Day 3 of the I ️ AI Challenge : AI in Your Daily Life

Hello, good morning, afternoon, good evening. It is Valentine's Day. Super excited to be here with you. As a matter of fact, one of my children was— is— their birthday is tomorrow. So this is a big week for us. And so excited to be here with you. Welcome to day 3 of the I Heart AI Challenge. I always want to give— we're a little— not— we're not early, we're right on time. But I always like to give a few minutes for people to find us and connect and have a chance to say hello in the chat. When you get here, just say hi. Let me know that you're here so I can— we can have some fun together. I will be sharing with you today. It's a slightly different take. Yesterday and Monday, we spent all of our time focused on marketing and sales, not just for your business, but also for your kind of professional life. Today is a little bit different because it's more about AI every day, AI in your daily life. And I'm really excited to tell you about it because many times people hear about how we use AI in our household, and they're like, oh my gosh, you're like, I don't know, 3 generations ahead of everybody else. So I'm going to share all of that with you. Good to see you, Wes and Sky. The community is here. Great to see you. Thank you for coming and joining us. We, like I said, we're just getting started. I did finally get the whole link to post automatically in the community. So yay. So hopefully we'll see our friends joining us here soon. Now I'm going to go ahead and show you a little bit, just draft out our plan for today. And I think it's gonna be pretty exciting. The reason that we are doing this challenge, for those of you who are brand new or who are catching this, you know, I noticed yesterday, if you're on the events page, you might actually just see this and be like, what's going on over there? So if you're here and you are not part of either our I Heart AI community or you're not a follower of me specifically, I always like to start with a tiny little introduction of who I am and what I do and why I do it. But today, because it's about AI Every Day, you're gonna hear how really artificial intelligence is infused into every aspect of my life, from taking care of my children to taking care of my dad to taking care of my pets to everyone. Awesome, Joe, good to see you. Welcome to the party. Happy Valentine's Day. I already saw some really cool stuff where, uh, somebody did a post. I totally want to do one later today about, um, Galentine's Day. One of my friends, Barbara Pender, she's another responsible AI, uh, enthusiast and advocate and speaker, um, and she sent me a message this morning wishing me a happy Galentine's Day. So for all you gals out there, happy Galentine's Day to you all too. Um, so we're going to take a whirlwind tour through what we're going to cover over the next hour. I'll give you a little bit of information about me and what we're doing, um, and then, uh, super excited at 11 o'clock, 11:05, we'll be bringing on, um, our guest speaker today, which is Erin Reddick, and I'm so excited to have her. Uh, we became friends actually through LinkedIn, as I met most of you, um, and we became friends on LinkedIn, um, and I started to find out all the things that she's doing with Black Black ChatGPT with her speaking and her advocacy and the advisory services that she offers. So super excited for her to come and tell us a little bit about what she's learning in her area of the world. So super cool. So I'm gonna try not to get ahead of myself because this topic is so exciting for me, and many of you know I'm pretty excited generally speaking. Yes. Oh my gosh, you will definitely see me at Build. Michael, great to see you. I love Build. I actually was a Dev MC a few years ago when the Microsoft Developers Practice decided to put MCs. Oh my gosh, I should totally do that this year and just go and be my own Dev MC and post like shorts and longer interviews to my YouTube channel. I'm— oh my, Michael, great idea. We're doing that. Good morning everyone. Good afternoon everyone. So glad you are here. Okay. We do have a global audience. We have over 1,000 people who have joined this challenge. Not everyone can show up live, but I also am streaming on 8 different channels, so multiple Facebook pages, couple different areas on LinkedIn. So people are invited to join me from everywhere. I try to synchronize all of those chats in one place. If I miss you and I don't call you out, it's not because I— you know, it's kind of like on Peloton, like they do their best to hit everybody, but sometimes they miss. I remember a couple years ago was my birthday and I got on a live, which I never do, and they, they'd never called my name out. So I wrote the instructor and they're like, I totally did, but it was 5 minutes before. They do this like pregame hype on live, on live Peloton rides. Little did I know. Um, so I am a big believer of trying to give all y'all some love. Talisha is here. Happy Valentine's Day to you. This can be a great fun day.

Okay.

So Everyday AI, this today we're going to cover a bunch of different areas. The first one I'm going to talk about really is around smart home. And I know some of you have been around long enough to remember when smart home was like a thing that we were doing, and it was a struggle. Like, it was just a struggle. We had all these different like Z-Wave or Zigbee or anyway, all these different Devices spoke on different wavelengths. It was hard to set up. Everyone had their own app. The world is not too much better from that perspective, but spending time at CES this year, I realized that I'm actually kind of a product of a lot of the things that I'm starting to see come back in this new age of AI readiness that the world is in due to the advent of ChatGPT over the last year and a half. So I think it's really exciting, but I'm going to demonstrate what we use every single day and who uses what and how they do it. A friend of mine, Joel Evans, he has a blog and a bunch of really cool stuff. And one day he came up to me, he's like, we really have to write down all the things that you do with your smart home, like household. And I was like, oh my gosh, we totally should. We didn't. But this will hopefully give you a little bit of a sneak peek into the world of Noelle and the Russell household and how we use AI every day. The second thing we're going to talk about is how I use AI as a personal assistant. And we are not yet at the stage where I'm writing it a little slant. Anyway, you don't— no judgment. But one of the things that, you know, I imagine a world where we are going to— it's going to be like Her. Do you remember that movie, the Her movie? Anyway, the good part, like, I don't know, was there a good part? Anyway, it did exemplify a world where we could just talk to our tech and that tech would sound very human but also was very transparently robotic. And so, but in that scenario, we talked to one voice, that voice in our ear or on our computer, and that voice did a bunch of stuff for us. For us, right? It would check our calendar and book our flights. Today it doesn't work that way, not because technically, like technology-wise, we're there. We have services that could easily build multitask orchestration engines leveraging large language models and even do it in a pretty cost-efficient manner. We're more, and this I see all the time with really good tech, I saw it at Alexa, I saw it at Microsoft AI in the early days, like even though the technology was capable of doing things, it really requires people like us developers, entrepreneurs to see a problem and solve it with the tech that's available. Most people, most problems don't have someone who's willing to marry the two together. And I know there's a few of you on here that are exactly this type of person, so know you are needed. We call ourselves— we're kind of like magical unicorns, right? Where we could see a business problem, we're like, oh my gosh, I know exactly the tech that should be used to solve this problem. I believe over the next 2 years we're going to start to see a lot more of these multitask orchestration engines. We're going to see Alexas and Siris become more of these orchestration agents that help us have a single point to talk to, but then lots and lots of agents will be communicated to. Yes, exactly, exactly. I love it. Thank you for that, Transistors. Okay, so personal assistant.. So we have a lot of stuff in there. So as opposed to me saying, go talk to your her bot and all will be well, I have about 6 systems that I use, but they help me amplify the work that I can do as a single individual. As one person, there's only so much hours in the day, and sometimes I get so frustrated with my inability to do more. Like, I want to do more, have more, be more, and I, I'm limited. Like, I only have so many hours in the day with the life that I have. However, this technology, I have stacked it up and used it to add hours. Sometimes I feel like I've added days to, to my, my life, um, in a single 24-hour period because I'm able to do so many things simultaneously, uh, to help me increase my productivity in the world. Okay, and I'll talk a little bit about whether or not you're doing this as an intrapreneur, meaning you're inside a company and trying building this internal productivity to make you a top performer in your organization, whether you're an entrepreneur starting your own business or whether you're just like me, like you want to do good things in the world and you don't have enough time to do it. So I'm going to give you some tips and tools about how I've done that as well. Um, all right, yes, right, Sky? What— hey, amen. Um, you know, being— I call us the sandwich generation, right? Where, where I'm in that world where I sit right between raising my children and taking care of my dad. I'm sandwiched in the middle, and I think it's relatively unique generationally. Like, I don't think our— my parents certainly didn't have to do this. Um, like, I was already in college when their parents started to get sick and need help. I think it's very interesting to me that I'm now like raising a 6-year-old and taking care of my 76-year-old dad. So It's an interesting time. So once we go through personal assistants, I'm then going to talk about a critical, critical topic. And I actually had some conversation about it, yeah, an event I was invited to, and it's called mental health. So AI can help you in many ways, but it can even assist in some mental health activities, and I'm going to share with you how I personally use it. Again, I'm not encouraging you to infuse AI in all the things that you do, but you should be aware of where— at least where I use it— so that you can choose. Like, oh my gosh, yeah, that sounds awesome, I want to do that. Um, and maybe you do no smart home and more of the mental health stuff, or maybe you do a lot of personal assistant stuff. Like, find opportunities to amplify your existence as a human on this planet. And I have found a lot of joy in doing that, and I hope you will as well. Finally, I'm going to leave you with, um, some top tips. And these are things that don't fall into these buckets because basically what started happening as I'm brainstorming the things I wanted to share with you, one list got way too long. And I was like, you know what, I'm just gonna put a bunch of them at the end, and if we get to them, great. If not, it'll be a nice blog post. Later on. Um, so amazing. Uh, oh my gosh, awesome. Yes, working with seniors. I mean, it's an interesting world we live in now where people are living longer, and thankfully I want to live longer, right? I expect I'm gonna live like a really long life. I'd like to live over 100 years old. My grandmother lived to 98, almost 99 years old, and, uh, I'm like, okay, great, I'm gonna do that, maybe more. And longevity, of course, if you are into that thing, biohacking. My husband and I, we do a bunch of stuff in this area, but I'm going to talk to you more about that. I also want to share with you as we go through these topics that we're actually in a world where we want to blend humanity and AI, right? Create a symbiotic relationship. So like Venom, but in a good way. But we want to make sure that we have this opportunity to leverage AI to amplify our human ingenuity, our capability. But we also wanna realize like we're not removing the human element, we're making it better, more capable, more accessible. And so I think that is gonna be a theme you see running through all of these topics. All right, so let's start with the first one, smart home. Now, like I said, this is not a guide. It's not intended to be like, you should do these things. It's just hopefully going to help you see some opportunity where— and this is only going to get better over time, especially with the advent of the world understanding artificial intelligence. So I remember, you know, oh my gosh, I guess almost 10 years ago, I was running around CES, the Consumer Electronics Show. This is when refrigerators and appliances and lighting systems and household items and robots. Oh, I have one of my robots here, so I'm going to share it with you. But the smart home world, we were— I mean, the technology was there. Being able to command devices in our home, being able to let it infuse kind of efficiency in our home has been around for quite some time. But I think adoption was difficult because people didn't really understand it. And when you don't understand something, it's really hard for you to trust something. Now, one thing you will know did take off that was also released during, um, a CES, a Consumer Electronics Show event, was the iRobot, right? The— is that the right name for it? I think so, right? The little robotic vacuum cleaner. There's so many of them now. I think it was originally called iRobot, but now I'm like, wait, that's Asimov. I can't remember. So whatever that thing is called, um, but Shark has one now, right? Like, I'm sure there are other vacuums. If you go to Amazon, they have lots of different ones. And these robotic systems leverage artificial intelligence. And most people don't really think about it that way, right? They don't think about it like they think about Alexa, like, oh my gosh, you know, like, it's funny to me that people are worried about Alexa. One of the most common questions I get when I mentioned that I was part of that team is they're like, I mean, does Alexa really listen? All the time. And I always like kind of joke, and I'm like, yeah, but you know this thing that we have, our smartphone? Yeah, Alexa is not the one telling Instagram that you were just talking about a protein shake and it showed up on your feed. Like, that is not— that's how it works, right? Your phone is listening way more than an Alexa device. But similarly, I'm like, are you, are you worried about iRobot knowing every inch of your house? Right? Like knowing exactly where the crevices are in every corner of your house in order for it to vacuum up dirt. Here's the reality. When we're looking at smart home, we are almost always willing to take on the risk, privacy risk, of using a smart home device in order to get the convenience that it operates and that it offers us, right? And the reason that dynamic exists is because And I've done this many times. I've been on stages in front of thousands of people, and I'll just ask them by a show of hands, you know, if— and I use a scenario, a fictitious scenario— what if I could, at the airport, for example— now granted, usually when I'm talking to people, they will have just flown into an event, right? So they already— this pain is fresh in their mind. And so I'll ask them, if you're at an airport and I could offer you frictionless drop-off and pickup, like in and out, no traffic, no stopping, no rushing, like beautiful experience. If I could offer that for you, but I would have to surveil you, right? I'd have to know who's in your car. I'd have to know, right? Because if it's a girlfriend, maybe it'll take longer. If it's a husband, maybe he gets out to kiss you goodbye. If it's, you know, your kids, maybe you like shoo them out the door faster, right? But there's different routines and patterns depending on who's in your vehicle. And so I need to know all those things. I need to know your face. I need to know their faces. I need all this data. Now, when I ask groups of people— now, these are not technologists by trade, right? They're not engineers. They're not data scientists. These are like people who farm for a living in the middle of the country, or people who are real estate agents, or people who are business owners of $1 million to $5 million in revenue. Like, they're not thinking about AI the way maybe you all would think about it. And I think it's really interesting because when I asked them, would you give up that data? And I don't phrase it that way because they wouldn't understand, right? Like, would you give up or would you allow that type of surveillance to happen in order to give you a frictionless experience? And because they all just experienced that pain, 90% of the audience raised their hand and was like, absolutely, I would do it. I would love to have that experience. And most of us listening to that are like, oh no, I don't want you surveilling nothing. But it's interesting, the world that we live in, that we're willing to sacrifice privacy data in order to get convenience. At a much more smaller scale, Amazon has been collecting data about us for decades, right? Has been collecting data about, about our shopping experience and has been using that to make our experience better. And the reason we do not reject Amazon's use of our data is because transparently they tell us, hey, we're looking at your shopping data and we noticed other people are buying similar things. You bought this, well, 1,500 other people, when they bought that, they also bought these things. And I'm like, oh, I never, I never thought about that. And it's like me at a grocery store. I'm like, oh yeah, sure, I'll take that. In that and throw that in the cart.

It's terrible.

But they understand that they're using my own data to create a better experience. But what else are they using it for? I don't even really ask the question. Okay, I asked the question, but most people don't really ask the question. They don't really even understand how that data is being used. So I think when we move into the world of smart home, it's important to realize that, like, you know, as with all things, I always try to come responsibly focused. Like, how do we become more responsible, um, in our use of this technology? So you should know when you start using these tools that all of them will collect data, and it's okay, and you might even be willing to give that data to someone, but you should be intentional about it. And that intention, I think, is critical. All right, so areas in my home— I already mentioned it— cleaning the house, right? Robotic vacuum cleaners. I I can't remember how to spell vacuum. I cursive'd it to try and— anyway, vacuum cleaner. So that's a great one. Another area of our home that we use artificial intelligence, of course, is Alexa. And I'll maybe even say voice-enabled devices. We have over 100 voice-enabled devices in our house, over 100. And I want to give you some examples of the types of things that you can do with these voice-enabled devices. And it's worthwhile taking a look. Samsung, um, LG, right? These types of vendors that generate, create these, um, devices. I am very interested, for example, in a microwave that cannot ever— because it has sensors inside to detect, right, what you're using. Oh my gosh, our blender! I love this one. Our blender. We got a new Ninja. The Ninja now, of course, is AI-enabled. And what that means is that it has sensing technology. It's really more IoT-enabled, but that doesn't sell nearly as well. So my blender has sensors in it to detect what did I actually put inside of it. It then blends and senses how that blending is occurring to create the perfectly blended, let's say, margarita. Even though I don't make margaritas, but it was on the box and it looks delicious. But I'll make a shake with like kale and blueberries, and in my old shake I would just have to— I mean, I can't tell you how many times I'd take off the lid and look at it, be like, no, put water in it or milk in it, close it up, put it back on, take it off, like 4 or 5 times every time I make a shake. Now it's cool, it senses the friction on the blade and it will actually tell you water or liquid is needed. So it gives you this insight. It also— I use, um, a technology. Oh my God, yeah, I use a technology, a vendor called VeSync, and I use VeSync in a couple different ways. Kind of strange now that I'm going to say it out loud, but hey, we're friends, right? Um, I use VeSync, I use their air fryer which is voice-enabled. It allows me to turn it on, turn it off, allows me to preheat it when I'm not right next to it, which is super nice, especially when I'm about to throw in tater tots or chicken nuggets. Um, but I also have a scale. Oh dear, a scale, a bathroom scale. Now what I love about this bathroom scale, and actually I was just working with this organization and they're doing something very similar, and it's called AIMD. But in this bathroom scale, just write that in here, AIMD, like WebMD but AIMD. And you could Google them, but they, you know, there's a scale that I step on it and it leverages sensor information, right? It's getting information from the sensors on the scale, but then it's also tracking behavior, like it's tracking data changes over time.. And it's giving me— I always call it, uh, when I'm on projects, I'll call it intelligent failure discovery. It'll use the data it's collecting from me in order to detect, hey, I see a pattern of behavior, skeletal muscle decline is happening. And it is able to sync up with Apple Health on my phone. And it's like, hey, you should do more strength training, right? You should actually increase it by a very specific amount, 3 to 5 times per week, because of the decline we see. And now this decline, when I look at the data, is like fractions of a decimal percent. But when it sees a trajectory, it— long before I as a human is— I'm going to be able to detect it, it's able to now give me advice about how I could do better. Similarly, AIMD, anyone can go get it. I absolutely love it. It actually takes a picture of your face. It is a scan, and through your face, it's a model, uh, this woman created, um, and a set of AI models, and it's able to detect your heart rate, your blood pressure, your cortisol levels for stress, all these things by just looking at— through your phone at your face. And the reason I love that is because not everyone can go to the doctor, uh, every time they're worried about something happening. Not everyone has insurance. And so what happens in a world where we can use our phone as a medical device in our own home and this concept of tele-doctoring, right? Where I'm actually talking to a doctor, but I don't have the cost of going into an office, right? Making healthcare much more accessible. Super interesting stuff. Like definitely take a look. Oh my gosh. Yes. All right. Thank you very much for asking. So somebody was talking, just mentioned, I can't see who you are, so you could tell me and I will call your name out. It just says LinkedIn user. But there is a device called Astro. And I was excited because Astro, if you remember, is the dog from The Jetsons. So cute, right? Love it. Astro is a little robot. And actually, I'll show you a mini version of what it kind of looks like. So it kind of looks like this, but it's much bigger and it's gray. And this is not Amazon's device. But it looks like this. It has wheels, it rolls around. Here's the thing though. And again, it's not because the technology isn't there. The tech— oh, maybe I'll keep this with me. I'll put on— the tech is there, but all Astro can do right now is connect with your Ring doorbell and kind of follow you around and tell you who's at your door. Oh, Valencia! Good, Valentina! Good to see you, Valentina. Um, awesome, good to see you, thank you. Uh, but yeah, so I love Astro, but it's limited in its capability. My vision— because I have a dad who has a traumatic brain injury, he actually just had a seizure. I told you guys yesterday, just had a seizure a couple days ago. And it just doesn't do anything else than connect to a Ring camera. So for my dad, who's less mobile, it's nice, like He can't get on a phone. So it's nice to have Astro come up to him and tell him, hey. But I would love an Astro that could not only do that, but it could also distribute pills for my dad, remind him of arm lifts and leg lifts, give him demonstrations of the exercises he has to do every day. So I see more for that product. It's not there yet. Amazon's relatively new. It's on market now. I think it's about $1,000. Um, maybe it's for sale. It definitely will be for sale for, uh, this coming summer, uh, right? They do— what's that called? Prime Day, right? So you can keep an eye. But I love the idea of Astro. It's just from a feature function perspective, it's not quite there yet. And it does make me wonder, on that team that's deciding what to prioritize, is there someone like me going, hey, have you thought about people taking care of their 76-year-old dad with a cognitive brain injury? There probably isn't, but who is informing what's important, right, for the people that will use this device? I don't really need a device that's going to tell me who's at my door. I need a device who's going to help me in much bigger ways. So that's just me, that's fine. Um, all right, so vacuum cleaners, Alexa. Oh yeah, the microwave not burning things. Also, we now are seeing, um, Well, yeah, dishwashers, but I was actually in my mind going to washing machines, like laundry. So laundry— I have 6 kids, right? So laundry is like the bane of my existence. It is always— there's always laundry. So I am excited to see some of the features that are coming out that are AI-powered here. One is, uh, dirt detection, right? How dirty your laundry is. And being able to use that information to remember settings as well as adjust the amount of soap you use. I mean, I don't know about you, but most of us just pour the soap in, and most of us pour extra soap in because we think it's dirtier, and so we think that's going to be helpful. We don't actually mess around with any of the other settings. Okay, I'm very much speaking for me and being vulnerable about maybe my lack of interest in changing the number of rinse cycles I need. But I am a big fan of a laundry machine that can do that for me, that can detect the level of dirt in my clothes because I wash the same clothes all the time. It knows the patterns that exist in, in my laundry. It knows how much laundry detergent, how much fabric softener, how much of this new, like, stuff that you can put in to make everything smell nice, which I absolutely love, how much of that to use. And how much, how deep the fill should be. I have to click that button today. How many rinse cycles there should be. Like, these are things that I now actively go and do. But I, again, at CES, got super excited about washing machines. Like, every device in your house being uniquely aware of the patterns. We are, let's face it, in a bit of a grind, right? Like, most of us do the same things every single day. And if we let sensors observe us, they could actually find ways to increase our efficiency without actually interrupting what we do. And that is kind of what I've seen, like vacuuming. Oh my goodness, we have one on each floor. Super excited. We're about to move to a new house. It was one of the things when I walked in, they also had a robot on each floor that vacuumed the floors. I'm like, ooh, but that robot's already trained. So I literally asked them, does the robot come with the house because it's already trained on the corners. It already knows where everything is. That training is tough. All AI model training is tough, but that training takes time. And so needless to say, I got them to throw that in. But these are the things, right? How— what a different conversation I'm having with, um, with, you know, the people that we're gonna take this house over from, uh, about how they're leveraging technology in their household. They actually had a device. I don't remember what it was because it was brand new to me, but they're from Chile. And in Chile, I guess there's this like AI sensor-driven cooking machine. It's kind of like an Instapot, but better. And they invited me to learn more about it. And I was like, I wrote it down, but they said it was like magical. So I can't wait. If anyone's from Chile or knows this device I'm talking about, feel free to throw it in the chat. But I was like, this is awesome. And they were, like I said, not really technologists, but they were using technology to amplify the time they wanted to spend doing other things, right? Set that Instapot thing to on its own and like make sure that you spend your time with your family or you spend time doing other things. All right. So this is my— oh, I mean, of course I should say we have plugs, light bulbs, Fire TVs, you name it. One of the— again, it's still a frustrating, a bit of a frustrating experience for me because I always have to mess around with what I name things. So the light in my room is called my light because it's mine. However, it turned out that I turned on something downstairs and I also called it my light. And so now I had to clarify which my light I was talking about, which is a bit of a mess. So, and I don't want to be like in my— if I'm in my room talking to a light Like there is technology today that can tell what device I'm calling on and the closest light in my proximity. That tech exists. It is not currently implemented as a feature in the smart home world that I live in. And so why? Probably because the person who's building these smart home features didn't think about how useful that would be. So this is why it's important for all of you, like no matter where you are, to think How does my personal life experience make me a better participant in the development of systems? AI systems. Every system is an AI system now, right? Even if only a little bit. And if they're not yet, they will be over the next 3 to 5 years. So no matter what you care about, no matter where you are in the organization, right? Whether you're a personal assistant, whether you're an executive assistant, whether you're a programmer, a user experience designer, a product owner. Like, it's so important that you use your personal life to fuel what might be important to the very people you're trying to help and trying to serve. All right, so that's, that's kind of a lot, a little, I don't know, I like it. I have a lot of fun in our smart home. It doesn't always work, but I haven't been locked out of my house yet. Oh my gosh, I remember in the early days of Alexa, we, when Ring first came out, people were worried that you could just use your voice to unlock people's doors. And, and in the beginning, you totally could. You could just yell at their Alexa if they had a Ring device, unlock the door, and it would unlock their door. Obviously, that didn't last very long, and it was fixed right away. But it was just fascinating to me because you— these are things you don't think about, like, oh, what if just somebody tries to, you know, is outside of my kitchen window screaming at my Alexa to order stuff? That's strange. Oh, so as a result, we have voice pins, right? So I have a— granted, I don't have a voice pin because I'm worried about people screaming at my Alexa through the door. I have voice pins because I have a 6-year-old who doesn't realize that when they order stuff on Alexa, it costs money. I'm trying to teach that lesson. But these types of things, it's so, I think, fascinating as we build out this ecosystem of products that we don't realize how the world changes when they start talking to each other, when different people, bad actors get involved. So it's a very rich and even now, dare I say, 10 years later, a new world of helping people in everyday life use this technology to make their lives better. And I will say my dad, I'll get to him a little bit more in the mental health section, but he uses Alexa every single day, multiple times a day. I mean, maybe 100 times a day. And for all sorts of things. Like, he'll be watching a TV show and he'll be curious about the actor, so we'll ask, you know. I remember he was watching, um, the latest— oh, it was— what was that? Oh, I think it was the Macy's Day Thanksgiving Day Parade, and he saw Cher was on in, and he was like, Cher? He goes, Alexa, how old is Cher? And he's— and it turned out Cher and him are the same age, um, and but obviously Cher was out there singing and dancing and doing things that he's not doing. So it's fascinating though, because in a minute, in a second, he had a companion he could just ask questions of and that would cheerfully answer his questions. Sometimes he has to ask the same question more than once because his voice is a little bit slurred due to the accident that he had. But he doesn't— he has a lot of patience with Alexa. I've heard him having a conversation with Alexa and apologizing for not being more patient, like my dad apologizing to Alexa. And I just, I foresee a world where we have this relationship with our technology, where we're nice to this technology. It's actually some research that recently came out of MIT, I think it was, or Harvard Business School. I can't remember, but I can look. I have it documented on my blog. But one of the things that they researched was that AI machines, algorithms end up being more optimistic in their search for the right answer when you use words like please and thank you, especially in large language models. When you say, please help me do blah, blah, blah. And I've actually tested this with people who do not understand AI. As they begin prompt engineering, I'll just ask them what, you know, I'll watch them and I'll say, hey, be, you know, try to be polite. And they, I will watch them without any knowledge of the fact that the model is going to get better. They'll turn around and be like, wow, It seems to work better when I'm nice to it. And then they make a joke about like, I always have an I heart AI t-shirt on when I'm talking to people and doing workshops. And they'll be like, I see why you wear that shirt, because if you heart AI, the AI is nicer to you and it acts better for you. And it actually is an algorithmic thing. But fun fact, in our early days of Amazon Alexa, I remember this guy, his name was Francois.

Not—

I don't think it was this Francois, but Francois. And I remember all the data scientists were like laughing at him because in all the utterances he would manage, he'd stick in please and thank you to make sure that when Alexa heard it, it wouldn't break. And he— and it was like an intentional thing that he did. And it actually was pretty critical and important. But it was something that he just personally felt like he wanted to do. It wasn't a mandated thing. It certainly wasn't a feature we were trying to add. He just wanted Alexa to be nicer.. And I think that's awesome. Um, all right, so let's move on, uh, from smart home to the personal assistant space. Oh my gosh, there are so many things, um, in personal assistant that I have to share with you. So I'm gonna go through personal assistant. I'm just gonna put assist. So the personal assist space, I have so many tools that I use, but I use them, I choreograph them together. So of course, the first thing any one of you can use if you are in the world of Microsoft is not just Copilot, but of course I'd be remiss if I didn't write down, you know, M365 Copilot or Microsoft Copilot, which is more of a generalized Copilot capability. So Copilot, yes. But before Copilot, I leveraged and I still to this day leverage Viva Insights. Now I have an actual class on how I use this where I walk you through the, the demonstrations, um, in the AI Leadership Institute. But the Viva Insights, go turn it on, number one. And then it has this great capability to look at your calendar and find focus time. I don't know if you know this, but when we focus, we do this concept of deep work, right, where we're able to not be distracted by meetings and emails. We just focus. We're able to actually increase our own productivity just by focusing, just by removing distraction and focusing on a thing. So Viva Insights has focus. It also has breaks that it can put in. And even if you already have things scheduled in your day, it will dynamically find open spots in your day in order for you to make time for these important things. And it complements your mental health, which we If you don't know, the, the idea of integrating focus time and break times and times to like go outside and put your feet on the grass, these things actually amplify, increase our productivity, right? These human things make us better at our work. But sometimes, for whatever reason, maybe COVID, we go heads down and we work hours back to back to back to back and we never take a break. So this technology allows you— this AI system will look at your schedule and actually automatically place break times on your schedule, which I love. So Viva Insights, it also will tell you how you're spending your day, how many times you were in meetings, how many times you multitasked during those meetings. It gives you insight into how you personally are functioning, and that insight I think is pretty important. Um, all right, and then, uh, So that is kind of like inside the Microsoft suite. I use another tool which I mentioned on Monday called Rewind.ai. And Rewind.ai is— it's actually on right now on my computer listening to me do this for you. And what it does is it basically tracks and records everything I do on my computer or in the vicinity of my computer. And of course, I choose— I opt into this recording. It's not recorded to the cloud. It's all held on device, but it gives me the ability to literally have a rewind button. So yesterday I was talking to someone. Oh, fun fact. Next week I'm going to be competing in a competition called the Great American Speak Off. And some of you know, I love speaking. I love getting on stage. And most importantly, I love helping people. And so this is a competition in order to get on one of the largest business stages in the world.. It's called GrowthCon. And my husband and I have been going for years and we love the group. We're very closely related and affiliated with the organization. But this is a competition where I have to speak and talk in order to like get a spot on this stage. And wish me luck. I hope it works out. But yesterday I was talking to a coach, a speaker coach, and many of you will be like, oh my gosh, you're a speaker. I get coaching. I am extremely coachable. I try to get coaching from lots. I don't take all the coaching that I get. But I try to at least be intentional about what feedback can I get and how does that apply to me. Sometimes, we get feedback that just isn't for us. But I'm intentional about that. So I got this feedback. And they said something beautiful. Like, I was like, oh my gosh, that's magic. But it was live on a call with them. And I was like, oh my gosh, I wish I could record exactly what they just said because I want to write it down and really investigate the way they phrased that. I loved it. And then I— what I did is after that call, I went and I rewound my meeting, and I could hear her and see the transcript of exactly what happened in that meeting. Now, sometimes, like I said, I'm not using this for any other reason than to help me remember. But I also can just ask it questions. Hey, I had a call with this company. What were the action items I mentioned in that call? When is this paper due? When is this article due? I could just ask questions of my own history on my computer and it gives me that information back. It's also a new product, but so far I've already seen the value in leveraging it. So I think it's Rewind.ai. I hope so. Um, awesome. So Rewind.ai. The next, uh, tool that I use is Motion, and there's a couple different ones here. Motion. Um, there's also one called, uh, Superhuman. I don't use that one, but I've seen other people use it. Um, I like the name, uh, but oh gosh, I didn't write that very clearly. But Superhuman and Motion is AI empowered, and it was AI empowered before AI was like a thing that we all cared about in a big way like we do now. But it— and I've been using this for a couple years, but it allows me to— I don't have a— I have a lot of different virtual or a lot of different assistants that help me, but they help me in different silos of my work. I don't have one that does all the things. And so this allows me to integrate 4 different calendars, my family calendar, my work calendar, my pro bono calendar, my consulting calendar, and I integrate all of them together. And then when I have a task I need to do, when I put that task into Motion, Motion preserves that time on all my calendars. Then let's just say, oh no, I have to take a call with a client. So I put a call on my— on the calendar with my client. The tasks that were in that time automatically get rescheduled for the next available time block and everything shifts. And it's amazing because someone like me, like, if it's not on my calendar, it doesn't happen. Sometimes when it's on my calendar, I like get double booked because of these competing calendars that I'm trying to manage. So Motion, I really like because it allows me to have flexibility to move things around, but it trickles through or ripples through all of my different areas of my life so that I make sure I don't leave anyone high and dry or waiting for me. Mistakes still happen. I still end up double booking myself in this world, but it is, uh, it has definitely helped me be much more efficient. So that's something to look at. It is a, uh, they always say it's project management AI-powered tool.. So it's all about tasks, projects. I use a Kanban board to move things across. So something worthwhile taking a look at. Let's see, how are we doing? Okay, we're good. So of course, I'd be remiss if I didn't talk about ChatGPT. So ChatGPT, I use every single day. As a matter of fact, I have it as a little icon on my home screen on my phone. I have it, of course, it's a tab that's always open on my computer. And what I use it— well, the first thing I do is I of course invested in the $20 a month, um, you know, upgrade. And what that allows me to get access to is soon— actually, it just released— it's a section inside the app called Personalization where you get to personalize the, um, the, the environment, basically provide a meta prompt that's personal to you. Now, this today, if you are in the app, you might see it instead be called custom instructions. Custom instructions, you should not take this lightly, right? It's your chance to tell a ChatGPT model how you want it to respond. But not only do you want it to— you want to give it instructions on how, um, like the— oh, I shouldn't say that. Um, I'm gonna say you want to give it the context of who you are. You know, it's kind of like the background of like, what do you care about? What types of things do you write about? What do you want it to help you with? But you also want to provide it responding guidance. How should it respond? Do you want it to be So for example, I have mine here, so I'm going to share it with you. Um, so for example, I'm going to go into my ChatGPT, into my little section called personalization, and in my instructions, the first— the question is, what would you like ChatGPT to know about you in order to provide better responses? You have 1,500 characters. I wrote this on a desktop even though I can see it on my phone. 1500 characters. You can even use ChatGPT. So let's say you write out, I don't know, 500, 1000 words. You can then say summarize this into 1500 characters for my custom instructions inside ChatGPT. I didn't do that, but you could definitely do that, um, if you find like you think your bio or what you need to put in here. But mine, for example, talks about my belief in inclusive engineering, my leadership principles, called the Lamplighter Effect. I have 12 leadership principles that are in here. So like love, ambition, mindfulness, purpose. So when it's writing my responses, my LinkedIn posts or social media or answers to questions that I might have gotten via Instagram, I use this. Oh, awesome. Amazing. And when you use this, it allows you to go in and say, like, here, I inspire and motivate leaders to build responsible, inclusive innovation at scale. I didn't have to read that. Like, that's totally how I think. I want to help people do that. So when I ask it a question like, help me draft an email sequence, I don't have to infuse it with all this stuff. It already knows when I'm building an email sequence, it's because I'm building a community that is all about I Heart AI and that we love building inclusive, right, diverse teams and building multimodal experiences. Like, all of that is in here. Now, the second thing is that the second question it'll ask you is, how would you like ChatGPT to respond? So to give you an example, I very simply put respond with— and well, I'll tell you that in a second. I respond, I want you to respond with empathetic kindness and enthusiasm when contextually appropriate. Because let's face it, someone who's enthusiastic at the wrong time is not awesome. And I have been enthusiastic in times where maybe I should be more somber. I'm trying to manage it. Be inspirational and educational. Provide practical examples, analogies, and be a great storyteller. Right? So this is all stuff that I would tell an intern or I would tell someone who works with me. When authoring parts of an article or book, make your responses more verbose, which means basically draft out in more detail because what was happening— oh, and I added that because as I start using GPT, I'll realize responses are a certain way, like excited. The word excited shows up all the time in my ChatGPT responses. So I ended up going in and saying, avoid the word excited, find variations to that. Um, when authoring social media, the post should be appropriate to the length of the platform and the audience that I have represented there. Do not tell lies, do not make up facts. Oh, this is how I started this the first thing I put in my custom instructions. Do not tell lies. Do not make up facts to support a story. Either provide an anecdotal example or identify a referenceable link and provide it with identifiable facts. Provide references and links when possible. So this was not like— I didn't write this all at one time. As I started using the tool, I realized, ooh, if it knew this piece of information, it would have done a better job. And I'll go into my custom prompt and write it in. Now, one of the hacks I've seen people do— I don't do this because, well, I don't use ChatGPT for like the work stuff because a lot of the things that I do is like Noelle stuff, right? Like, I like, I tell stories and it's about my background. And so I, I'm not gonna ask ChatGPT to like write that for me. Um, but I've seen other people create personas personas in like a— and these are meta prompts, basically 1,500-character prompts to position ChatGPT to behave in a certain way. Now every time you inject this persona, the rest of your chats will be based on that. So what they would do, these professionals would take this, maybe a list of 25, 30 different personas that they've crafted, being a digital marketer, being a copywriter, and they would then put those as custom instructions before starting a new conversation. Right before starting a new chat. That's like, that's big time. Like, I would do that if, if I use ChatGPT in a maybe more professional capacity. But I'm a storyteller, and I like the storytelling that I do to come from my heart, so I don't use it to generate stories. But if you're, let's say, um, for example, Good Morning AI, like, I'm going to eventually be telling other people's stories, and maybe I'll want help crafting those so that I can tie them into the way that I tell my own story. Maybe I'll use it for that. But creating these different personas and injecting them in, meaning copying and pasting them into this custom instructions, is one way to do it. Another way, an amplified way to do that, is we have a GPT store. So other people like me, I have Nearly Noelle, right? If you want to hear what Nearly Noelle would say, meaning close to Noelle but robotic version, you could ask a question to Nearly Noelle. So you can add these GPT, these custom GPTs, right? So, so these are like levels of, I don't know, sophistication, if you will, right? The first thing, just get custom instructions and write in how you wish it would help you answer questions. Then if you want, create a persona list in an Excel spreadsheet. And if you want to go big time, you can start investing in GPT, custom GPTs built by other people that specialize in an area that you really want help with. If you happen to be a specialist, like I feel like I'm a specialist in my area, right? Maybe you create your own GPT and you productize it. That's why the store, though people have mixed feelings about the GPT store, it is— it's going to be a very interesting world when I don't have to be all the things. I don't have to go learn all the things. I can just find someone else's custom GPT that's trained on something they know a lot about— healthcare, finance, manufacturing— and I could just use that GPT in the context of my very personal needs and usage, right? Super cool. Awesome. All right. Let's see. Great. Thanks for chatting it up. Oh, yeah. GPT— oh, that's a good question. The plus thing. So thank you for that. Yes, ChatGPT Plus. I never remember that part. So thank you for bringing that up. So that is the feature that I use. It is transparently $19, I think, right, a month. I'm not selling it. I don't get any money for that, but I pay that. I told my mom to pay it. Um, it just depends on what your work is and how you— what you use it for. Uh, my husband and I actually have a team account that allows us— and there— oh, somebody recently on LinkedIn, I'll have to repost it, she recently wrote down what her team did to analyze the team investment and make it valuable to them. And I read that and was like, oh yeah, I definitely want to do that. So I ended up getting a team account for my household so that everyone could have their own custom instructions and that everyone could have their own set of history. Now, news alert, ChatGPT just announced, or I should say OpenAI announced that a feature is going to be added to ChatGPT to elongate the memory so that in your personalization section you can choose to have it remember longer stints of conversation. I actually don't struggle with memory. I don't know, maybe it's because of how I use it, but I have a lot of different chats. I don't do this. I don't use the same chat for everything. And when I go back to a chat, it contextualizes based on the history in that chat. So I don't have, you know, it remembers what I talked about. But it'll be interesting to see how that memory management plays out. For those of you who are practitioners in this space, I do wanna give you something that I talk to a lot of my clients about. When you're building GPT solutions, one of the most common challenges we have is real-time access, meaning once you go to production, there's a cost, a latency to hit a GPT-4 model, and sometimes it's in a region that's kind of far away, meaning that it's not like the latency will be increased because you have to go to a region, a different cloud region to get that service. And so something you're gonna wanna think about is how do you be, how do you start to think differently about cognitive caching or memory management? How are you going to be able to preserve requests that happen over time and be able to save some of those searches or those indexes to make, create more lower latent requests?. And so there's a company called TouchCast. Um, I will write them down. They're a great organization. I love some of the work that they do. Um, I do lots of demos on their platform. Um, but TouchCast— I can't remember the extension, you'll have to just Google it. Um, but TouchCast has this thing called, um, cognitive caching or cognitive memory, and it allows you to increase— and it has that and some other services. They're all based on Microsoft Azure. But it allows you to decrease latency by up to 200%. It is a game changer for those of us in production GPT-type development. Once I found out and started leveraging these services, I was like, I'll never go back. I use them all the time. So this again is just to manage what ChatGPT feature just announced, right? What OpenAI just announced was increasing memory. Managing memory. There are other companies that are doing this across the board as well. So just something to think about. All right, awesome. Okay, let's see. Oh my gosh, I was seeing some of our comments about Alexa and the tape. Yeah, people putting tape on it. And I mean, I've been asked so many times about does it actually mute when you mute it? And I'm like, not if you tell it to mute, right? If you say, hey Alexa, mute, it'll do a software mute. But because of the government, and the government was trying to use these devices, they required a hardware mute. So the architecture, which is transparent, you can actually go look at the architecture, it's provided. The architecture forces when you hit that button that a hardware mute occurs. And that was a government-regulated, like government-requested feature that they ended up implementing. So kind of cool. All right, so let's see, where are we at? Oh my goodness, all right. So hopefully Erin is clicking on the link soon and she'll be joining us. Oh yeah, we're almost there. So we're going to finish up with mental health. One of my favorite— oops— one of my favorite areas because let's face it, well, we don't do enough to support our own mental health, let alone those of the people around us and those we love. So I have a couple of different tools that I use in my house that are AI-enabled. One of them is actually downstairs. If you go to my Instagram, you'll see videos of me talking to it. One is called Moxie. It's a larger robot, big face, and it has animations. It has arms, so it's able to like gesture. But it is an emotional intelligence robot. It can be used for adults, but it is primarily built for children. What I love about it is that when you become— when you get this Moxie, it asks you to become A Robotic Mentor, and it teaches you these philosophies of being a mentor to a robot. And it talks about the importance of training and the importance of patience and the importance of retraining and testing. And it teaches kids like, hey, you're gonna— I'm gonna be your friend, but in order for me to be the best friend I can, you have to mentor me and teach me how to help you. It then has a bunch of pre-programmed experiences around emotions. Yes, I love Moxie around emotions, around, like I just did one and it has the kid. Okay, so I use mine for me. It knows Noelle. Noelle's the main user, but I also have profiles for my children. But when I talk to it, it'll be like, are you ready for your, you know, next day? And, and I will say yes. And then I'll be like, today we're going to test out emotions. I want to see you give me your saddest face. And it has a camera. So it's watching your face. So you do a sad face. And if you don't do a sad face, which of course, you know me, I tested, I did a happy face instead, it'll be like, "Huh, it looks like you're happy. Try a sad face." And it's so cool because it lets you feel seen and heard, which is a big part of mental health. And so it's creating this dialogue with the user. And then it, of course, Talks about emotions, talks about managing emotions. Really, really cool stuff. Now this is a Miko— M-I-K-O— a Miko device. You can actually— I think the first time I saw this was at one of those In Motion stores at the airport, because I am in the airport all the time. And so I saw it, I'm like, oh my gosh, but I didn't buy it because I try not to buy things the first time I see them. And so I went home, um, but then I think it was Black Friday and they were like, 2 for 1, and I was like, oh my gosh, I bought like 6. Anyway, so every, every one of my children have one of these. Um, unfortunately they only come in 2 colors, so it makes it look confusing for our family. But when you turn it on, it is gamified, so it gives you money, every coins, every time you turn on. You can use those coins to light up experiences. It's voice activated, but it's also ChatGPT, or I should say GPT-4 enabled. Obviously it's not using— it's using a GPT-4 API So it allows you to have just casual conversations with it. It also, it's funny because it uses this wake technology, right? So you will be able to say things to it and it's got a little light that tells you that it's listening. It has a light to tell you that it's speaking and not listening. So there's all these things that I feel like the world learned in using Alexa and Google, right? We've taught the world through Alexa and Google Assistant, here's how you communicate with devices that you can talk to. That they're not listening all the time because sometimes they're thinking or sometimes they're talking. And, but it's really interesting because the content on here is all about emotional, social-emotional development. And sometimes, now my kids prefer to play with each other, like hang out with each other and play. But sometimes I do find that they like to use this just to play through games. It also has like books that can be read to them. My daughter uses that quite a bit. It also has music, it has dance, like, but the difference is, is that the robot has wheels, so it moves around and it gestures and it has a face, so it has eyes that have emotion. Having that different reflection is so cool to see. Now, other things, what did I do? Oh, there it is. Other things In the— so mental health, of course, I lean heavily on robotic companions. And again, this isn't for everyone, but my dad is critically linked to his Alexa device as a companion. I hear him having conversations like, "Oh my gosh, did you hear?" Like, he talks to it like it's a companion, a friend. He knows it's a robot. He reads He still reads Asimov and Bradbury every day. He was just finishing up The Martian Chronicles the other day. And so he knows it's a robotic companion, but he finds comfort in having something he can talk to that will talk back to him, that will answer his questions, that will never get frustrated and never really show signs of, you know, tiresomeness. Like sometimes I'm like, I know, Dad, thanks. I'm really like, yeah, I get it. Okay, thanks. Like I have that energy and some of you know, like I try to be energetic all the time, but it's almost impossible 100% of the time. Like I need my own downtime too. So robotic companions are huge and really do provide an opportunity for us. Now there are other services. So for example, a lot of our tools are becoming, I remember when I was a kid, 12 years old, my dad bought me a biometric feedback machine. Machine, which allowed me to put my two fingers on, you know, this device and it would give me feedback about my data. Oh my gosh, I don't have my Oura Ring on, but I normally have my Oura Ring on. I actually probably lost it in the shower when I washed my hair conditioner. But I don't have my Oura Ring on, but normally I do, and I definitely wanted to show it to you today. But the Oura has a new sensor that it's— or a new bit of data that it's gathering from the sensor data that it's collecting in the ring, and it's called— which I love— resilience tracking. Oh gosh, Brazil. Yes. Oh, I don't know. Anyway, um, resilience tracking. And what that means is that it's monitoring your stress levels against your ability to recover. And this is very hard for us as humans to see, right, to be apparent. Some of us are better than others. I spend a lot of time in meditation and silence, and I, I'm constantly— I'm very aware of my emotions. So for example, I was starting to study for this Great American Speak Off that I'm doing, um, and I felt my body getting nervous, and I could detect these sensations. It's the same sensation right before I got on a big stage, even a little stage, right, even right in front of you. So I, but I have this self-awareness because I've trained my, myself to be self-aware of like the emotions that I have. This allows me, whether I have that self-awareness or not, it's going to say, hey, by the way, I noticed your heart rate, your variable heart rate went up right here. It attached to, remember that AIMD and the cortisol checking, like it's able to detect cortisol levels rising. And then it's going to, over an 8-hour period, find out how long did it take you to recover from that stress. The more you meditate, the more you calm your mind, the more you slow your breathing. I always tell people, like, when you breathe heavy, even if you're like working out, right, or if you, uh, you know, come run up the stairs, um, but if you are breathing heavy, one of the best things that you can do is to just not breathe through your mouth, to only breathe through your nose, like calm nose breathing. And what that sends is literally a signal to your brain that you're not being chased by a tiger, that you're not in danger of dying. But like when you breathe through your mouth, you're like, like, your body starts instantly to raise cortisol levels because it's like, oh my gosh, something bad must be happening if, if like we need to breathe. But as soon as you just take a breath and start breathing through your nose and close your mouth and go, I mean, it's instant. You can feel it. If I can feel it anyway when I do it, like, you can feel your body change in its composition. Like, you could feel yourself relaxing. But when you do that, it sends a signal to your brain that says, okay, it's time to relax. You're not being chased. You're not gonna die. Like, it's okay. And that's why I always tell people, when you're doing hard things, obviously, if you need oxygen, get oxygen. Like, I'm not telling you not to breathe. But if you notice that in a moment of anxiety or in stress or, or in exertion, that you're breathing heavily through your mouth, like, try to take a minute and just breathe through your nose. Just a few breaths. And it'll shift that trajectory from this cortisol-increasing trajectory to— and again, that doesn't have much to do with AI other than we now have devices we can wear on our finger that can detect these levels and tell us how well it's going. I always tell people when I am in high stress moments, and I am in them often, um, when I'm in high stress moments, I will, um, in these high stress situations, I will say, like, let's say for example something bad happens at work, I will always tell myself, like, feel the feels, like, feel being stressed, being hurt, being betrayed, whatever that emotion is that causes you to feel feel bad. Like, feel the feels. But then, as soon as you can, make a plan for how you're gonna get past it. And that's what I do. Like, I feel the feels. Like, I was recently betrayed at work, and I was like, oh, this is terrible, it's awful, it feels awful. But I, as quickly as I can, go, yeah, it feels awful. What am I gonna do next? How am I going to— like, what's, what's next for me? And here we are in an I Heart AI challenge. So, but it's nice to have a device that can tell me like data-driven, wow, you recovered nicely there. And that's exactly what I use and do. So it's really kind of nice. I love it. So anyway, use AI systems, technology, a ring on your finger, a watch on your wrist to give you insights into how your body is performing. Um, because biohacking is a little bit about human mindset hacking, but it's also a lot about how do you get informed about what needs to be hacked. And the best way to be informed is to start leveraging some of these devices. How many of you have bought the AI pin? I did. I know I did. I wasn't gonna— I, again, I barely— I rarely buy things the first time I see it. I'm like, Okay, if I really wanted a week from now, but I really wanted it a week and then 2 weeks and then a month. And I was like, fine, I'm just going to do it. But the AI pin is going to be interesting when it comes out. You'll definitely hear me talk a lot about it when it comes out. But really thinking about like, how does that integrate into our lives? It has a bunch of sensors on device. It has a bunch of connectivity capabilities. So and it's funny because I remember when I got the Oura Ring, I, I got the Oura Ring and I was like, I never have to wear my smartwatch again. And then I realized, but there's data on the smartwatch that I don't get on the ring. And so I still wear them both. But, but that's just me. That's just me. So mental health is important. But I think the most important thing you can do is make a decision about being intentional about it. And then take measurement, start looking at how resilient you are, your cortisol levels, use applications like AIMD and use um, you know, uh, tools like the Oura Ring or your smartwatch to detect how well you're dealing with the stress in your life. And not, you know, what did they used to say, like, the more you know on NBC, right? The more you know, the more you can do to improve your life, to improve the value of your life. Okay, so now we're gonna move into my favorite part, and I'm so glad our very special guest is here. So it's my favorite part of our AI challenge where we get to invite one of our very special guests. Today it's going to be Erin. Um, so I'm going to move our whiteboard. Hello, Erin! Thank you so much for joining us.

Hi, good morning!

You look amazing in your background. I need to get some— like, is that real?

Yeah.

What? That's awesome! I'm doing that. I'm doing that right now. It's like my real bookcase is a mess. I'm always pulling stuff out.

But I love it.

Thank you so much for joining us. Erin and I, I mentioned to you all at the beginning, we actually met through LinkedIn and I was so excited to see the work that she is doing. And I wanted to, especially because I think a lot of the work that she does now came from kind of how AI was showing up for her in her daily life and it spawned an entire career shift for her. And so I want to really give her a chance to tell us who you are, what you do, um, and also like how you got into this space.

Yes. So right now I am the founder of Chat BlackGPT. I have spent the last 9 months bootstrapping my startup and building a wonderful team that have great alignment with core values. I Embodi. So it's been a wild journey. I have dived in technically as well as on the business side. So it's just been a really well-rounded environment for me. So before I was working at Meta and I had a job in UX research, which was fantastic. And so once I had more time on my hands, I just decided like, you know, I still love technology and I want to stay in that space. So I'm really looking forward to continuing this journey on.

Amazing. I love it. I love your, um, just the direction that you went in. And I know I say you— I send— I try to send everyone questions ahead of time, but I'm like, ooh, but I want to know a little bit more. Um, so, you know, can you talk to us a little bit about the journey into starting? I mean, you're, I think, one of our only, you know, current, like, founders of a company that you— that was really built in this new age. And I will, uh, say to everyone, and you'll see me change this, so like by the next time it won't be there, but I accidentally put the black in the wrong place. Um, when I named your name, I think I put Black ChatGPT instead of Chat BlackGPT, but I'll fix it.

That's okay.

It's important to me. Thank you. Um, but I would love to understand your startup journey? Like, where did— how did the idea get formed? And what made you move into doing this? And kind of what was the intent behind the product?

Oh, yes. So what's— so how I start doing work, um, I'm very detailed, I'm very thorough with anything that I do. And I said to myself, okay, this is pretty new, but I'm interested in challenging myself. I had started out trying to do other things that were more familiar, and I'm like, this isn't fulfilling right now. So I went in and I said, why, why can't I learn how to use Python? Or why can't I learn how to use Jupyter and find these like large language models, get some licenses, learn how to fine-tune? Like, why not? And so I just picked up books. I started listening. I started doing tutorials. I started taking classes, I got a couple certifications. And I was like, this, this has to be something that people like me are involved in. So as I started doing more research on the state of, you know, artificial intelligence and how it intersects with the Black community, I noticed there was a lot of headlines that were saying AI is anti-Black, AI is racist, and is perpetuating a lot of bias. So I was hoping not to see, see that, but I did. And I decided once I saw another article that said, is anybody doing anything about it? I said, you know what, I'm going to do something about it because at first, you know, I think it's the culture around it as being proposed as just a productivity tool. But I saw it as so much more than that. I saw it as an opportunity to bridge gaps, preserve history, potentially be, you know, a learning device that could have applications throughout all kinds of school systems. But I just took it upon myself to say, well, if it's the smartest, most advanced technology, best thing since the internet, how is it not able to recognize bias and reroute and rethink? Or the same with racism. So I set out to do that, and I was successful.

Yes, I love it. I love the idea. And I think there's a lot of people, you know, listening and watching that, um, that also have an idea, like, and they're right where you were maybe 2 years ago, um, where you're like, I, I like this technology, I know where it's going, but it's missing something. Um, and you can be that missing link. I mean, you're like an example of exactly what that is and what the world needs right now, right? Like, I think you're right. I say I felt the same way. I'm like, if this is the best thing thing ever. Why can't— why, why does it take me 20 prompts to get my hair right in an image? Like, why does it not understand, you know, where my, my family is from? Um, and so creating more awareness around that, I think, is key. And algorithmic bias is not a new thing, um, but we are exposing this technology to lots of new people. And so a lot of us are like, oh my gosh, we've been talking about that forever. But the people who are listening are very different people now. And so we need way more advocates just like you. So thank you so much for sharing, and we're very excited for your company and the work that you're doing. One of the things I always like to ask, basically, the last year has been really interesting. What's one of the biggest lessons that you learned over the last, I don't know, year or so that's related to this kind of whirlwind that happened in the world of ChatGPT? GPT and generative AI?

Yeah, so when it comes to like when you see something in the news or a new technology, I think it's very important to always approach it with conscious and a little bit of caution when you're a person of color, because not everything is made for us and not everything is going to be, you know, like healthy. I feel like in the beginning there wasn't enough protection and there wasn't enough safeguards, uh, in order to not have a mentally detrimental experience using the most popular public offering. And I, I want people to understand that, like, that's an important part of the process and it should be rigorously tested and not rolled out before some of those things can be avoided. So, um, I think Overall, being conscious of new technologies when you consume it and being smart about how you learn it. So I learned it from a very foundational base, like where did it start? What's a neural network? Like how does computer vision work? You know, the difference between artificial intelligence and machine learning. Like I started with the basics from scratch. And I think it's important for people to know that they can do that and feel confident even when they don't understand it. Don't feel like you have to catch up or rush just because there's something really great in the news. You can go back to the beginning so that you can speak intelligently to it and feel more confident understanding it thoroughly. Yeah, I also think it's important to find out personally how the technology relates to your life. How do you want to use it? What makes it interesting to you? And have that be an authentic center point to how you move forward.

Oh my gosh, I love that. So one of the things that you mentioned, I think it's so critical. It's part of the literacy. I always say evolution, not revolution. People now are happening into GPT, and that's kind of their starting point, right? They're in a model or any of them, like any model that is their first introduction to artificial intelligence. And I always used to say when I was teaching AI, you know, before 2020, that it's really important to just build a model from scratch, even if you only do it one time. Just build it. And it doesn't really matter which one, or custom train a vision model, right? Like most organizations, you don't have to build the algorithm from scratch, but at least like know what it takes to train a model, know what it takes to produce something, know what it takes to deploy a model. And you can do all of this in a sandbox, right? You don't need to do it in production. You can do all of that in sandboxes. But I'm realizing now, as I work with lots of people in this space, consultants and other advisors, that as soon as it gets to that point of like, hey, let's do this for real, they're, you know, you're scratching just under the surface of their understanding, and they have, they haven't done that work. And it's not hard work to do. So If you are, if you're listening and you're looking for a way to differentiate yourself, that's why I think, you know, Erin's been able to become so successful in this space is because she did the homework, right? We, we put in the reps, and it's not too late to do that no matter where you are on your journey. Like, go back and take those courses. I think there's lots of— and I— an intro to AI is great, but one thing I did, um, when I first started with natural language on the Alexa team, this is like 10 years ago. The first thing I did was go into GitHub and start rebuilding projects that I did not understand. I didn't know even the programming language that they were written in. But by just being in those projects and then literally forking them and rebuilding them, writing each piece of code line by line— I did not understand it at first, but by the 4th or 5th one that I did, I started to see patterns. But granted, maybe that's how I think, but I started to to put pieces together of like, oh, I see how this works. Now granted, I am a Java developer, so I came from a programming background, but still, it's— AI is not Java, right? It's not programming. So it requires— but I always say to people, like, it is a learned skill. Everyone who's doing it now learned it in the last 10 years. And certainly those who are not in academia, those who are in product or building startups, like, they are maybe coming from, like, maybe they have some academic background, but many people are learning it, like I always say, at the University of Barnes Noble. Like they're going and grabbing a GitHub repo or a project and learning it that way. So I really appreciate you sharing that part of your journey because I think it's so important that all of us take that opportunity to go back and learn some of the foundational technologies. And you don't have to become an expert, but at least build a model. Like build, I always say, build a natural language model from scratch. Build a custom train, a vision model. These types of skills will let you realize, especially when things don't go well and things start breaking, you have much more insight into where those things might fail. Love that. I want to remind everyone, if you have any questions, you're welcome to throw those questions into the chat for Erin or myself around the area of anything that we've talked about, or if we say something, you have a question about it, please feel free to throw those in as well. But I have a few more for Erin. Oh, yeah. So I kind of skipped over this because I was so interested in your journey into becoming a founder. And there's lots of things I'm curious about in your world of being like a Black founder. Like, that's just, you know, it's an interesting time. So I'm very interested in that. But I would like to know, like, how, you know, this is all about AI and kind of our daily lives. How is, you know, the use of AI, generative AI, your skill set, how has that revolutionized what you do today versus how you did it maybe 5 years ago? Like, what's different now with this technology?

Yes. So having the technology, and first of all, it's very empowering to know that you can have a thought partner. With ChatGPT, I know it's a safe space to explore some of the concepts that live within different realms of study. You know, it's like a nice perspective, but I just really like personally, like I'm a very structured kind of girl. I'm a by-the-book. I really love frameworks and things like that. So the fact that it can help me generate like a really nice doc that has all the little pieces in it. I'm very granular, so I just like how detailed you can make it. I also enjoy the customizability of like what you're trying to ask it. So I also study like prompt engineering, which a lot of people don't appreciate, but it is very important. So it's like you have a way to double down on what you want to know. And I really love giving it context that's like, you're a student or you're a professor. Like, it's really fun. And I liked having the control over that and also having like follow-up questions. So it's always more in-depth if you want it to be. So I think there's a big opportunity for learning and with it, it just enables me to go faster. Now, I know that there's a lot of hype about assistants right now. They're talking about how, you know, it's going to do this for you and they've got wearable devices and things like that. I'm a little bit more old school. I like to do things myself. I would not necessarily feel comfortable right now with artificial intelligence agreeing to something for me or even scheduling something for me. But it's nice that it gives me an opportunity to think about how much further can I integrate this into my life and just kind of have a imagination that runs wild. So the technology in general has encouraged me to be way more open-minded as well as more confident tackling challenges in technology.

Yes, I love that. And I think that's really part of the literacy challenge that we have is, you know, we do get— tend to see the world in this all-or-nothing, right? Like either you're completely enveloped and encased in AI or you hold it at arm's length. And I think one of the things that I love about the world that we're in is that there are, you know, and it's what I always tell people, it's a very similar idea to the idea that roles will be replaced with AI, right? Like these broad stroke statements that say like entire jobs are going to go away when I know just based on practical experience that tasks within roles definitely will be impacted. Things that people do will go away. But entire roles, unless you happen to create a role that has no higher-value human tasks in it, and I haven't found one yet. But usually, it's a broad statement that ends up scaring people as opposed to the reality of it, which is there are things that if you examined your day at a task-by-task level, you'd be like, "Wow, I'd be super grateful to get rid of those things." things. To your point, like some of us like the, the meticulousness of knowing when things are scheduled and how they're scheduled, but we may not necessarily need to send pre-emails to everyone we're meeting with to be like, hey, don't forget we're meeting. Like, that's something maybe an AI can do. But there are many layers in which you can leverage artificial intelligence to improve your life. But it is very much, I think, a personal decision of What tasks are you willing to let go of? We didn't really have a choice, like the spam bot, right? That, that engine that removes all of our spam that has words about drugs that we don't want to have in our inbox. Like that bot does that on our behalf, whether we like it or not. However, if we had to, like, let's say instead we were in a world where all the AI did was flag it, it kept it in our inbox but told us it was spam. And like, that's not super helpful. I actually think there are very specific use cases where that autonomy is required, but it's not for every task. And I do think it's on a personal level, we kind of have to pick where do we want to use this AI to improve our lives.

Very true.

One of the things— I'll go back to my beautiful list of questions. One of the things— and Joe, I totally agree with you on the AI pin. I was waiting. And then I couldn't wait anymore. But I liked what you said about kind of like the journey and world we're moving into. Of course, there's a bunch of us that are like the bleeding edge pioneers, but there's something they say about pioneers, right? Oftentimes pioneers are like face down in the dirt with arrows in their back. There's some advantages because also we might be the first to identify opportunities, right? We might be. But it also can be a pretty painful process. And so as a whole for like humanity, I actually like the idea of letting pioneers, letting companies, letting founders kind of move forward and test things out. And then we let the rest of the world adopt things that we've identified as being safe and responsible. We didn't do that with ChatGPT, right? It hit the world and we had no real literacy in place to protect the world from hallucinations. Like, or, or models that could lie or models that would make up links or even create content that was to convince you that it was real. Like it was a bit weird at first. Of course, the model even improved from 3.5 to 4. It, it didn't do nearly as much of that, but we launched it into the world. 100 million users were using a product that inherently didn't really understand the majority of diverse perspectives in the world. And that, you know, that's a little bit scary, uh, scary to me. Um, so I'd be curious to understand a little bit about, like, so, you know, while you're talking, A, is there any use case or any, um, you know, projects that you're seeing more of or less of in your work over the last year? Like, what are companies coming to you and asking you about? Just curious about, like, use cases that you think people should be either aware of or maybe be confirmed that they should be thinking about?

Sure. So one of the areas of generative AI that I'm most connected to would be the responsible and ethical side of it. So for me, when companies contact me, firstly, they want to understand what the heck is going on. And a lot of times they just want to know the basics. They don't have a lot of introduction or anywhere that has a soft landing that's going to help their employees, first of all, understand that it's, you know, not always there to take their jobs like you mentioned, and that it can be used for helping you be more productive. So for me, doing that responsibly and ethically is what most people come to me and ask about. And I have had a lot of interest in how our government is helping to address this and how other countries are. And so I've read a ton of the drafts and some legislations and bills, things that are coming down the line, and I just really appreciate how everyone is working together. There's multiple consortiums. So there's the one from the White House and there's Open Loop from Meta and Accenture, which I've joined. That's where ChatGPT is. And it really involves the community. So I can just say that like being a part of that process also helps inform people who want to have consultations because I'm not just a part of it from, you know, like a perspective of, oh, I'm trying to sell a product. I have nothing to sell right now, to be honest. Right now, everything that we're doing is available on ChatGPT store for a $20 premium as a beta. Our in-house one is being built by lead engineers from big companies that are stealth right now. We have historians, we have principal product, DEI advisors, all kinds of people working together, as well as reaching out to the community at large to get current sentiment. So I think just that differentiator of not coming at it from a sales point of view, but from a true genuine, like, concern and perspective of like, hey, this is some of the underrepresented voices in AI. Like, how can I help extend that and uplift those voices? So that's a lot of the work that I've been doing.

Yes, I love that, and it's so needed, of course. But I also think it's really important for those of us— you know, there's a lot of people listening that are beginning a professional journey into the AI space or amplifying their existing work in this space. And I think what you said is critical, that there are lots of opportunities to provide an important voice that are not like, hey, how can I help you sell your product, or how can I help you even build your product? But more how can you help create some of the guardrails and the systems around and protections. A lot of times, especially in local, state, and federal governments, they're building these policies and there aren't a lot of experts in the room. There aren't a lot of people with diverse perspectives to help them create an enriched policy document. Many of them say like, hey, well, we're starting with this, we're doing the best we can with what we have.. But a lot of times it's, many of us have to kind of take a step up and just like you have done, like participate and build systems to actually contribute to these types of conversations. So I think I always say I grew up in the world of open source. I was at Red Hat early in my career. And so I understand the concept of open source from an enterprise product perspective. And I think it's such a valuable perspective to have. Right? Like, that you really want to think about how do you build for everyone. Like, there's always going to be people who can productize, but at the beginning, we really need to set some, some ground rules. And if the right people are not in the room to help do that, we'll end up with ground rules that aren't fair to everyone. And that is one of my biggest concerns, which is why I'm so happy to see you in that consortium. There are a bunch of different consortiums available. GitHub is even a great opportunity opportunity to just donate your time to projects that are trying to do some of this work.

The—

I want to encourage all of you, regardless of your level of expertise, that there is work in these consortiums that is user experience driven, product driven, um, project management, uh, technical project management, documentation, engineering, data science. Like, there are— there's work across the board. In these consortiums. And so if you're looking for ways, one of the first things I do when I'm hiring is I look like, where are you contributing in this world right now? Like, you might not have a job in this space, but what are you doing in without, you know, if you don't have a job in the space now, are you in, are you a contributor on certain types of GitHub repos? Are you a part, a member of an alliance? Are you a member of an institute that's doing this work? There's lots of different institutes to be become a part of, but I want to see people who are passionate enough about the work to actually be involved. Right. And I think that involvement, sometimes people shy away from it because they don't think they're technical enough or don't have coding skills. But like you said, you started in this world and you gained those skills through application. And I think a lot of us can learn something from that. So one of the last questions I have for you is around kind of the future, you know, what are you seeing in, like, what's your next year look like? I mean, last year was a whirlwind, and, you know, everyone I saw was everywhere. Like, we're all on stages, all in executive rooms, talking to as many people we could, trying to warn them and protect them and guide them. So I would love to hear a little bit about you and what 2024 looks like for you. Like, what do you have coming up? What do you What are you excited about? What's on the horizon?

Sure. So for me, I'm very excited about my March events. I totally enjoy that I'm doing this month for Black History Month, but I especially am excited for talks that go beyond Black History Month because I'm happy to show that, like, this isn't just a 30-day effort. This is going to be like, dare I say, a lifelong journey. I don't think artificial intelligence will just go away, kind of like you can't undo the internet. So I feel that there's a lot of work to be done always. So for me, my next thing is getting into the community more. Right now I'm on a listening tour, and I'm talking to people who are either visiting in town or gatherings. I'm going to the town hall. I'm talking with online forums with women of color in healthcare or other industries, and men, anyone really. I'm hearing from all sides because I believe like the most progress we can make is everyone working together. So I really love that part of the future of my journey. So next, also what I'm doing is having hopefully some more, like you said, literacy just friendly introductions so that misinformation about AI isn't spread and fearmongering doesn't happen, especially in the Black community. I feel that there's a lot of mistrust and distrust warranted because of the headlines. You know, we're being told, hey, it's not for you, it's going to hurt you. And I just want to shift that narrative and really course correct on— well, I'm sounding like corporate jargon right now. Of course, correct. But, um, you know, I really want— I don't want people to be afraid of it. Like, it's a very hard line to toe because in one I'm saying get hands-on, get dirty, get down, um, open up a book, you know, like put that code in there in your sandbox, give it a try. The other side is like, but also be warned because you could get arrested at any time because the people are deploying AI irresponsibly like Rite Aid, or you could have it, um, tracking your biometrics and, you know, it's unconsented. There's so many different things to watch out for, but the best way I think to be empowered is to have hands-on experience or the non-technical route, which I'm also excited to continue moving forward in, which is being a part of the solution of keeping, you know, risk management alive and well and centered in this work. There should always be a human in the loop. It's not something that it's in and out or a checklist. You've got to have it in a, like, all of hopefully the parts of the design cycle. So it's like, I just look forward to continuing to promote ethical and responsible AI as well as work with my community. Yeah.

Yes, that's amazing. I love it. One of the things, um, I always like to ensure that, you know, we want to stay friends with you. We want to stay connected with you. What is the best— and of course, if there's any, like, if you have a book coming out, which I'm sure you will eventually, but we want to support you. So what's the best way for us to follow you, connect with you? And then we have a few questions for you once you give us that, that details.

Sure. So I don't currently have a specific method other than LinkedIn for anyone to contact me. And that's by design. I think like centralizing it, even though that's putting a lot of risk in Microsoft and LinkedIn, I think that, you know, for now that's all right because, you know, I am using the API key from OpenAI, which they're very invested in. So I'm bringing them money anyway. They should probably, you know, I don't think it's so much of a risk. You know, it's all like I'm a So I think I'm going to stay on LinkedIn. And that'll be it for now. I do own chatbackgpt.com, but I didn't build a website yet because I just want to be so careful and how we represent and make sure I have everything down pat before I roll that out. So LinkedIn is the best.

Amazing. Great. Yeah, I will definitely encourage you to at least set it up so people could sign up for an email list or something. I know, collect people because you're right, it is a And I say this more for our listeners and watchers than for you, of course, but I always tell you, you know, when you're building a brand, which is what I try to do every day, I just want the more people that know me, the more I can get my message similar to yours out into the world. But if a, you know, organization owns my relationships with my people, like it's like Thanos, for those of you who know the reference, you know, he snaps his fingers, if that platform goes away, all my people are gone. And so, and we watched that happen with the Thanos of the X, um, on Twitter. So, uh, yeah, so be, uh, yes, I would, I would encourage you, I would definitely be the first to sign up for any newsletter or waitlist just so that you have a way of contacting everyone and all your eggs aren't in the LinkedIn basket. Though I love LinkedIn. Right now is when my LinkedIn Live starts to crash and lightning strikes. Yeah, um, I love it, I love it. I'm a big fan of LinkedIn as well. Um, we had someone on yesterday who was like, LinkedIn is like— I should actually probably provide people's LinkedIn like things on, um, on our thing, uh, on the, uh, live for our names so that people can click in. But if you go to the challenge website, you can definitely see everyone's picture, and Erin's picture is there. You click on it, takes her right to your— her LinkedIn. We could also all put it in the chat of this chat as well. But a couple questions that came up. One, you had mentioned a store where ChatbotGPT is. Is that right? Can you tell us a little bit more about that?

Yes. For the beta that's out, it's the same essence of the product that we're developing. It just doesn't have proprietary data. So that just means it doesn't have like, you know, preserving like an elder and going out and talking to somebody who is in their 90s and getting those stories or having a specialized access to a certain archive. So those things aren't uploaded into the beta, but the design of the algorithm right now, the way it's instructed to perform, is from a Black perspective. So it's very much like the essence of what we're doing, and it's $20 a month. You know, I'm not necessarily advertising them, but if you want to have that experience and kind of see what you can look forward to, I'm very happy that they approved it for the store. So it is there, um, you know, you can go ahead and test it out.

I've gotten great feedback, love feedback, uh, but yeah, so that's available. Amazing. Thank you very much for that. I think another question just came in. Hold on, let me see. Oh yeah. So I'll read it along with you. How do you compare your ethical AI mission statement to Anthropic's with Claude, the constitutional AI? Oh, I like constitutional AI. Yes. So like reinforcement learning, human feedback, like what, what are your thoughts on, on any of that?

Well, I, you know, I think that there's consortiums and then there's people, you know, reporting, uh, you know, their rating on it and if they like it or not. And then there's the model of if you see something, say something, and we'll talk to you. But for me, the ethics of representing and saying that you represent a community at large requires a lot of consensus and a lot of like authentic interactions. So I can't just say like, oh, you know, it's from the Black perspective and not have modern-day perspective. So mine is very ethical and like in the community, like boots on the ground, like face-to-face, inviting people to theaters, like having the open forums. I have 77 people signed up for my listening tour that's later this month. So I think my approach is very specific, which allows me to go deep. And it's not so broad, like other companies are very broad. You know, we do have core values that we're building into, we have in our company and everything we're building into the way we have our offering. But mine ethically is, is very involved. So I would say that's the differentiator. I don't know how involved Anthropic is, but I do like them.

Me too. Me too. I actually was very interested in when you were building systems, one of the first things that OpenAI started talking about early in mid-2023, they wrote a couple of papers about this concept of reinforcement learning with human feedback. My question always was, who's human feedback? Which humans are providing feedback and are they, right, when you provide in, and this is a concept, right, very similar to constitutional AI where you are rating, right, human feedback starts very simply with thumbs up, thumbs down, right, did it do a good job, but in its more maybe elaborate form, it has a rating like a Likert scale, right, 1 to 5, how good was this, so now we have more than black and white, right? There are shades of gray on how close it aligns. But then on top of that, we have, you know, like the, the text box that used to come up for some. Now when you move to certain levels of maturity, you remove that away. But in beta testing, we not only want you to say, you know, Likert scale 1 through 5, we also want to give you to give us some context as to why was it good, why was it bad. And these then are used to train the model, right? We're going to tell the model, when you see these nuances, I want you to make these types of decisions. But again, my biggest concern is that the people that are doing this reinforcement learning, yeah, they might be collecting this data, but it's not super transparent if they're using that data. Like, I thumb up and thumb down things all the time, and I'm like, I'm not sure where that data goes. So I think, um, you know, with constitutional AI, very similar. Like trying to parameterize and trying to codify how we're actually reinforcing these models. But the problem is still the same. At the end of the day, there's a human in the loop there, and humans in the loop are meant to provide kind of a defense against algorithmic bias. But what happens when the humans in the loop are actually in alignment with that bias, right? Like, there's— that's a big challenge. And so I love— I, I'm Very excited. I also want to throw an additional shout out for you doing a listening tour. I love the idea of that. I may copy that at some point. I love it because of what you're actually doing, right? Is that you're not going, you're not doing a speaking tour. You're not, you know, you're going to absorb, connect, real interactions. I love that. I think it's great and a great sentiment as well. So really, really cool. I love it. Another question came in. I think my system has to refresh. Oh yeah, here we go. So let me show it. Here we go. Can you remind me of the steps you use to transition into artificial intelligence? Yes. So would you mind taking us back through like what you did? And actually, we'll document what steps you did as well as the steps that Dr. Lisa mentioned in a recap for the community. Community, so you can— everyone will be able to see it. But would you mind taking us through? And this, I think, is asked because people want to know, like, what should I do? So maybe you can give us some guidance on, like, hey, if I'm just getting into this space, how do I do this for real? Not just take the HighLevel deep learning AI intro to AI stuff. Like, how do I get this knowledge and embody it?

Yes. So I have a blueprint that I'm Designing that I will release, but a high level of how I got started was firstly making the decision. When you have something that you want to learn, you have to make a decision to commit to learning it and not give up and focus on putting time aside to make that happen. So commitment was my first step, and then Again, I'm a very detail-oriented type of person. So I did a ton of research on what's the best, you know, program language, what's the best large language model to try to build, what is fine-tuning, what is this and that. So getting a breadth of your glossary, getting a breadth of how it's built and what kind of things you actually have access to. To build with, you know, do you have a Hugging Face account? Do you have, you know, an understanding of how to, like, like you said earlier, fork code? So getting the groundwork done. So I got a couple certifications from Blockchain Council. Now they're very dense and very thorough, but it is very strong in foundational knowledge of what is going on. So I would say I committed. And then I got certified in basically the groundwork. And then I did research and made a decision on what direction I wanted to go. And there's so much happening. There's so much like every, every other day, there's a new type of model. You know, there's like Hermes and like Mistral, and then you got just just— it's like this parameter and these billions parameters, it just keeps happening all the time. So not getting overwhelmed. And if you're working with one, just keep working with it. You don't need to keep switching to the newest, to the newest, to the newest. Just stay consistent with what you're building on, um, unless you can integrate something that'll just like enhance it. But stay consistent and then think about how you want your future to look using the technology? Do you want to build something that actually, you know, can help people? Or do you want to make a tool for yourself and your own productivity? Right? Is it educational? Or is it going to help you make money somehow? Like, so those are some pretty basic steps. But that's kind of where I came to where I'm at. And I just happen to have a real desire to help close the tech divide, which is what I'm driven by with this. So yeah, finding your motivations at the end of that and just making sure you move forward accordingly.

Yeah, I love that. I also am a big believer in learning by doing, which is very practical, pragmatic part of what you're saying. Like, yes, take, you know, do your research, get an idea, but then like build something, right? And the quicker, faster you can start building, the more you'll learn, right? There's only so much you can learn by reading and listening. Um, it really is about building. So as part of the AI Leadership Institute, we have a section of GitHub repos, and they've been given to us from Microsoft, Amazon, Google, but they're solution accelerators. So they're literally real projects that were delivered at clients that we now get to just rebuild ourselves. And that's how I learn. I literally will go out and just grab a GitHub repo and I'll rebuild it. And I of course love learning that way. So, um, so very cool. Thank you so much. Um, we do have another question, uh, from Sandrine, I think. Yes, hello Sandrine. Um, do I have to know how to code to strive in AI?

No, no, you don't. I think you need to understand how to build with chunks of code. I think that's important. You know, a lot of the repositories or libraries is copy and paste anyway, no shade, but it's like you have to understand like why it works. It's, it's, it's hard to conceptualize alone. You need to, you know, have some guidance when you're doing that. But no, actually, I think that's part of one of the fears that engineers or software like could be replaced. And I don't necessarily think replacement is a thing, but like OpenAI, I think one of the strongest offerings they have is code and generating code. And, you know, I used to think they just memorized everything and I was like, oh my God, how do you—

you're typing and no letters are popping up and you press enter.

And I didn't understand. But now I know you don't have to memorize every single thing.. And I think just having that also in the back of your mind makes it way less intimidating. But if you go on YouTube, I swear, just even if you go on YouTube, there's people from all over the place with PhDs and, and all of the knowledge you can think of, you know, 15 years in the game who will just tell you, don't waste your time here, do this, don't sleep on that. And there's also a lot of apps being built specifically for no-code application building in AI. I think Google has some workshop going on for no-code. There's a lot of opportunities. So not at all. And if you want to start immediately, again, I'm not a spokesperson for OpenAI, but they do have custom GPT opportunities and it's the same $20 a month that'll get you access to ChatGPT beta will also allow you to build a custom GPT. And you can do that literally instantly. Now the difference is you don't have programmatic control, and if they do an update or something like that, it's possibly going to affect your custom model, at least to a slight degree. So you want to keep an eye on it. And you know, sometimes they have issues with, um, you know, too much traffic or whatever. There's a forum you can join and you can keep up on all that. But anyway, so you can start right away, even just if you're doing it with open access, not really open source, but it's open access over, um, through that customizable option. Yeah, you can start right away.

Yes, I love it. And I agree, I, uh, I think it's always— I think what'll end up happening for those who don't know how to code is you will learn how to code over time. Like, it's as you look at things. Oh, like me, when I first started working on Alexa, I didn't know Node.js, but I saw, I saw it consistently. I saw it being used, I stepped through the code, and I learned it. I learned enough of it in order to understand how it works. And I kind of think of it like HTML today. Um, so one of the things I also encourage as you move down this path is to look at some of these no-code platforms. So thank you, Daisy, great to see you. Um, so on the low— on low-code, no-code platforms, there's AutoML. So AutoML is for junior data scientists or senior data scientists that want to expedite some of the rudimentary tasks in building models. But it really was originally built to help equalize earlier stage, uh, data scientists to build models. So it does a lot of the work for you, but it actually does it in an explainable way so you can walk through it. There's lots of great tutorials for that. The other thing I'd also encourage you to look at is Power Platform. On Microsoft. Microsoft has Power Platform. They have this— Power Platform allows you to assemble in WYSIWYG style, if you remember that term, right? What you see is what you get. It's a graphical user interface and you can drag and drop components into a flow and that flow will allow you to build a, you know, the backend might be a GPT or an OpenAI service that you're talking to. It might be another API. You might integrate it with other services. But you're integrating them as boxes that connect together with inputs and outputs. So you still have to understand how it works. I always tell people like AutoML or, you know, automated machine learning isn't really— no, I mean, it does allow you to build machine learning without any code. It doesn't mean you can't get away without knowing what machine learning is though. Like you still need to understand all the knobs and whistles and bells that you can turn and twist in order to make it work for you. So, but there's plenty of opportunities. Like you said, um, Aaron, I loved what you said about like learning from others like ourselves, like people on, you know, people from Harvard or MIT who are now teaching for free on YouTube. Like, yeah, learning from them, they'll tell you, don't worry about learning Node.js, use it in the context of this application. Or learn it for this reason. I actually helped build the AI program at Miami Dade College, and it's a community college, one of the most diverse colleges in the world. And I did it because I wanted to create accessibility. And when I was in high school, I wasn't going to be university-bound. I ended up going to university, but I didn't see an opportunity. I didn't think I was good at math. I didn't think I was good at coding. I ended up— I'm actually great at both of those things. But at the time, when I was a 17-year-old, to 20-year-old, it didn't occur to me that that was a path. And when I looked at the math that was required to do the things I was excited about, like physics, I was like, there's no way I can do that. And I opted myself out. So when I built this program at Miami-Dade, what we ended up doing is we ended up changing all of these like hardcore math classes into what do you actually need to do the work. And in the world of applied AI, AI that you're going to use and apply to business problems. Most of these models are already built. You're going to extend them. So that goes to what Aaron was saying about just learn enough code to piece things together. You don't have to be a hardcore programmer. You don't have to go to 4 years of university to learn how to code. I went to the University of Barnes Noble, so you could do the same. Or you can go to GitHub, um, or you can go to one of my favorite online resources, which is freecodecamp.org. And they provide daily practice building projects very similar to, to some of the things we've talked about. So, but get started, but it's, I think it's very accessible now more than ever to help you do that. So thank you so much for that question. We have just a couple minutes left. I'm gonna see if we have, oh, we have just a lot of love for you. So that's awesome. Yes, let's see here. I'm just checking to make sure we didn't miss any questions. So maybe as we close out, Erin, any advice or— I mean, you've given tons of advice for us today, but any final words you'd like to leave us with as we close out the day today?

Yeah, I have. You know, there was once a time in my life where I— well, it was my first time contracting. At Facebook when I was at the Oculus site and read Oculus site for research and development. So I was walking by like the most incredibly intelligent people I could have ever imagined. I just, I never had the confidence to really have conversations with them. I didn't I didn't feel worthy of, of taking up space in their time, in their day, even though they would approach me. And, you know, they're the most humble people that you would ever meet. But I just didn't have the confidence. And I have people on my team who resonate with the mission, who are very educated, talking PhD, MBA, ML, AI graduates, from top universities like Carnegie Mellon, like really great, super intelligent people. And I got over that little voice inside of me that said, oh, you don't have a PhD, you can't talk to other PhDs, you know, like that. So I just want people to understand, like, even if you don't know something, you don't need to know everything. But you can connect with people who specialize in that and know more than you do. And just trust them to do their part if you've connected with them to build. For example, you know, like, I can't validate Black history because I was raised on a very whitewashed version of history, and I'm not going to be able to provide that validation in my app. That's why I'm working with historians who have been teaching Black history for 10 years, who have doctorate degrees, right? I can't specialize in all of those things. I can't go back in time and do 8 years of school before I try to do something good in the world. So don't think that you're behind just because you don't have that. I have like so many programs where you can take all the classes for free, and you can get your marks and get your grades, but then you have to pay to get the certification. I'm not paying that. I'm not paying you $30,000 for what I already learned just so you can like what, you know, if somebody asked me a question, I can answer it. And that's what matters, right? So I understand there's a lot of value in getting that piece of paper. There's also a lot of value in learning and representing what your knowledge is, and then partnering with people who specialize deeply in those fields to make up for areas you don't necessarily have that super in-depth knowledge. So for me, I know that my historian isn't going to want to do business stuff. He doesn't want to set up, you know, internal architecture of like how the business runs or how we're going to apply for grants and things like that. He does Black History, you know what I mean? So it's okay to partner with people who have done incredible things because you're also incredible. So I think that that's a really important message. Just get out there and connect And don't be afraid of people who specialize in things.

Oh my gosh, I love it. And it's so important because I think a lot of us, we do get into these spaces where we don't feel like maybe, you know, we're a little skittish about entering a room or having a conversation. But I will say that so many times, just like you have experienced, like, it's definitely not them. It's always us. Like, we're always the one saying, oh, I'm not really sure. And they are the nicest, most welcoming. Maybe on occasion, like 1 out of 99 would be not as welcoming, but, and usually then it's because they're not super social. Like, they just don't, like, it's not, um, it's usually never a situation where they wouldn't be happy to talk to me or have a conversation with me. So I think it's really important, and they're actually quite happy to have that interaction. Um, and so Reach out to people, connect with people on LinkedIn, have conversations, show up at these listening sessions, and be willing to not just listen, but interact, create real relationships. Because in the world where AI is going to make our lives, I think we're going to have extreme productivity advancements, but that should open up some room for us to have much better human advancements. So, so excited you are here with us today. I want to echo what Joe said as well. You are enthusiastic and authentic and just my kind of people. So thank you for spending time with us. Good luck in all your endeavors. Everyone give it up in the chat for Erin. I'd love to— for her to later go back and see all the love. Um, thank you for coming. Happy Valentine's Day, everyone. I am actually about to hop on over to another event with T. Um, she is hosting, uh, We Love Generative AI. It's like a, I think, a panel. It's been going on already, but I'm going to be joining for the last hour. So if you want to stay online, see me over there. Otherwise, Erin, thank you. It was awesome. I'm so grateful. Yes, have a great day. Happy Valentine's Day, everyone.

Bye.

Bye, everyone. Yeah, thanks everybody. Have a great day.