--:-- Mel Robbins: Hey, it's your friend Mel, and welcome to the Mel Robbins Podcast. What if I told you there's a tool that can save you time, help you make a lot more money, get more done, and improve your life? And get this, it's free, it's right in front of you, available to you right now, and you're either not using it or you're not using it correctly.
--:-- Mel Robbins: The tool I'm talking about is AI, artificial intelligence. Haven't you noticed? It seems like everyone everywhere is talking about AI. But see, that doesn't mean people know how to use it. 99% of beginners don't even know the basics of AI. Here's a guarantee. You are not tapping into the AI tools, the tricks, the skills that you need to know right now, and you are leaving so many opportunities on the table.
--:-- Mel Robbins: If you think you can avoid AI, you're wrong. You're already living with it every single day, whether you notice it or not. And so I want you to be the one in charge of how you use it.
--:-- Mel Robbins: My guest today, Allie K. Miller, is someone I'm calling the AI whisperer because she is in the ear advising some of the world's leading brands on AI, and she is the most followed AI expert on all of LinkedIn. She has impressive experience at IBM and Amazon on the AI teams, and she is here making her podcast debut.
--:-- Mel Robbins: Now, I've been searching for someone to come on and talk to me and you, not to scare us, but to empower you and me step by step on how to unlock the transformative power of this remarkable tool. Allie is going to give you the AI basics that you need, and she's also going to explain it so simply in everyday examples so that you and I can leave this episode and know exactly what to do to unlock the four different levels of AI.
--:-- Mel Robbins: You can use it to solve any problem that you have, to make anything that you're doing even better, to help you earn more money, save time, find a great job. The ideas and ways you can use it are endless. Allie K. Miller says, AI isn't something to fear, it is something to embrace. And I want you to know something. I'm going to be sitting here with you shoulder to shoulder because I know I need to embrace this tool too.
--:-- Mel Robbins: It is time for a real human conversation about AI. One that's going to give you the truth, the confidence, and the step-by-step simple moves that will help you take control of your time, your money, and your life.
--:-- [Theme music fades in and out]:
--:-- Mel Robbins: Hey, it's your friend Mel, and welcome to the Mel Robbins podcast. I am so excited that you're here. It's always an honor to be together and to get to spend this time with you. And I also love it when we're learning something that's going to help us.
--:-- Mel Robbins: And if you're a new listener and you're here because somebody shared this episode with you, well, I just want to take a moment and personally welcome you to the Mel Robbins podcast family. I am especially glad that you chose this episode because today you and I are going to learn how to use AI to save time, to make money, to improve your life from the person that I've just dubbed the AI whisperer because she is in the ear of some of the top brands and CEOs on the planet. Her name, Allie K. Miller.
--:-- Mel Robbins: In 2025, Time Magazine named her one of the 100 most influential people in AI, and it's easy to see why. She is also the number one most followed voice in AI in business. She's a graduate of my alma mater, Dartmouth College, and she's also a graduate of the Wharton School of Business. She launched the first multimodal AI team at IBM, which if you're like me and you had to look that up, it means she pioneered AI that understands text, images, and other data together, not just one at a time.
--:-- Mel Robbins: Then, as global head of AI for startups and venture capital at Amazon Web Services, she advised the biggest founders, investors, and machine learning researchers in the world. Today, her insights are trusted by Fortune 500 companies, leading startups, governments. Today, she has flown here to our Boston studios for one reason, she did it for you and me. She's here to break all of this down in terms that regular people like us can understand. And she's also here to empower you to not just lean in and learn, but to actually leverage the full power of the AI revolution.
--:-- Mel Robbins: So please help me welcome Allie K. Miller to the Mel Robbins Podcast. Allie Miller, welcome to the Mel Robbins Podcast.
--:-- Allie K. Miller: Thank you for having me.
--:-- Mel Robbins: I am really excited to talk to you because I know this is going to be a conversation where I selfishly am going to learn so much. This is a topic I've been dying to have an expert on. I am so glad we could pull you off all the stages where you're speaking around the world and have you here in our Boston studios. I would love to start by having you tell me, how is my life going to be different? If I take to heart everything that you're going to teach me today about AI and I put it to use in my day-to-day life.
--:-- Allie K. Miller: If you take everything that I'm about to share to heart, you are going to learn how to use AI, which is the most basic value that I could deliver to you. You are going to save time. You are going to get more support that you need in your life, in your work. You are going to expand your capabilities and your superpowers. And you are going to be shocked by what you can actually get done with these systems.
--:-- Mel Robbins: I love that because you talk a lot about the fact that you can use AI and the thing you're most excited about is that it can help you become the best version of yourself.
--:-- Allie K. Miller: Yes.
--:-- Mel Robbins: You actually believe that?
--:-- Allie K. Miller: 100% because I've seen it in my own life. I teach millions of people how to use AI. I've seen it in their lives too. Whether you are a 91-year-old grandmother, whether you are just out of college and you're freaking out, not even knowing what to do, I've seen the change happen. So I am sharing with you every single thing that I've shared for the last 10 years online, hopefully as fast as humanly possible, and I can't wait to share it.
--:-- Allie K. Miller: In addition to saving you time, because I always want to do that, right? Efficiency is key. There are some transformational ways that you can use AI to improve your life, to lead the life that you want to be leading, whether that is getting research done on a topic you've always been interested in, whether that is developing a workout plan in the way that you've always wanted to do it, whether that is having a better relationship with your kid. I want you to live the life you want to be living and not be held back by the environment or context around you.
--:-- Mel Robbins: Let's start at the beginning.
--:-- Allie K. Miller: Great.
--:-- Mel Robbins: What is AI?
--:-- Allie K. Miller: If I could give the most simple explanation.
--:-- Mel Robbins: Dear God, please. Because it's overwhelming, Allie. Like I wanted you here because every time I turn on the TV, I don't even watch the TV. Every time I log on, especially YouTube, it's coming and the robots are going to kill you and steal your jobs and we're doomed. And I'm like, well hold on a minute, I'm not even sure I understand what it is and how to use it. So let's start at the baseline thing. What is it and how does it work? Can you just explain it for those of us that kind of think we know?
--:-- Allie K. Miller: Yes. So AI as an umbrella term has been around for decades. The term AI was invented in the 50s. So AI at its core is just a system attempting to do a human-like thing.
--:-- Allie K. Miller: That could be as crazy as self-driving cars. That could be your Roomba in your house. That could be your spam filter in Gmail. All of that counts as AI.
--:-- Mel Robbins: Don't come after my vacuum. That's all I've got to say. I don't even have one of those.
--:-- Allie K. Miller: Your vacuum might come after you.
--:-- Mel Robbins: We don't have that. Okay. So you could think about AI as any computer program that is attempting to do what a human being typically does.
--:-- Allie K. Miller: A system, yeah, attempting to do a human thing. Whether it does it in the method that a human does it, kind of open for a question.
--:-- Allie K. Miller: But it's whether the end user, upon seeing the final writing, the final tweet, the final image, the final video, goes, yeah, okay, a human could have done that. Generative AI is a subset of that, which has also been around for decades. The new thing now is good generative AI, high-quality generative AI, generative AI that changes the way that we might check emails, write emails, or completely build our business, right?
--:-- Allie K. Miller: Generative AI is an AI system that is looking at big amounts of patterns. Picture the whole internet. Picture all of Wikipedia and a whole bunch of the internet thrown into this model. And the model picks up on patterns. And so it's looking at patterns that we as humans are probably going to miss out. It might be picking up on, every time you say the word zebra, the word black and white tends to be around that word. And of course tends to be around that word, but penguin is nowhere near it, right? So it's picking up on a bunch of these patterns, and then it is using all of that pattern recognition to very awesomely generate net new things. So it's not copying and pasting. It's generating brand new stuff, brand new images, tweets, emails, novels, movies, blog posts, whichever. So generative AI, subset of AI, where it makes new stuff.
--:-- Mel Robbins: So what about a plane flight? What about a plane flight?
--:-- Allie K. Miller: So in the old days, like three years ago.
--:-- Allie K. Miller: Well, in the old, old days, you would call a travel agent, right? Then in the next iteration of that, you would go to the airline website.
--:-- Allie K. Miller: Then in the next iteration, which is where I am.
--:-- Mel Robbins: So I'm stuck in the kind of modern old way of doing it. I go to Google. I go to the Google and I put in my flights on a date and then I get a list. But when I get the list, I have to look and pick out the flight.
--:-- Allie K. Miller: You're not going to do that anymore.
--:-- Mel Robbins: Okay. What am I going to do? And can you explain how AI is making my life easier?
--:-- Allie K. Miller: Yes. Go. Okay. Let's say that the reason you're looking up a flight, let's say that you're planning a family vacation or something.
--:-- Allie K. Miller: Okay. Instead of what we used to do for the last 20 years when we've Googled these things, you would Google flights from Boston to Atlanta. Right. You would then have to filter nonstop, one-stop, different airlines because you got points there, points there, whatever. And you would still have to be that big filter mechanism. New age, you're going to go into one of these systems. It's going to have access to the internet. And instead of going, pull me all the flights from Boston to Atlanta, you're actually going to say, my family of five wants to go somewhere warm in September. We are thinking about Charleston, Savannah, but we want to try something new. We're thinking three days, maybe it's five. We've already gone to Scottsdale. Here's three reasons why we liked it. Here's two reasons we like Texas. Here's four reasons we're thinking about Bermuda. My son's allergic to strawberries. My daughter really wants to stay hydrated. My other daughter wants to do yoga, whatever.
--:-- Allie K. Miller: You're going to be able to feed in that amount of context. And before you even decide that you wanted to pick Atlanta, it's going to act as that co-pilot because you're bringing in all that context. So in addition to it finding your flights, which again, you could totally Google that. I still Google things, to be clear. In addition to finding those flights, it is going to help you create an entire action plan around this vacation.
--:-- Mel Robbins: So it's basically going to make a recommendation based on all the things you told it and crunching all the data and what other people have searched for and given a thumbs up and thumbs down?
--:-- Allie K. Miller: Based on not even things that people have searched for, based on just any time people have written about things that are kind of similar to Atlanta or Savannah or someplace that is totally not similar, right? A sauna. It might be some sort of relation where people go, okay, in warm places, maybe you do this hobby instead of that hobby.
--:-- Mel Robbins: That would be so helpful.
--:-- Allie K. Miller: It's incredibly helpful. And what I think a lot of people miss is that these systems can add so much more action into your life.
--:-- Mel Robbins: It also immediately made me feel, as a mom, because you're managing so many different variables on anything that you're actually searching for, that being able to turn all of those concerns and variables and but this, but that, but this versus what are the right flights to get us to the same airport or train station at roughly the same time, I felt like a giant exhale.
--:-- Mel Robbins: And one of the reasons why I was so excited to talk to you is it's already here. We use it in just about every aspect of the way that we work at the Mel Robbins podcast and 143 Studios. And I have not started using it in my day-to-day life.
--:-- Allie K. Miller: Okay.
--:-- Mel Robbins: And it is kind of everywhere. And I noticed that, and maybe you notice as you're listening to Allie or watching this right now, that your phone needs more updates than ever because every app is having an update because it's got AI in it. And so it's already here. And so I was excited to talk to you because it has exploded. It has accelerated. And I don't want to get left behind. I don't want women in particular to lean back and get left behind.
--:-- Mel Robbins: And I'd love to hear what is the takeaway for someone listening to this conversation about the opportunity that is available to you if you lean into utilizing the power of this in your day-to-day life?
--:-- Allie K. Miller: I like to think of it in a couple different categories. The obvious one that I think people read a lot about and pick up a lot more quickly is the productivity side. Doing things that you're already doing today faster.
--:-- Mel Robbins: Give me a couple quick examples.
--:-- Allie K. Miller: Writing emails faster, writing blog posts faster, taking your blog post, creating a video out of it faster, right? The idea that just speed and the idea that we can synthesize an article, right? All of that is things that you would already be doing. You'd already read the article, but it's able to do it a lot more quickly.
--:-- Allie K. Miller: Or, at a bigger scale as well, right? I can synthesize 10,000 pages in a paragraph in like a minute. That is the category of doing things faster.
--:-- Allie K. Miller: Second category is doing things better. And this is what everyone is missing out on, which is, yes, I could use it to cheat on my college essay.
--:-- Mel Robbins: You're not recommending that.
--:-- Allie K. Miller: I am never going to recommend that. But to think that you should not use AI in that process might also be wrong. So anytime that I'm coming up with a big plan, right? Let's say that I'm coming up with a plan for how I want to show up to this podcast. I might ask AI to interview me and say, go full Mel Robbins on me and just say, hey, ask me 20 questions to get out more information that I can work from. I might say, here's my plan. What are five risks that I might not be thinking about? And what are ways to mitigate those risks? I might say, what are 10 crazy ways to make this more interesting? And maybe it tells me to bring a yellow pen because it's all male branded. And obviously that's a small example, but your work can be so much better.
--:-- Allie K. Miller: And I think so many people fall into this productivity trap. Whereas, my team, we put on this AI First Conference. We ran our entire agenda through AI and said, think from the viewpoint of these five different people, right? We had these synthetic personas. Think from the viewpoint of the busy CEO, the really busy parent who has too much stuff going on with their work and their kids and their parents. Think from the viewpoint of these three other people and now review my entire conference plan and give me 10 ways to improve it, 20 new ideas we haven't thought of, 30 ways that it could go wrong, 40 ways to make it a better bonding experience for the team. And so making it better, not just faster, so that you also don't feel like a drone trapped behind your computer with all these thousand pop-ups.
--:-- Allie K. Miller: And the third, which is even harder to figure out, is doing new stuff. Right. So doing things that you're already doing today but faster, things you're already doing today but better, and then net new. Holy cow, can you even believe I did that? And I'll give you one example here just because it's personal life. It's not life changing right when you hear it, but then you hear a little bit of value around it and you go, wow, they really did that.
--:-- Allie K. Miller: A woman that I talked to joined a Mahjong club in New York, obviously. Shout out to my mother. She taught me Mahjong.
--:-- Mel Robbins: Really? I love the clicking noise and I love playing it.
--:-- Allie K. Miller: Okay. So heck yeah. You and your mother and I are going to be best friends.
--:-- Mel Robbins: [Laughs]
--:-- Allie K. Miller: So this woman wanted to become a better Mahjong player because it allowed her to bond better with these people in this club. She used AI as a non-coder to create an entire app to teach her how to play Mahjong, to drill her on the tiles, to drill her on the combinations so that she wasn't using her time buried in this little notebook of the different rules so she could spend that time actually hanging with these people and creating lasting friendships.
--:-- Allie K. Miller: So again, the AI component of that is not the coolest part of that story. The AI component generating her own app, building it in a couple days, and now she uses it literally on the subway to train. It allowed her to create more value in her life that could not exist as a non-coder before. These systems are so accessible. We have never had tech be as accessible as it is today. We've never had the ability for non-coders to jump in.
--:-- Allie K. Miller: And women are adopting AI 25% less than men. And I just think about what societal opportunities we're missing out on, what economic opportunities we're missing out on. It is such a big jump that people feel that they have to take, and actually it's really just about opening up this thing, testing out a few prompts and just getting your feet wet. Your gears are going to start turning, right? Your listeners are brilliant people. I read your comments. They are brilliant, brilliant folks. It is all about giving yourself the best chance of being able to capitalize on these tools and build the value that you want in your life.
--:-- Mel Robbins: What do you say to the person who's nodding along and is like, that sounds really cool. Never thought about how I could use it to be faster, better, or do things I never even imagined. But they're not sure, and they're kind of waiting for the right moment to jump in and learn AI. Allie, what do you want to say if that's you?
--:-- Allie K. Miller: 5, 4, 3, 2, 1, right? I want to dispel people of the myth that there is perfection in our lives, period, right? In our financial decisions, in the way that we decide to make dinner that night or hang out with our kids that day, we're waiting for something that doesn't exist in our lives. And so at least when I look at people that I look up to that are successful, it's people who jumped in and did the thing.
--:-- Allie K. Miller: And I would also say that it's not that big of a leap. The people who are winning in AI are not these big, crazy, risky decision makers. It's people who are taking these quick little wins and quickly iterating and creating a little system of adaptability. It's people who actually think a little bit smaller and get their feet wet.
--:-- Mel Robbins: Here's how I think about it. I think about it like having a personal executive assistant, like all of the things you wish somebody else could handle. Coming up with the perfect workout routine if you want to have more defined calves. Learning a better walking loop in your neighborhood when you only have 20 minutes. I just feel like there's so many ways you could use it that I personally have only just scratched the surface.
--:-- Allie K. Miller: There's like four interaction modes that I think about. And again, most people are stuck at step one. And so for the person listening, please do something to try one of these three others.
--:-- Allie K. Miller: Okay, number one, microtasker. That's like the make a meal plan for these 20 people that are going out to dinner. Two people are gluten-free. One person only likes ham, whatever. And you're going to be able to very quickly do that. That might also be the flight search example.
--:-- Allie K. Miller: Number two is as a real-time companion.
--:-- Mel Robbins: Okay, what does that mean?
--:-- Allie K. Miller: You can just pull up these systems and be in a live video chat. And so as an example...
--:-- Mel Robbins: Why would you want to do that?
--:-- Allie K. Miller: I went to a board game bar and my friend and I had 45 minutes and I could have spent 20 of those 45 minutes evaluating every single game that existed. And we'd only have 25 minutes to play. Instead, I opened up video mode and I am just scanning through and I go, we have two people, 45 minutes. I want an easy game. I want to have fun. Tell me which one.
--:-- Mel Robbins: What you're saying is that you can open up video mode, scan an environment, and it's almost like having a guide and a decision maker to help you assess what's happening. Could you do that if you're lost somewhere?
--:-- Allie K. Miller: It's scary good at picking up on locations, especially if you're in something recognizable. If you're in the middle of nowhere and you're using Google because it has Google Maps tapped in, it might be pretty good. In general, I wouldn't trust it for being honest. I would just open up Waze or Google Maps or something.
--:-- Mel Robbins: Gotcha. I understand. I think a lot of us have discovered the ability to take a photo and then search what's in the photo by putting it in a search engine. You're basically saying there's a second step where you can use the video scanning or open up the video. I didn't even know this existed. So already I'm like, you can do that?
--:-- Allie K. Miller: My willingness to get something done goes crazy high when I know that I'm not alone in that task, whatever it is. So I have been in a live video stream with AI where I'm screen sharing what is on my screen, and I am navigating Etsy to pick out the perfect gift, right? And I'm just having a chat back and forth. But it's like being in a Google Meet with an AI that can see everything that you see.
--:-- Mel Robbins: So it's literally the same as you typing in the meal plan that you want, but instead this is like open up your fridge and scan it with the video AI mode and go tell me what I can make with this? I have needed this 20 years ago. We've got to know. How do I turn this on? Like this is the level of which I'm at.
--:-- Allie K. Miller: So I'll give you one example. I am a terrible cook. Everyone that knows me knows this. But my sister told me that cooking is just chopping things up and heating things up and so I'm trying to get better at this but the recipes part eludes me. And I take a photo of my fridge, a photo of my pantry, and I hit enter and it tells me exactly what I can make. It gives me a couple recipes that have ingredients that are missing and it tells me the exact grocery shopping list that if I go to Trader Joe's I can grab.
--:-- Mel Robbins: So that is something that has saved me literal hours and money and food that didn't go to waste.
--:-- Allie K. Miller: 100%. I'm no longer that person. And emotions that I feel because I feel like a bad person for wasting the food. And then I feel like an idiot for not being organized enough. I live in New York, so this is a lot easier for me to do. I'll walk down the street and have an entire conversation with an AI system. I will talk through a problem just to be devil's advocate. Am I the asshole? Right. And I will just talk through this idea as if I'm on the phone call with someone.
--:-- Allie K. Miller: I can do this at two o'clock in the morning.
--:-- Mel Robbins: What's the third interaction type? So if you've got the prompting, now we have the sort of live video voice thing, real time, acting like an assistant, helping you out here.
--:-- Allie K. Miller: Okay.
--:-- Mel Robbins: What's the third?
--:-- Allie K. Miller: Two others. You've got delegate, which is really happening right now where you can give AI a 20 minute task and it'll come back to you with an answer. So you might say, let's say that you're a teacher and you really want to come up with a new lesson plan for chemistry. You can say to any of these agent tools, I'm a teacher. I want to be able to pull off a new thing in chemistry. I can upload screenshots upon screenshots of all of the years of reviews that I've gotten from students. I can feed it all into the system and I can say, I want to come up with a new chemistry plan. Go online, find me a hundred other examples and create an entire spreadsheet for me, an entire document summarizing this and an entire pamphlet that designs the PowerPoint around it so that I come back to it 20 minutes later and I already have this fully done report for me.
--:-- Allie K. Miller: So I am constantly delegating big, big planning tasks, particularly things that are rooted in research or data entry, because that's where AI is still really good.
--:-- Mel Robbins: You know what's super exciting about that? For anybody that has typically a business or anything that's kind of just you, you're a realtor or you're a teacher or you're a nurse or whatever it may be. And you're like, who do I give this to? I need a website. I don't have anybody to delegate it to. I don't have the money to do that kind of thing. You're telling us that there are tools available now for free that are your team that you can learn how to use pretty quickly that can do all of this work that for years you've had nobody there to do. Whether it's social media stuff, whether it's a business plan, whether it's a website, whether it's an app, whether it's a marketing plan, whether it's analyzing what the realtor competitor that you hate is doing that you want to do? Like all of it. It's like a free research assistant.
--:-- Allie K. Miller: And I think solopreneurs used to feel like they were deserted on an island and that no one understood them. They had no help, right? AI gives you 20% of a marketing person, 20% of a customer support person. We're still going to use amazing video editors for stuff like this podcast, right? But the average person is now able to record themselves for an hour, upload this video to a tool and immediately get 15 clips that they can post. And those 15 clips come already pre-cut, already captioned. I know people that are taking their Instagram videos and immediately turning them into Spanish and posting those on a second channel. The reach that you can have, the impact that you can have has increased by 20%, 10x, whatever. And so many people aren't taking advantage of it because I think deep down they feel like it's wrong or it's cheating or something like that. But people that take advantage of it now are going to gain this velocity that is going to be really hard to catch up on in the next two years.
--:-- Mel Robbins: What is the fourth interaction type?
--:-- Allie K. Miller: Teammate, which for folks that might be at bigger companies, for folks that are maybe they're part of the marketing department or something. Think about, yes, you said they all had their executive assistant. What if your entire team just got a little helper? So as an example, maybe you record all of your meetings with one of these tools. Suddenly you can have an AI system that is sending out automated reports every single Friday morning to your entire team going, what did we not do today? What's the status of this project? What's the latest? Because it's able to grab from documents. It's able to read Google Drive. It's able to look at Gmail. And so it is lifting up the tide of your entire team.
--:-- Mel Robbins: Allie, I have so many questions. I'm so glad you're here. I just want to take a pause so we can give our sponsors a chance to share a few words. And I want you to also have a chance to share this with people in your life that you know need to learn this. These are skills that are going to save you time, make you money, improve your life. This is a tool that can help you solve problems. Every one of us needs to lean in and learn. This episode is a free, life-changing resource. And so I want to thank you for being generous, sharing this with your parents, with your grandparents, with your brothers, your sisters, your kids. I'm sharing this with everybody I'm related to and all of my close friends. But don't go anywhere because Allie and I are going to be waiting for you after this short break. So stay with us.
--:-- Mel Robbins: Welcome back, it's your friend Mel Robbins. Today you and I are getting to learn all about AI from the woman that Time Magazine called one of the most influential voices in AI today, Allie K. Miller. So Allie, let's just jump right back in. My next question is this: what is the biggest mistake you tend to make when you start using AI and how do you fix this mistake?
--:-- Allie K. Miller: I'd say the average person is not bringing in enough context. They're coming into these systems and they're going, plan me a family vacation. Who's your family? What vacations have you taken before? Or you're coming in and let's say you're building a house or you just bought a new apartment. And you come in and you just say, help me fix my apartment. Help me be more organized in my apartment.
--:-- Mel Robbins: So how would you do that with the apartment? Because my daughter just moved into a new apartment.
--:-- Allie K. Miller: Fantastic.
--:-- Mel Robbins: Every day I'm getting a call. Overwhelmed because you forget that when you move into an apartment, there's not a spoon. There's not a hanger. There's not a waste paper basket. And then it's overwhelming. And so how would you use it? Because you said the biggest mistake is context. So I get it with a vacation. You should be like, my kids are these ages. This is many days. These are the dates. This is what we like to do. So the more context, the more it could help. How do you do the apartment?
--:-- Mel Robbins: So apartment, let's say that you give a photo of your apartment, the square footage, photos of your previous apartment, concerns that you had about your previous apartment. "I didn't have enough storage. I didn't have enough place for my board games. I didn't get enough natural light at my desk." Right? So you can share all that. You can say, "And I'm also worried that someone's going to walk in and see my bed unmade." Right? Boom. Think about how AI might solve that. And I'm concerned that people are going to think I don't have enough furniture, or that I have too much furniture. Or how do I make it look like this is full when I only have the money for second-hand? Or could you actually say, "Find me a couch"? Yes. "I have this much money. Scan online that has delivery." One hundred percent.
--:-- Allie K. Miller: Oh my God. I went to an AI system and I said, "I want to find a watch. I want to find a watch that's less than $50, AI-themed. I only want it to be square or circle. I don't like anything that's rectangle. I only like black and gold. I need it to be whatever." I give it 15 parameters and I said, "Go." It is working. And by the way, I can see it working the whole time.
--:-- Mel Robbins: Why? Because the wheel is spinning?
--:-- Allie K. Miller: No, because I'm literally watching it navigate websites. Imagine that it goes into another room and it opens up the laptop and it just works on its own. You're watching it, right? Just like an IT person would tap into your computer. You're literally watching it navigate and scroll and click.
--:-- Mel Robbins: Wait, so is it controlling your computer?
--:-- Allie K. Miller: It is controlling a virtual computer that you're just watching like an observer.
--:-- Mel Robbins: Are there risks to using agent mode? I mean, this sounds amazing, but I'm just...
--:-- Allie K. Miller: There are risks of everything. The main risk to think about would be, let's say that you're saying, "I want to buy a couch." And at some point you're going to go add it to my Wayfair cart and I want to check out. Right? Or I'm buying this thing from Ashley. How do I give it my credit card? Right? You're going to take over that screen. Got it. It's not looking in that moment. You type in your credit card details and you say, "OK, I'm done." And the AI model goes, "Sure." And you go, "Yeah, I'm done." And then you come back in and you keep going. So anytime you're logging in, anytime you're giving financial details, that's going to be an extra layer of concern. But these systems are not tracking that remote control.
--:-- Mel Robbins: That is so cool. I don't want to talk to you anymore. I want to go try this. We can spin up the laptop right here. I love this stuff. It's pretty incredible. As somebody who's advised top companies and even governments on how to use AI, what is one simple trick that everybody misses that would instantly save time for you if you try this and it would save time every day?
--:-- Allie K. Miller: Can I give you two? One that is very easy, one that's weird.
--:-- Mel Robbins: Give me five. Yes, go for it.
--:-- Allie K. Miller: Okay. The one that is very easy that everyone can start with is having AI interview you, right? Coming to it with a problem and just saying, "I need to redesign my apartment. I need to come up with a plan on how to keep my mother entertained when she visits." Right? You come to it with a problem and then you, instead of coming with this whole long prompt, you can say, "I don't really know how to solve this problem. Help me help you. Ask me five, ten, twenty questions." And then you're going to turn on dictation and you're just going to talk and complain and ramble. And you're going to say, "I'm thinking about this. I'm worried about this. I tried this. This didn't work. Here are five things that I know I'm good at, five things I think I'm bad at, three ways that my boss is yelling at me, two people that I want to hire." Whatever sort of context you need to bring in. Ramble, ramble, ramble, enter. I do this when I'm at the hair salon, right? I just have it ask me questions. And while I'm sitting there with the dye on my head, I just whisper to it for 20 minutes. And I'm able to get four hours of work done in 20 minutes. That is a crazy easy one, right? Go Mel Robbins mode. Go Barbara Walters mode. Ask for AI to interview you.
--:-- The second way, weirder one, that is my favorite hack is synthetic personas. It is creating fake profiles of people that you might be pitching to, people that you might be working with. I know a woman who created—this is insane, by the way, but I love it—I know a woman who created a synthetic persona of her husband because she wanted to have the family go to Disney World. And the last time she asked, he said no. So she practiced against the synthetic persona and then was able to come to it with, "Hey, here's the perfect pitch." And the husband goes, "Oh my God, sounds like a great idea."
--:-- Mel Robbins: Well, I recently, I think, did this. I went to one of the models. I went to Microsoft Copilot and put in—okay, I'm not going to tell the long story—but I basically was trying to write a strategic apology to somebody with a very challenging personality in order to defuse a situation. And I went to it, explained the situation. It spit out the card. It worked like a charm because it acted like the person that I described. And then I told it the situation and said, "What could I write in a card that I'm going to mail to somebody that would defuse a situation?" I know. It was genius. This is brilliant. And I want people, whether you are extremely extroverted and you feel like you're pitching all the time, right, you can still be better.
--:-- I know of a gentleman that told me he was terrified to quit his job. So terrified that he almost didn't quit his job because the fear and the shame of that moment of "maybe I won't do it right" was halting his progress and ability to go to the dream job that he actually wanted. And it was so paralyzing that it was going to change the trajectory of his life if he didn't do this. And he went to AI and he goes, "How am I supposed to do this? How does someone do this? I've never had to quit a job, right? How do I deal with a boss that's going to push back? What if they offer me more money? How do I say no to that? How should I write this letter? In what order should I send this letter?" And that sort of life change that someone can experience—you got to get through that in a much lower stress way. You got to get back to the thing you actually wanted to do.
--:-- Allie K. Miller: Well, you know, what's interesting about it is, first of all, you keep reminding us the amount of context you give it is critical and is directly related to the value of the information you're getting back. The second thing is, what you're doing when you take the time to think through scenarios and you take the time to get really present to either the thing that you're worried about or the thing you really want to achieve, and then you utilize a tool like this to make yourself smarter and more effective, is you're just using all of the foundational psychological principles called if-then planning. You are using all of the things that human beings have done forever. You're just utilizing a data set to help you do it faster and better. And then that makes you more confident and more equipped to go into your real life and follow the advice that feels right for you. It's like practicing.
--:-- It is a brilliant view into this space because so many people look at it as faster Google when it's actually a prosthesis for reinvention. There is so much you can do with it that just searching faster almost feels limiting. I have this post-it on my desk that says, "Use AI to become the person you want to be." And it helps me get out of that productivity trap where again I'm just using it to write emails faster or I'm just using it to find information. It reminds me that the real challenge of these systems is, wait a second, how can I take all the excuses that I've had over the last—I'm not going to say how old I am—but how can I take all the excuses and get rid of those excuses, right? If I had had this when I was starting my business and I could go to it and say, "How do I start an LLC? What are the big concerns when I'm picking a lawyer? How do I pick a good accountant? What are 20 questions to ask my first hire?" I would have been in such a better spot. And so again, it's using AI to become that person you want to be. Not over-relying on it, not misusing and abusing it, not lazily offloading to it, but using it as that method for reinvention, tool for reinvention.
--:-- Mel Robbins: What if I work for an employer who isn't using AI?
--:-- Allie K. Miller: Prepare to quit. Like, we are three years into the AI revolution. And if your employer is actively banning this technology and in three years has not yet carved out a safe, responsible AI policy that allows them to use it in the work, you are at a massive disadvantage for your work, your life, your career. You're going to be less hireable in your next couple roles. You know, maybe if you're in manufacturing or plumbing or HVAC or something, it's fine. I'm talking about the knowledge workers who could be leveraging this. Your company, whether they're doing it intentionally or not, they are putting you at a massive disadvantage for the next several years of your career. For that person, I would say, learn AI, raise your hand, try and have AI be at that company and say, "Can I lead it? Can I take on the first project?" Right? That's an opportunity to be a big leader in your org. If you are met with "no" and they say, "I don't want to use the tool, I don't trust it," you can take on that project, leave your company. And I know that that sounds like a privileged statement and it is to a certain degree. You need to make a plan to do that, even if it means leaving and working for yourself and being a coach that uses AI, that is able to be more efficient and is able to have more clients that they can help. But we're three years into this. A year into it, I wouldn't have said that. Three years into it, I'm saying it.
--:-- Mel Robbins: Okay, we hear the call. We need to leave. Now I'm looking for a job. What is the best way to use AI to help me find a job that I love?
--:-- Allie K. Miller: Number one is I would describe to AI what you have done in the past and talk through all of your previous roles. Describe the tasks that you took on and very specifically the tasks you liked and didn't like. I don't care if you've been an accountant for the last nine years. Maybe you don't want to be an accountant anymore, right? So this is an opportunity to give all that weird nuance that you can't really give into a Google search. So what have you done? What tasks have you taken on? What did you like? What did you not like? What were the concerns you had at previous places? What types of companies do you enjoy? Big companies, small companies. What entices you about going into the office five days a week? Working from home the whole time? Remote and you get to fly to Italy once a month? Or never traveling because you're afraid of planes? Whatever the thing is. You want to add in all that context, and then you're going to say, "Give me three jobs that I'm a perfect, perfect fit for. Give me five jobs that you think I could be a fit for if I just told the right story. Give me five jobs that I could be a perfect fit for if I just took a couple courses—Google courses, Microsoft, LinkedIn, whatever courses. And give me five jobs that you think that I really, really want to reach for, but would be absolutely nuts if I went for it and would take me a year to make that pivot into." Right? And maybe that's going to tell you to go to a big boot camp or get your master's in some degree. That is the type of action plan that you can get with AI.
--:-- Once you get that back, you're going to then say, "Great, here's my resume. What are 20 changes I should make? What are 13 ways that I'm missing out on making this the perfect resume? Go out and find 150 examples of great resumes. Go and find 20 blog posts from Google, Microsoft, or from KPMG or BCG, wherever you want to get hired, and have those blogs synthesized. Give me five best practices and give me exactly how I should edit my resume." Great. Now you have an updated resume. Now you have a stronger action plan. Even the way you're going to do the writing and the outreach is going to be AI supported, is going to be AI first, right? "How can I make a splash and work for you, Mel?" Maybe it's going to tell me to show up at your offices and sing a telegram, right? Like, we don't really know. But you can ask for ways to stand out. You can ask for ways to pitch yourself. You can ask for ways to create your narrative. And even when you're in the interview, you know, "What are 20 questions I can ask this person to stand out?" Every little part of that job search process can be AI first. And then of course, being someone that knows how to use AI is going to make your resume that much stronger.
--:-- Mel Robbins: Sitting here listening, now I'm going, now I know I'm not getting hired by anybody because I'm not doing any of those things. But seriously though, isn't it also important because doing all that optimizes your resume to be scanned by AI?
--:-- Allie K. Miller: Yeah, there's this weird AI-eating-AI moment. You know, even when we're shopping for things online, if I have an AI agent shop for me and the car brand that I'm trying to buy from has an AI agent answering all of its sales questions—what are we doing? It's two AI agents acting as proxies for these people talking to one another. So it can feel very weird when you are creating things with AI that is then read by an AI. What I also want to advocate for is there are so many ways to stand out that have nothing to do with tech and online application. You could have AI help coach you through how to ask a common friend for an introduction. A lot of people feel very uncomfortable around that. Have it coach you through that moment of discomfort so you can push back, right? There are so many ways that you can use AI in the process, not just doing the work for you, writing your resume. And so having it coach you to ask for that, having it help you post on LinkedIn and say, "I'm sorry, I got laid off. I'm in a situation where I have these skills and I need help. I don't usually ask for help this publicly, but I need you." You've never written that post before. AI can find a thousand people who have posted that before and can help you get through that obstacle, that friction, so that you can get the life you want.
--:-- Mel Robbins: I love that because you're right. All of the things you just walked us through will help you leverage it for positioning yourself. But you keep reminding us that AI can also be this coach almost that can help you do the preparation, figure out how to have the conversation, practice the interview so that you're preparing so that when the real life stuff happens, you've actually prepared. Yeah. Using it that way is almost more important because you're not hiding behind it. You're using it to help you be more of yourself and to be a better communicator and be more effective.
--:-- Allie K. Miller: Yes. I think there's a lot of online discourse that AI is ruining our authenticity. When there are some people that could lean into using AI and actually help you live a more authentic life. I'm a weirdo in my life. I organized this big dumpling taste test from my friends. I had a friend who's a violinist come play and all of us laid down on the ground and just stared up at fake stars that I put up on the ceiling. Using AI to come up with weird whimsy ideas because Google is not going to be able to do that. You can live a more authentic life. Again, I'm not offloading to AI. I'm having it support me in the way that I want. So I think that's such an important idea because I'm still bringing myself into all these conversations, all these relationships, my job, my client conversations. You still have to be the person who's authentic, the person who's confident, the person who's earning trust. You're not going to—until we have brain-computer interfaces—it's still you, no matter how much you're using AI.
--:-- Mel Robbins: I would love to know, are there top-of-mind ways for caregivers to use AI to save time or find support and help that you've heard of that you can think of?
--:-- Allie K. Miller: Absolutely. First, let me say, there's an AI use case for everything. As a caregiver, one of my followers sent me an entire app that he built out. Again, does not require code. He is not a coder. He is just someone who played around enough to make this thing work. Okay. It summarizes all the emails that he gets from his kid's school. Yep. So that he knows exactly what's happening at the school. It summarizes every week. It gives him a calendar. It gives him action items. Okay. It even looks at the emails that he gets from his partner to be able to put that into the summary. And every single morning, automated, it gives him a summary to look at. And so the caregivers that I meet with, whether they're looking after children or family members or friends that they've taken in or parents, there is just so much noise. And for whatever reason, we've decided it's a good idea as a society to have like 20 different sources for this noise. AI can act as a really strong synthesizer that can pull in sources and can summarize things for you and make it digestible and can automate that sort of check-in.
--:-- Mel Robbins: Allie, let's take a quick pause. I have so many more questions that I want to dig into, but let's give our sponsors a chance to share a few words. And I want you to share this with people. Even if your kids are using this, there are ways that they could be using it better. Share it with them. Everybody at work, on your team, you guys need to adopt this. This is a free resource. I don't want you to be confused and worried and sitting back. There's no need for that. This is an incredible tool that can change your life and we are breaking it down so that you can use it. So share this episode. Don't go anywhere. We'll be waiting for you after this short break. Stay with me.
--:-- Mel Robbins: Welcome back. It's your buddy Mel Robbins. Today, you and I are getting to learn all kinds of things about AI from the woman that Time Magazine called one of the most influential voices in AI today. So Allie, the next thing I wanted to ask you, is there a particular prompt that if you're listening and you're like, "Okay, what's a problem you're dealing with?" Whether you have to plan the first birthday party for your kids and you're newly divorced and you need advice, or you are asking for a raise at work and you're scared to do it, or you have a neighbor that plays their music really loud and you don't know what to do, or as it was me this morning, I couldn't turn on my new Dyson blow dryer, right? So like there is a problem that you have. It could be anything. What is the prompt that you would recommend to the person that is leaning into this for the first time that helps you dip your toe into the water to solve something big or small that you have in your life?
--:-- Allie K. Miller: One structure that you can use is, "I'm a blank who's trying to blank." And by the way, these blanks are long bits of context. "I'm a 57-year-old woman and mother of three who is trying to turn on my hairdryer and I can't figure it in the hotel. And I'm trying to turn on that blow dryer. I have tried plugging it into multiple outlets. I have tried hitting the reset button. I have tried turning it off and on. I am nervous that I'm going to electrocute myself. I have double-checked the manual." Right? You can give it things that you've tried before, things that you want to do, things that you're worried about, methods that you might want to get done. Really, all I want to do is curl my hair, right? And then you can say, not just "what's the answer?" Right? And maybe blow dryer is one where you just might say, "What's the answer?" But if it's something more complicated, you're not just going to say, "What's the answer?" You're going to ask for tons of options for answers. And then you will also ask the AI to rank and score the answers.
--:-- So you might be—you had said yelling at your neighbor. I can give you an even more profound example. "You're a caregiver for your aging parents. Dad is coming to dementia. Your three siblings who live in different places are not helping. You're at your wit's end." That sounds like a problem.
--:-- Allie K. Miller: So you write in there, "I am a caregiver, and this is the situation, and this is what I'm looking to solve, and what are all the different answers?" Yes. And you can go crazy deep into these prompts. In addition to asking for the answers, you can also say, "What are five ways I should even think about this problem?" Right? And help me solution each of these ways. You might say, "I've already tried these three problems. Here's how it blew up in my face. Give me new ways of approaching this." You might say, "I think I already know the answer to this problem. Give me three ways this might go wrong." So you're going to bring in that context—things just about yourself, about the situation, the context that you're in, the environment. It is a different solution for every little problem. And the joy that I have when I use these AI systems is I tell it how weird and unique my situation is because there's no way that you can help me in my unique situation. I am a perfect little unique thing. No one's ever lived this life. And it helps me think through that problem.
--:-- Mel Robbins: Let's talk about accuracy. So where is the tech at this point? It's 2025 in terms of just general AI and accuracy of what it's spitting back to you. I'll give you an example. Last week, if you did a search for me, you would find out I was divorced, that I drive a Lamborghini. There would be all other kinds of things that are untrue. So the accuracy of these systems right now, the best models have a hallucination rate.
--:-- Allie K. Miller: A who?
--:-- Mel Robbins: Yeah, let's first—yeah, good call, good call, good call. A hallucination rate? It's like taking ayahuasca or mushrooms? Just making things up.
--:-- Allie K. Miller: So when we talked about how these systems are trained, right, we said give it tons and tons of millions of gigabytes of information. Okay. So the first thing is that these systems were not trained to be factual regurgitators. So the fact that it's so accurate all the time, even with these couple mistakes, the fact that it gives answers that outperform PhDs is actually pretty miraculous. The remainder of it, when it does hallucinate, we're getting to the point where models have about a 1% hallucination rate, meaning like you ask it a hundred questions and maybe 1% of the time it doesn't answer it on the first or one of the first 50 tries. Different benchmarks, whatever. But hallucinations have dropped a lot.
--:-- Mel Robbins: Wait, so is hallucinations just a term for it's wrong?
--:-- Allie K. Miller: It's just—it's when AI is spewing incorrect stuff that it's just like, "Maybe Mel has a Lamborghini." We're calling it hallucinations.
--:-- Mel Robbins: Well, what I like about what you just said, because now I get it, is you're doubling down on the fact that it's not, quote, fact, it's information. And there are ways to increase the fact so you can give it access to the internet so that things are cited and you can check the sources. You can then check the source of the Lamborghini thing and prove that it's not true.
--:-- Allie K. Miller: So it's got to guess. So if it's asking for a car and it doesn't know, it might be like, "Well, based on what we've heard and the fact that she has this, I don't know. We're feeling Lamborghini, not pickup truck." I don't know. Whether you knew it or not, you just said something that took researchers years to figure out.
--:-- Mel Robbins: What do you mean?
--:-- Allie K. Miller: We're just now seeing research around this space of why do we get things wrong, okay? Knowing that there are ways to improve it. We can ground it in information. We can use more state-of-the-art models. We can give it access to the internet, check citations, all this stuff. Why does it still BS us? Why does it still hallucinate? You just hit the nail on the head, which is we told these systems, "Be helpful to me." And these systems converted that task and said, "Oh, you want me to be helpful? You want me to always answer." Because when I say, "I don't know," that's not helpful to you. So because these systems were not given an off-ramp to say—they're not allowed to say "I don't know" because you have trained them. You've rewarded them by answering you.
--:-- Mel Robbins: I want to ask you, as one of the world's leading experts on AI, you're speaking on stages all over the world. You're a consultant to brands that people really trust. What are you most concerned about as this technology picks up speed?
--:-- Allie K. Miller: The first is the pace of change in AI. And I think it's really important to just level set on the type of acid reflux that even people in AI are feeling. I started in AI almost 20 years ago, and the pace of change is even faster than I would expect and that people in the field are expecting. Education heavily concerns me. The fact that companies have not yet leaned in and skilled up their employees, that's a really big one. The fact that parents have not leaned in to have these open conversations with their children about the risks, about mental health risks, about over-reliance, about misinformation, about cheating on schoolwork. I want more real talk happening in homes, in schools, in work, on the subway. I want that.
--:-- I think there are also very real concerns about data privacy and data use. I think there are very real concerns about the environment and how much energy or water usage these models or full systems are using. And just if you're not tracking with that, it's because they have to be powered by something. Everything is powered by something. Which means these huge cloud and server farms are powered by something.
--:-- Mel Robbins: Creating the ice that is in this water.
--:-- Allie K. Miller: Yeah, it doesn't live in the air. It's actually on a computer server somewhere.
--:-- Mel Robbins: Yeah, we call it the cloud, but really that means a data center in Arizona.
--:-- Allie K. Miller: Yes, exactly. And so these concerns are very, very real. There are some stats that have been shared by these companies. And I think one is by leaning into these systems and by being a user, you get to have a voice in these conversations. And you get to be a voice and say, "I've used it, and here's what I've seen, and here's what I think is stupid, and here's what I think is great." That's true. You get to be a loud, active contributor. And again, a concern is that there are going to be some people listening to this podcast around the world who are going, "I'm hesitant to use this," and their voices are going to be lost in the conversation.
--:-- Mel Robbins: I'm so glad you're saying that because I will personally say that I do fall into the camp of believing that this is the single biggest innovation, tech, human revolution that we will experience in our lifetime, that we can't even comprehend how it's going to change life for the better and in some ways for the worst in the next 10 years, but more for the better. And I appreciate you connecting the dots and saying, this is here, it's accelerating. And if you don't understand how to use it in your day-to-day life, you don't have a voice in demanding more regulation or demanding that things get labeled as AI-generated. Or, if we are creating tools that can create things, then we should be creating tools that can also police things. And so I really see the connection there. And if you're worried about it, don't sit back. If you're worried about it, this is when you lean in.
--:-- Allie K. Miller: I am also in the camp that some of these concerns are made a little bit more dramatic than they actually are. And that demanding more transparency and documentation from these providers has been very fruitful in shedding more light on that. Video streaming for an hour versus AI chat for an hour. You want to guess the energy consumption, the comparison?
--:-- Mel Robbins: Zero idea. I'll be honest with you. I don't even think about this. Yeah. Because I'm thinking it's coming from my job. Right. I'm not even thinking about the larger implications of this.
--:-- Allie K. Miller: There are two things that are true at the same time. It is absolutely a concern. We should be voicing our concern for it. We should be asking for more transparency and documentation from these players, from these builders. And it is not as dramatic as people make it out to be. We can compare it to video streaming. Video streaming uses over 4x the energy of AI chat for the exact same amount of time. So you know, using Netflix less and chatting with AI—that might actually be a trade that's good.
--:-- Mel Robbins: What do you want to say to somebody, Allie, about the fear that AI is coming for my job?
--:-- Allie K. Miller: I think we will have job loss because of AI. We need to be very, very honest about that. And I don't know at what scale, and I don't know on what timeline. But I feel strongly that I should say that out loud to be a responsible citizen. The second is that every single job that we already have out there—marketing manager, legal, finance—will be AI supported. And you'll have a switch in the types of things that you are doing. And so maybe if you are a marketing manager, let's say, and right now you are writing a lot of copy, you are constantly going back and forth and checking on stats, you might have an AI that is literally just constantly checking your metrics for you and flagging when things are out of sorts and offering 20 potential solutions that you could pick one of, or you could say, "I know my business better than you. I'm going to pick the 21st." So the job of each person is going to shift, right?
--:-- Allie K. Miller: Even in legal. I know people who are using it to do contract comparisons or clause-like risk analysis.
--:-- Mel Robbins: Hey, and this, by the way, as a solopreneur, I do this too. Maybe instead of saying AI is coming for my job, the reframe is AI is a part of my job. AI will be a part of everyone's job. AI is coming for some jobs and there will be new jobs because of AI. Can you unpack why women are slower to adopt AI than men?
--:-- Allie K. Miller: I think a lot of people in AI are men. And so when you're looking at people talking about it, it's going to be largely men. And so there's going to be a little bit more of like, "Ah, that future is not for me. That's the tech bros or whatever." So one is just that they don't see people like them. That is one reason why I spend every waking minute trying to share more information and make this world more accessible and why I've educated millions on this space.
--:-- So one is, "Is this future for me?" That also means that the use cases that are shared might also not be as relevant. Like women more often are taking on care for others, care for their children, care for aging parents, teachers. And so those stories are just told less. And so we get into this toxic flywheel of those stories not being told.
--:-- We also maybe have, again it's anecdotal, but when I speak at conferences, I am more often asked about data privacy and environmental concerns from women. And again, I want to give a path forward to those folks that feel that hesitancy. And there is a very fast way of finding action there. I'm going to give this as a tip, okay? I want everyone who feels this way about AI, that you're worried about data privacy or you're worried about maybe environmental usage: you can download an open source small model and you can run it on your computer. It will never go into the cloud. It will only live on your laptop. The only electricity that is used is downloading it and the energy that your laptop needs. It's a smaller model, so it's also going to have a smaller footprint. I am able to use AI in the skies with no internet access because of this local deployment. So that is the path. If you are still hesitant, please try out small local models. You can do it in less than five minutes.
--:-- Mel Robbins: You know, Allie, one thing that I saw a couple months ago was kind of the first—it wasn't really a study, you probably know more about this—it was done here in Boston at MIT. And it was the first look at cognitive decline of people using AI. And the results were alarming. Like there was a significant decline in people's brain power. That's not the scientific term, which basically in my layman terms, I read that and was like, "Oh my God, people are getting stupider using this. Their brains are rotting." And it wasn't a clinical study, but it was looking at people overly relying on AI and the impact it has on your thinking skills and your brainpower. Is there such a thing at this point that we know of as relying too much on AI?
--:-- Allie K. Miller: Over-reliance is a risk of many tech systems, including AI. And that study, I think, thankfully illuminated a key point, which is: yeah, if you use these systems lazily, you're going to get lazy. So in the same way that we still teach our children math, even though they have calculators, we still need to teach our children taste, curation, critical thinking, creativity, writing, the ability to cast judgment on whether a fact is right or wrong. We still need to teach children that.
--:-- That study was about people using AI to write essays. And the outcome was that people couldn't remember what they wrote in an essay.
--:-- Mel Robbins: Of course you couldn't remember. You didn't write it.
--:-- Allie K. Miller: You didn't write it. Exactly. And if the goal is to be able to remember what you write, then yeah, you should still do the writing. You can still use AI to interview you to get more information out. You can use AI to review it from the viewpoint of Abraham Lincoln or Mark Zuckerberg or whatever. You can have AI review it and make it better.
--:-- Mel Robbins: Got it. So there is a spectrum of right and wrong ways to use AI. There is a role that humans play in our world, which is bringing heart and empathy into situations. There are things that I also think are gray area that some people have said, "Hey, I'm going to do this." I've heard of people using AI to write obituaries or statements at a funeral.
--:-- Allie K. Miller: If you go to a wedding, a lot of the speeches sound the same.
--:-- Mel Robbins: Absolutely.
--:-- Allie K. Miller: Right. Let's delve into their relationship, the landscape of the tapestry of love, right? That could be a gray area. I think, honestly, one of the biggest takeaways that I've had over the last seven years in the Gen AI space is that urgency is creating toxicity. If you are under the gun, you have to write this report, you have to write this essay, it's five minutes before the wedding and you forgot to write the speech, that is when you're going to lazily offload and abuse these systems and not get the great writing out of it and not speak from the heart and not build a better relationship with your friend that's getting married.
--:-- So I think the more that you can do to eliminate urgency, which as a procrastinator is absolute hell, is going to help you use these systems in the way you want to use them. Again, there are going to be some people in that gray area that still say, "Hey, that's fine." And that's everyone's prerogative is to have that voice in this conversation. But urgency—and removing it—is going to help you make better work, use AI more responsibly.
--:-- Mel Robbins: You know, when you think about AI long-term, what excites you the most? Seriously. Like, first I want you to talk about what excites you the most. And then I'd love you to talk to me and to the person that's listening and watching about what might be coming in the coming months.
--:-- Allie K. Miller: Two things that excite me, and they're very closely related. Number one is the accessibility of these tools is only increasing. So two years ago, you had to be this like perfect prompter. Now you can kind of type a couple sentences and it gives you a really strong output.
--:-- Mel Robbins: You can also speak to it.
--:-- Allie K. Miller: You can also speak to it.
--:-- Mel Robbins: You can also film and upload photos to it.
--:-- Allie K. Miller: Yes. There are so many ways to interact with these systems. So the accessibility, and to me, the inevitable downstream impact of more accessible systems is that people that are burning inside with this amazing idea that they've never been able to accomplish, or this problem that they wish they could have solved seven years ago, or this kid that they want to bond with more, or a parent that they want to help more—everyone has this burning thing inside. It might take a little bit to figure out what that is, but the ability to accomplish that thing, those obstacles are dropping very quickly.
--:-- We are going to have billion-dollar companies with a couple people, and we might see billion-dollar companies with one person. The ability for someone to scale their authenticity and their impact and the types of change and helping each other that they want to have is going to explode even more than we've already seen.
--:-- Allie K. Miller: So I want people who feel left out to lean in even more because again, that ability to go from idea to execution on anything is going to compress. That is what excites me.
--:-- The things that we should expect to come—and again, I can declare my predictions, they might change all the time. Experts are always sharing their predictions and we are constantly changing it. So again, listen to a variety of voices. Anyone that declares for sure that something is happening the next 40 years, whatever, they're guessing. Everyone's guessing.
--:-- Okay, number one: it feels like it is very, very likely that we will have a much more multimodal world. What does that mean? That means modality could be text or vision, like visual things, or audio. So the ability to not just type in and say, "Make me an image of Mel Robbins posing as Wonder Woman on top of a hill," but the ability to go in and out of these different inputs—not just text to image, but image to sound, sound to movie, movie to blog post.
--:-- I legitimately think that we will be able to talk to our pets in the next 10 years.
--:-- Allie K. Miller: Because these systems, again, the ability to translate is an emerging capability that's coming out of these really, really big models. So I could put a phone at my dog's face and be filming him and go, "What is YOLO thinking?" It's trying to tell me. It's a guess, but it feels more likely than not. And there's research happening, by the way, already in dolphins.
--:-- Mel Robbins: Really?
--:-- Allie K. Miller: Yeah. So when Homie puts his paw on me, I'm like, "Okay, yeah."
--:-- Mel Robbins: Love it. Home slice. Homie. Yeah.
--:-- Allie K. Miller: It's again, multimodal. You could view it as it's easier to put in information and easier to get out information. But that also means that if you want to learn quantum computing and you really like podcasts or you really like video, maybe instead of reading a 700-page book that is really scientific and dense, you could say, "Hey, can you make me a 25-page PowerPoint?"
--:-- Mel Robbins: Well, I'll tell you what I'm excited about. I'm really excited about the fact that so many people—and I know that I have absolutely felt this way in my life—feel alone. And you feel like it's all on you. And the way that you've explained what is already available right now that is there for free to act as an extension or a team member or a thing that you can delegate a task to that then expands your time, expands your capacity, awakens you to options, helps you create a plan, saves you time—you're not actually alone anymore.
--:-- Allie K. Miller: Can I also just—
--:-- Allie K. Miller: A hundred percent yes. There are going to be people who just heard you get it and they're going to go, "Well, now there's too many options. And how on earth am I supposed to change my whole life when all I see is a blank page?" So I just want to also tell the person listening: it's okay if you don't have that moment of reinvention for the first couple of weeks you're using it. Give yourself the space to fail, to be weird with it, to ask new questions, to try and break it. And it's okay to delay that light bulb moment. Don't punish yourself if you don't have that early on. That's so normal. And I don't want that person to feel like they are behind in any way because, in fact, they're quite ahead if they're willing to do this.
--:-- Mel Robbins: Well, the fact that you've just spent all this time listening to or watching this means you're ahead.
--:-- Allie K. Miller: Oh my God. Very much so. Yes.
--:-- Mel Robbins: On that note, if I were to take one action—I mean, you've told us so many exciting things, specific things to do, things you're concerned about. But if I were to just take one action, what's the most important action you should take after everything that you've taught us today?
--:-- Allie K. Miller: If you have not been using AI, use it. Not because I'm telling you you have to use it every single day or else the world will explode. But I'm saying, I want your voice in the conversation. And by you experimenting and seeing the strengths and weaknesses of these systems, you will be a stronger voice in the conversation. And you will be included. You will get to say, "I want these systems to serve me." And right now we are missing some voices.
--:-- So for those who are hesitant, use it. For those who have been using it, it is not Google. And you have to get out of that mindset. You have to treat it like this alien. You have to try and do some real-time interaction. You have to try and delegate a 20-minute task to it. You have to try a live voice conversation. You have to get the superpowers out of it. And that means gaining more clarity, using it for more forethought, using it to 10x—maybe even when you're naming this podcast, ask AI to come up with 250 options. You're still going to be the human that curates and picks and moves things around and maybe rewrites it all, but you need to lean into the superpowers of AI, not just better browsing.
--:-- Mel Robbins: One thing I'd love to have you end on is you have said repeatedly, you're excited because you can use AI to help you become the person you've always wanted to be. Can you speak directly to the person listening and tell them what that means?
--:-- Allie K. Miller: I'm going to give you an example from my own life. I moved to New York three years ago. I had just come off of a three-year road trip. I lost everything that I owned when my apartment flooded with sewage. So I'm moving to a new city that I've never lived in, into an apartment with zero furniture, zero spoons, zero lamps. I am sitting on the floor. My butt hurts because the floor is so hard. And I burst into tears while eating like Annie's pretzels.
--:-- And I'm talking to my therapist the next day. And I'm saying, "I can't do this. I'm depressed. I can't even get out of bed. I'm literally eating dinner by myself, sitting on the side of my bathtub because that's the only thing that's elevated." She goes, "Wait a second, Allie, did you just say that you have an empty apartment?" I was like, "Yeah." She goes, "So you have a dance floor in New York? How many people have a dance floor in New York?" I was like, "Say that again?"
--:-- And she completely flipped how I thought about this problem. And suddenly I literally hosted a dance party in my apartment. I had friends come over. We had a YouTube video. We did Zumba stuff, acting like an idiot, in a dark, empty apartment. I also organized a New Year's planning session where we covered the entire floor with post-its because I could.
--:-- That gave me an idea to go to these systems and to say, "Here is a transformation that I've had in my life because of this woman, because of one sentence that she asked me. I need to do this on repeat. I need every single time I come to you with a problem, you're going to give me the reframe, you're going to give me another reframe, you're going to give me a motivational sentence that tells me I can absolutely accomplish this, you're going to give me action items that I can get it done."
--:-- And so I built—again, zero code, took two minutes—I built a repeatable way to go to these systems with a problem and to see it through a new light. It completely rewired my brain. I used to go to this thing multiple times a day. I haven't had to go to it in the last couple months because that's just how my brain processes bad things now.
--:-- So if I am very stressed about meeting with an executive or whatever, I go to the system, maybe, and I say, "I'm really stressed about this meeting." And they go, "You're probably stressed because you know that it's important. You have a successful career because you've been given this important meeting. Good for you for being successful. Own that success and know that with success comes stressful moments. And you got to where you are because you dealt with less stressful moments, but that bar is going to keep increasing. Good for you for already surviving everything you've gotten."
--:-- That is the type of transformation that I am working with these systems on. And again, it has completely rewired my brain. I now look at stressful situations, at anxiety, as an opportunity for reinvention.
--:-- Mel Robbins: Amazing. I just want to thank you. I want to thank you for making the time to learn about this exciting tool. I mean, I realize there's so much I have to learn. So I'm so proud of you for listening to this. And I'm proud of you for watching this on YouTube. And thank you for sharing this with people in your life. We all need to lean in and learn how to use this tool that's right there that could make our lives better.
--:-- And one more thing, in case no one else tells you today, I wanted to be sure to tell you as your friend that I love you and I believe in you. And I believe in your ability to create a better life. And I'll tell you something, after the conversation today, I am 1000% convinced that you can use AI as a tool to create a better life. And I hope you feel empowered to start doing so.
--:-- Alrighty, I will see you in the very next episode. I'll be there to welcome you in the moment you hit play.
--:-- [Pause, then informal chatter]
--:-- Allie K. Miller: I, um, here's how I think about AI. Yeah.
--:-- Mel Robbins: Is everything okay?
--:-- Allie K. Miller: I feel like all of a sudden a whey protein shake burp is coming. Here we go.
--:-- Mel Robbins: I burp all the time. Let's do it.
--:-- Allie K. Miller: Now, can I ask you a question? I think you've been doing it for a couple hours.
--:-- Allie K. Miller: Only just now realizing how many cameras.
--:-- Mel Robbins: You are way better prepared than I am.
--:-- Allie K. Miller: Well, it helps that I have a suitcase that's the same.
--:-- Mel Robbins: I went to sneeze on stage and literally peed, and I could feel it hit my ankles in front of an audience at J.P. Morgan as I was wearing a skirt.
--:-- Allie K. Miller: Long one, thank God.
--:-- Mel Robbins: That was going to be my follow-up.
--:-- Allie K. Miller: Yes. Like, skirt. If I had felt the pee when you're wearing pants on your ankles, that's more of a waterfall, Allie.
--:-- Mel Robbins: Fabulous. You nailed it. That was what we needed. Thank you. You killed it.
--:-- Allie K. Miller: Okay. Okay. Yeah. Okay. That's how we do it.
--:-- Mel Robbins: Oh, and one more thing. And no, this is not a blooper. This is the legal language. You know, what the lawyers write and what I need to read to you. This podcast is presented solely for educational and entertainment purposes. I'm just your friend. I am not a licensed therapist. And this podcast is not intended as a substitute for the advice of a physician, professional coach, psychotherapist, or other qualified professional. Got it? Good. I'll see you in the next episode.
