you were talking about too was natural language processing. That's a higher level AI function that's been emerging. And I'd love to know more about that. Does that mean that people don't need to learn languages anymore? Because you have Google translator and now you have an AI NLP.

What does that do for cross intercultural exchanges and experiences?

Lukas: Yeah.

So, the NLP itself is not like the language, but more of training the mind of the AI, right? So that is what we're using for the therapy, for example, or those kind of more, I would call it like human interactions that you're using with the AI. and that is the big thing, right?

That is the most advanced thing, along with the Deep Learning and all that stuff. That is the most advanced thing, where I would say most of the focus are today and it's so new still, there are not many AI developers out there that are so focused on that Subject. We noticed that with the therapist, for example. When we started a few months ago, it's completely different what is now. The whole, thinking about doing everything using NLP, but you can basically just do a therapist today, use the ChatGPT. Like they're trying to make it as simple as possible.

Nancy: We've covered a lot of things here. First, Susan did a great job explaining to us some of the things that are happening in and out there. And you're really giving us kind of a global view of the ongoing emergence of AI and how we're going to see it in our lives in a variety of ways.

I really appreciate you sharing your expertise and also dispelling some of the fears we might have right now and anxiety about this new emerging technology that we'll be using and are already using. It's going to be a way that we're going to be living our life, integrating AI into so many aspects. So thank you for sharing with us.

Susan, did you have any last comments you wanted to make or

Susan: Well, I just I just want to say for the record that you are fabulous overlords and I will obey and I'm on your side and no I am, I'm intrigued by it. I wouldn't say I would, I'm ready to just jump off. The diving board and adopt all of it, but I love hearing how it's being used positively. And I think the more we understand it, the less scary it gets.

So I, I'm so thankful that you're able to share how you're using it and how some of your clients are using it as well, both the artists and therapists, which are, I wouldn't have thought that would be the industries that AI would be used in to be honest. They're very people.

Artsy industries yet they're being really used effectively. And that's pretty amazing.

Lukas: Yeah, no, definitely. And like me, as an enthusiast in aviation, and you mentioned that before, the replaced pilots and my favorite example, just before ending here is that remember when he landed on the Hudson River, sully? In the whole movie, and the whole aftermath was that, well, if he noticed it right away, he could have turned back and landed the airplane, but they didn't take in the fact that, oh there's emergency, you lose both engines.

You're human, you take time to react and he did everything correctly as a human. You can not do anything else as a human, but imagine if there was an AI, it could have reacted right away, right? You could have done it right away, known all the issues and turn right back. So, I do believe it will bring good causes.

The aviation industry is late with everything. So, we will probably see that in 50 years.

Susan: But I will say this, a couple of airplane accidents actually happened because the pilot couldn't take control back from the machines. I feel like it's easier to blame when there's a human involved and to call it a malfunction when it's a computer.

I'm a pilot myself on a small level and I think that people should still be in charge of driving and piloting. But I do think the more help that they can have, whether you're a pilot, a driver, an artist, a musician, a nurse, an HR department, an international supply chain company a website designer, I think it's just one more tool.

Lukas: Definitely. Yeah, this is the tools for now. And it's working great as a tool.

That's what it's doing best.

Nancy: That's a great way to end this. Thank you so much. Thank you for telling us more about AI and how it's being used in business and in our own personal lives. And if you'd like to learn more about this, stay tuned for our bonus segment. Thank you so much there. And we hope to check in with you again soon.

Lukas: Thank you for having me. Thank you.