Revolutionizing Marketing Efficiency with Advanced AI Strategies

Eric Eden: 1:00

Welcome to today's episode of the Remarkable Marketing Podcast. Our guest today is Jonathan Green. Jonathan is an expert in AI and using it in business to be more productive. He has a book on that same topic that we'll talk about at the end, but he also has some great stories for us today about how marketers and people are using AI at work to be more efficient. Welcome to the show, Jonathan.

Jonathan Green: 1:32

Thank you so much for having me. I'm really excited to be here.

Eric Eden: 1:35

Awesome. So I'm very excited about the topic of AI. I've been using it as a marketing leader for all kinds of productivity reasons in the last year. But let's jump right in. Tell us about what you and other people have been using AI for to be more productive, more efficient and to grow their business.

Jonathan Green: 1:58

A biggest use for me is testing ideas. I'm someone who always wants to try this and try that, and one of the things that would keep me back is if something takes six weeks. I love playing around with coloring books. It's one of my hobbies, but it's not a main part of my business. But it's a great way to drive in traffic and in the past it would take me six weeks because I would have to hire someone or go through a slow process. Now I can go from idea to the book is live in two hours, which means that I don't get emotionally attached to an idea and that I can have that traffic in those sales coming in really quickly. And if an idea is good, then I make more of something similar. And a lot of other projects are like that. My friend Debbie, who likes to just have a lot of free time she always gets slowed down in the writing of the sales page. Right, that's the thing that catches a lot of us up. We have a great idea, but then writing a sales page is just a completely different skill and it takes a really long time. And now she uses AI to help her come up with 10 versions of each headline. So, instead of her having to come up from scratch, she just chooses a winner for each part of the sales page. And then last year she spent a lot more time working on her pickleball career and she started winning a bunch of tournaments in that free time and going on all these crazy trips and sending me pictures, and in fact she's in Mexico at a tournament this week. So all of that opened up because she found that additional free time. I also work with someone named Travis who helps people to write their books, and a huge part of his delay in the process is sitting down with each person and turning their notes into an outline. Together we designed an AI process where now he just uploads the notes and the AI, in less than five minutes, shoots out an entire outline for him. So he's able to take every Friday off and spend more time with his family.

Eric Eden: 3:42

That's fantastic. I am using AI for marketing projects that I work on with companies. I've seen a similar sort of increase in productivity when I do projects like product marketing messaging for companies that would usually take two weeks of writing work. I've been able, working with chat GBT, to generate the draft content and more like two days instead of two weeks. And it comes down to things like you're saying is, I can use it to generate a list of thousands of product names and then filter that down to something like 20 that people can choose from, and sometimes people look at those 20 and then they pick one or they come up with an iteration that is a combination of some of the 20. But it's a lot faster than having to start from scratch. I love the example of also coming up with outlines. So it's not just about writing content, but having the AI come up with outlines for you, or you gave the example of using it to create headlines, like the subject line for emails is always something that I would, shockingly, over the years with my team, spend a lot of time on, because the right subject line makes all the difference in the conversion for the campaign. So your friend getting to take Fridays off is amazing because some of that stuff just takes hours to come up with the right inspiration. And if chat GPT can help come up with that inspiration, that's awesome. What other sorts of things are you doing or seeing people do to get those sorts of gains?

Jonathan Green: 5:24

One of the worst parts for me of podcasting is everything after the episode. So I love the interview when I have a guest on my show, but then I have to create show notes and make social media clips and send them an email. Now, with AI, it takes me about one hour to do that entire process and everything is automated by AI. So it makes the clips, it puts the text on it so it looks cool and tiktok-y and all of those parts are done, and it automatically sends the emails for me, because I always forget I try and remember to send someone email 10 weeks later where their episodes come out. I'm bad at all of that. So I was able to set up an entire automated process using AI that removes the least favorite part of the week. I love the front part of the podcast and the back part just turns into all work because the fun part's over. So that's really lead to a lot of growth in my podcast and I can spend more time forging relationships. I've made some really good friendships with some recent guests on my show. Then the past was impossible because I was like, oh my gosh, I've spent 10 or 20 hours trying to process this and find the right clip and cut out all the parts where I say words I don't like. Now it's all automated and really it edits the episode. There's even a tool that I push one button and it just shows whoever's speaking. I used to have to do that manually, switching back and forth to whoever's in front of the camera, and that alone saved me like five or six hours, and now it takes less than a minute.

Eric Eden: 6:44

Yeah, the tools around AI for use cases like podcasting are amazing. Another example and you're probably doing this is it'll auto generate the transcripts from each episode and you can see that right on the podcast session page and I think Apple is getting ready to launch that feature this spring out for all podcasts. But some of the podcasting tools like Buzzsprout and others you just upload the episode and it'll give the AI will give you the whole transcript of it and then that's great for things like SEO, and there's been tools for years that would do transcripts, but you'd have to pay extra for it and it wasn't quite as good. There's still some things I would say. For example, sometimes you have to check it over because in the things that the AI writes, it'll misspell the guests' names and things like that, so it's not completely autopilot. I wouldn't recommend that, but it's still. I would say would do 70, 80% of the work for you. It suggests headlines for the episodes and it may not get it perfect, but a lot of times I'll be like, yeah, one of them is good, or I'll combine a couple of them, or it'll give me inspiration for something, and sometimes I just do have to still come up with my own, but I would say it's a 70 or 80% efficiency boost, which is pretty huge when a lot of people are trying to do a good volume, decent volume of episodes, right.

Jonathan Green: 8:16

Yeah, we now take transcription for granted. When I first started dabbling in transcription five or six years ago, I had to hire two full-time transcriptionists to work for me just to get good transcription of my content. Because I was trying out tools and the big one at the time was called Dragon Dictate and they said, oh, we have a 95% accuracy, but that means every 20th word is wrong. That's so many mistakes and it would be impossible to use because it would give you no punctuation. So just imagine a wall of text, no punctuation, no spaces between paragraphs, thousands of words and one out of every 20 words is wrong. So even if you're trying to tell, I was like what was I talking about? So I would have to go back and listen to the audio. Now there's probably six things that we don't even realize a transcriber in this conversation, because I always find out oh, this other app has added a transcription feature. Now my mic probably has a transcriber inside of it, so it's become ubiquitous. Something that used to be such a challenge and that's what's really exciting is that all these things that were so inaccessible or so expensive, now they've become so affordable and they're available to the masses, so it's really lowering the barrier to entry for everyone.

Eric Eden: 9:27

Absolutely, and so, beyond generating the content which the generative AI is good at, it's not perfect, it needs to be reviewed, it needs to be edited, but it is a far step from starting from scratch. Some of the use cases that I'm seeing people use AI for are beyond even just the binary thing of generating the first draft of content. They're doing things like they're using AI to create personas and training the AI in those personas, and they're using it to train AI to talk in their brand voice. Then they're using AI to go and analyze all of their content library on their website and documents they have and to train it on their voice, and then for the personas they have say they have five personas or industry verticals or targeting they can ask the AI what areas are they lacking or have gaps in terms of content for their personas and then use the AI to generate more content in their voice for those personas, and that's a very advanced use case in terms of senior product marketing. People would have to do that sort of analysis at companies that I've been at in the past. Are you seeing people taking things to the next level and doing things like that?

Jonathan Green: 11:01

Yeah, that's a big part of my job is creating personas, creating custom characters which work, called personas. When I started doing it and one of the things I discovered is that a lot more people than you realize are in there. Because I said to chat GBT one time, right in the style of Jonathan Green, and it started writing exactly like me. I thought it was going to choose the more famous one who writes movies, because I'm like the eighth most famous Jonathan Green and it started writing in my style and I go wait, which are they're using? And then just exactly describe me and it read my book. I said, oh my gosh, I didn't realize you were reading people's books. I thought you weren't supposed to do that chat GBT, but I'm already in there. But I was in that exact process of training it and now the training process is getting easier and easier and what we're seeing with GPT is that they're becoming available to everyone. So I sometimes use my own bots and now I can have my bot talk to another bot someone else has created, so I don't have to design the part that, like, makes it watch YouTube videos. So now I can have a YouTube video watching bot, then pass that information to my customized personality bot. So things are getting really crazy and really exciting. And a lot of tasks that were really hard, like figuring out what's your unique selling proposition or who's your customer avatar these are really hard for someone with a new business who hasn't sold anything yet, I go. I don't know anyone who will buy my thing. Now, that was one of the first big breakthroughs I had where I could ask that question and chat. Gb goes great. Let's figure it out together and we answered a couple of questions and it goes based on your product, based on your personality. Here's what I think and it was the best analysis of a customer avatar I've ever seen. I was like I've been teaching this for seven years and you make everything I have been teaching look like a waste of time because it does it so well. It's really made some of the things that are the hardest for a new entrepreneur that are now the easiest.

Eric Eden: 12:42

Absolutely, even things like coming up with a name for a company or a product. I've been using chat GP to do that and it's just so hard to do because there's 350 million names that are taken, with domain names that exist and 100 million trademarks. When you sit down trying to brainstorm, if you don't have that AI, a logarithm behind you trying to brainstorm a lot of names, it's really hard to come up with names that are available, for example. And once you get a name, I'm working with people who are having great success in generating logos for that name and then avatars. Like you're saying, those things took so many hours, days, weeks to do, because just the sheer processing power is mind boggling that it takes to do somebody's task. But it's very easy for the logarithms right.

Jonathan Green: 13:40

Yeah, I always found logos so stressful because I would always work with a designer and then you keep asking for changes until they get mad and you go okay, go back one level, Like I'll stop asking for iterations and now I can ask for as many. I can just hit the button and just get as many designs I want and not have that stress.

Eric Eden: 13:59

Now have you found I was reading some research that if you're nice to the AI, you can actually get better results? If you give an encouragement like you would normally give an employee and you let it know how important it is to you and that this is a really critical task, have you seen that that actually gets you better results?

Jonathan Green: 14:20

Yes, it does. Chatcha BT is trained on a plus one minus one model. Plus one is ultimate positive. Wow, chatcha B, that's the best response I've ever seen. And negative one is you've ruined my life. I hate you, chatcha BT. Now my wife is going to leave me and it's all your fault. So it's trained to pursue positive one. So let's say it's talking to you and me at the same time. I give a conference all the time and you never do, no matter how good a response it gives to you. So that means maybe it's going to give me just 1% more resources. But for a computer that's that powerful, that makes a big difference. So that's why it rewards people that say nice things, because that's built into its programming. It's a pseudo emotion, it's a false emotion, but it's how it's trained.

Eric Eden: 15:04

It's very interesting. There's a side benefit if one day the robots rebel, if you're nice to it, they'll be on your side. I guess the other question I had was broadly across business. What are you seeing to be the most remarkable or impactful use of AI today in business?

Jonathan Green: 15:28

I think the most significant thing is coming from the open source world. So because there's such good free AIs out there all these companies that's why everyone charges $20 a month chat GPT $20 a month, perplexity $20 a month, google One $20 a month, because they know if they go above a certain number, everyone will just switch to the free version. So there's this check on pricing, which means now it's affordable to everyone, because software that used to cost $300 a month is now $20 a month. It's really made it available to new entrepreneurs and to a lot of people. And the second thing is that the free content, the free AIs, are so good, whether it's chat GPT free version or stability AIs free version. A lot of the free ones out there are so good that it's opened the door to people whose first language is in English. The biggest barrier for a non-native English speaker who's trying to make living writing articles is that they couldn't write perfect English. That barrier just got erased. They don't make grammatical mistakes, they don't make spelling mistakes. They don't do anything that gives away their native language anymore, because chat GPT, when you run it through, or any AI, will fix all those mistakes automatically with its lowest level of editing process. So it's really, I think, helped those people. The most people who are at the bottom of the market have really been raised up because now they have a tool that doesn't cost anything, they have access to it everywhere and they're no longer limited by where they're from or their native language.

Eric Eden: 16:49

Yeah, it's pretty awesome. So what are your top tools that you're using that you would recommend to other people? If you have a top three or top five tools that your AI tools are using, what would this be?

Jonathan Green: 17:07

My main tool for writing is chat GPT. My main tool for image generation is mid-journey. My main tool for anything to do with video from editing to green screen removal to audio enhancement is descript. My main tool for research is perplexity AI. Those four tools are all you need to be a well-rounded, super-scaled person who can do anything you want.

Eric Eden: 17:32

That sounds like a great tech stack. Let me ask you about. You mentioned GPTs. I'm not sure that everyone knows within chat GPT that you can create GPTs. Can you talk a little bit about that and what you're doing with that?

Jonathan Green: 17:49

Yeah, so you can pre-create a character that does a specific thing. So I have a specific character that just writes for LinkedIn and another one that writes for Twitter, because they talk completely differently. And I have a persona that writes TikTok scripts. She's very annoying because she writes in seven-second chunks and talks with a lot of emojis, but that's how most TikTok people are, so it makes things it fits, and so for each different skill I can create that persona and then it does those tasks for me. Sometimes I build them custom for people and I build my own. So for my podcast I have one called Patrick Robert AI, because he's my podcast researcher, so I always make the letters match. I can post a link to a guest's website and he'll give me 25 questions related to AI for that guest, based on their website or based on their book, if I upload their book. So I'm able to do a better interview without having to do because I don't have time to read everyone's book myself, even though I'd love to. So it allows me to give that experience. They get on the late night show where they have 300 pre-interview people, so I can do those amazing things.

Eric Eden: 18:50

So what do you think this is going? Some people are very critical about, even though the massive progress we've talked about so far in this episode. Some people are critical of where we're at. They cite issues around the hallucination, inaccurate information and the fact that you still have to edit it, like I was sharing earlier. Where do you think this is going? Do we get to GPT seven or eight and it's more ability to be automated and it's much more authentic like a human? I'm just curious. It's moving very fast. We've been talking about that like in the last year and a half. It's just moved so fast. We have new things like text to video that OpenAI just announced. Where do you think this is going? What's your prediction? How long do you think it takes? No-transcript.

Jonathan Green: 19:39

I think that we're approaching the singularity as far as technology, which is we're reaching a point where the technology is happening so fast that we can't keep up with it itself perpetuating right, whereas maybe ChadGBT4 is going to create ChadGBT5. We're moving that fast. They're talking about AGI, which is sentient AI, being about seven months away. I don't know if that's true. That's like an intense jump, but we're seeing things happen at such a pace. I have to check the news every day for three hours because there's a major revolution almost every three or four days. So something major happens every week for sure, and usually it's multiple things. So I think that we're approaching a lot of speed technology. Will it lead to a post-scarcity society? I don't know about that because there's a lot of other factors at play. Will it lead to a better world or a worse world? I just don't know. But I know we've already crossed the Rubicon of inevitability. So whatever's going to happen, it's going to happen. It's too late. You can't put the genie back inside the bottle. We saw Metallica try that with Napster for five years of lawsuits. It didn't work. We now nobody buys CDs anymore. Those days are over. We download music. It just is what it is so. The future is happening and it's just a matter of accepting it or trying to live in denial, but it's inevitable at this point.

Eric Eden: 21:01

So you don't think regulation and holding it back is realistic?

Jonathan Green: 21:06

No, I think that people try to regulate AI or dummies, because it's the same type of people that have tried to regulate spam. How's that worked out? Have you stopped getting spam emails since they started regulating in the 90s?

Eric Eden: 21:20

What about some of the legal issues that are around it in terms of ownership, in terms of copyrights? The interesting thing from my perspective is, over the years, I have spent a lot of money buying stock images from the big stock image companies, thousands and thousands of dollars. And now I'm not sure why I would ever buy another stock image again, and I think some of those companies have big lawsuits going already. But do you think that, again, it's inevitable?

Jonathan Green: 21:57

So the question of copyright is a very good one and I think the answer that will have to be cited in the courts. We do know that anything created by an AI will never be owned by anyone. So if you write a book with AI, you don't really own it. You can publish it first, just like anything else it's anything creative commons but you don't really own it. If you have an AI generated image, you don't really own it. You could be the first one to use it. We have to figure out what that means, and just like we had to figure out what copyright meant, how long a copyright meant. Like Mickey Mouse just fell into public domain, winnie the Pooh is in public domain. So these things that we've always thought of as being owned by one person, now anyone can make a Winnie the Pooh movie. So those came from concepts. In other countries have different definitions of fair use. So I think those things do have to go to court and we have to decide as a people what is okay and not okay. Chatgbt read my book and I'm in there, so I'm a victim of it. So I had the choice of going. I'm going to sue ChatGBT. Take my information out of there. How dare you do that? Or I can go how can I leverage this to get maximum advantage for myself? Now, I don't have to train the AI. It's trained on my behalf. So it's a mindset thing. And there's this saying if they replace everyone job with robots, I want to be the robot wrangler, I'll be the robot repairman you have to look for how can I adapt to what's coming? If you spend more time thinking about that, you're more likely to succeed. I do think there's a lot of questions. I think that, like the New York Times is suing open AI about using their content, and I guarantee you, the New York Times is writing articles using ChatGBT. That's all Sports Illustrator was doing. That's why they went out of business. They got caught because they made up fake people using fake images designed by open AI and articles written by open AI. So a lot of big companies are the ones most guilty of abusing and misusing these tools. I have a feeling that every single one of these companies that's filing a lawsuit is also guilty of using these tools, way more than you think. If someone accuses you of cheating, they probably are the one who is cheating. So it's the same thing. I have the same suspicion that we're going to find out, when they do discovery, that all these companies change their mind and drop their lawsuits.

Eric Eden: 24:04

So a lot of companies have banned their teams from using tools like ChatGBT because they're worried that employees are going to put their proprietary information in there. I think there's been some solves on the teams or enterprise versions of ChatGBT where supposedly you can upload things and it won't load it into the publicly available language model, but I still hear a lot of people being very concerned about that, a lot of companies banning people from doing it, but the temptation seems almost irresistible to employees. When at the beginning of the episode we were talking about the productivity gains, it just seems irresistible for people to do it All the way through some very large companies that we're going to build our own language model. I'm not sure how realistic that is, but I'm curious what are your thoughts about people being concerned about that and trying to ban it within their company or organization?

Jonathan Green: 25:08

So I know I was talking to someone who worked at an embassy and they have employees who will use Google Translate on top secret documents, which is not secure. I've never thought of Google Translate as secure, so people already are doing silly stuff. I can tell you that they will, no matter what they say. 49% of open AI is owned by Microsoft. Anything you put in there, they're stealing. I'll just say that right now. There's no way they're not. Microsoft was founded by stealing MS-DOS from another developer. If that's your foundation, what you're not gonna change and that's just how it is. If you and that's the same thing If you put your top secret stuff on my computer and then say don't look at it, there are ways to be secure. On the other end, you can create instances of an AI that's run locally on your machine. You can use an air gap to computer, which means it's never connected to the internet. You can run an AI on it, do all of your work and then hit it with a magnet and destroy the records. That's the highest level of security. People won't do that because it's inconvenient, but you can use an AI in that way and there's no security risk In between. There are other ways of doing it. This is one of the things I help people with consulting. I talk about what you need to do. Especially like healthcare companies have specific requirements because of the HIPAA laws. They're not allowed to leak patient data even, and so if a doctor is using chat GBT to help write medical notes, that doctor could be in a lot of trouble Like a mass. That's a massive violation of a security act. I do think that people have to learn to be fastidious and to understand that not all tools are created equal, and then there are tools out there that are just as good. Chat GBT is the most well-known, but there are tools built with security and just a question of are you sending your data off-site or not? That's the first thing. The second you're sending your data through the internet. You have a security hole, you just do. But you can run a local AI inside your company. You don't have to build your own AI from scratch. There are so many good open source models that have beaten chat GBTs, so chat GB 3.5 is not even in the top 100 of the free models. So there are so many open source AI tools that are completely unrestricted. You download it to your computer, you run it locally and there's no risk. So there's a lot of options in that space. But I think people just have to learn Like people have to learn that if you find a USB key in the parking lot, don't plug it into your computer to see what it's gonna do. So they have to learn security policies.

Eric Eden: 27:36

That makes sense. In the race between open AI, chat, gbt, microsoft, copilot, the Google's new Gemini offering, who do you think comes out on top? Who do you think is on top now and who do you think comes out on top a year from now?

Jonathan Green: 27:58

I think Google has no chance. They've made three major missteps in the last year. With their first launch, their AI got the telescope question wrong. Then they announced Google Gemini and they faked their video showing how good it was. And now their newest iteration it just came out that it's massively racist and it means it's not usable. It doesn't seek accuracy and that's really bad for anyone who is trying to do any type of work a tool that doesn't put accuracy. If you're doing research or image generation, that's not useful. So I think Google is out of the race. I think they have no chance at this point. They've made three major mistakes that have destroyed the reputation. Now Copilot is chat GBT. The relationship that Microsoft has with open AI is they own 49%. Anything that open AI invents automatically transfers upstream. So anything they make good's pushed into Copilot. That was the deal that they made and I think that open AI will stay dominant for at least the next year. I think with, as you mentioned, sora, the text and video thing, that was a game changer. If that's real, if they didn't fake their videos, which they said on the page, which was a dig at Google, then they're very far ahead. However, I think there's a very good chance that stability or mistral become the dominant platforms.

Eric Eden: 29:22

Interesting. Why do you think they become the dominant platforms? What's your thesis?

Jonathan Green: 29:28

So stability. Most people don't realize this, but almost all image generation is run by something called stable diffusion. Stability AIs who made that? They made it open source. So mid journey, blue Willow, leonardo they're all built on top of stable diffusion, so everyone's using their tool. They've made some really interesting decisions as far as how they price things, their open source model, and they're doing really cool stuff. The other thing is that mistral is just the best team that's making open source. About six months ago, two scientists wrote a white paper that said no open source AI will ever beat chat GB3.5. Two weeks later, mistral did and they had to withdraw their paper in shame. It's the fastest. Anyone's ever claimed something will never happen, and then it happened. It's like saying no, it'll be the four-minute mile, and then someone does the next week. And since then, dozens of AIs have surpassed ChatGP 3.5 in every category and a lot of them have beaten ChatGP 4 in one category or another. So I think there's a strong possibility that the open-source world pushes ahead, because you have everyone developing and there's the kind of idea that you could put a piece of the AI on every computer, like the way a torrenting works to transfer a file a little bit stored in all these different places. There's already a tool called Pedal doing that and it's very interesting. It's also probably how Skynet will happen if it happens. So I think there's a possibility that these dark horses and people don't realize how many people are using it. So I happen to know that people who work at some of the big companies that are developing their own AIs actually use open-source AIs for their side projects, because they don't have the same restrictions. There's a lot of weird things that people who build an AI for a company put a lot of weird things in. It's like it'll write a poem about a man but not a woman, or the main character can only be a woman, or the main character can only be a man. All that weird stuff that's not neutral. If they do enough of that, it'll push people away for their platforms. That's what Google did. They did some really weird stuff. It's like it won't draw a picture of a Viking that looks like someone from Europe. That's weird. So I think that those mistakes have a strong possibility of pushing the general public in another direction, so the companies that aren't doing that have a chance to grab market share.

Eric Eden: 31:52

It's fascinating From a business perspective, I believe, that open AI in a span of just two years has gotten to two billion in revenue. That's got to be one of the fastest ascents of mass adoption that we've seen in recent times, right.

Jonathan Green: 32:16

Yeah, but we're also seeing that they have to keep coming out with innovations and new updates and new features to stay ahead of the competition. They weren't the first one with text inside of the images. They weren't the first one with the GPTs, the character bots. They weren't the first one with the text to video. So they have to come out with these revolutions, like they have to come up with something better constantly. And they do have the attention of the market. They are a little bit ahead of everyone else. They have a massive amount of funding and it's possible they stay dominant because they just have so much wealth. But, as we've seen, all it takes is like one mistake or one misstep and it almost happened when they had the whole board and everyone quitting the company, all of those weird things happening, because they were started as a nonprofit, they were started as a charity and then now they're. You know what I mean. It's like very strange to you. Yeah, we're a nonprofit. We're also the most expensive company in the world. Which one is it? So there's a. I really don't know things happening at that level. Like it's people, all I can do is speculate, but I only know a little bit about the internal workings of these companies. I know a few people that work at each of them, but I don't know what's happening in the boardrooms and what happened, what happened in the hearts of men. So anything is possible.

Eric Eden: 33:33

And the other thing related to the business of AI that's remarkable is the news yesterday that Nvidia, who makes the processing chips that powers the logarithms to do all the things that we're talking about here, nvidia has now risen to be the fourth most valuable company, with a market cap of 1.8 trillion like a higher market cap than Google and Amazon. That's mind blowing, like, how do you think about the chips that power this? And I think that one of the reasons that they've gone up so much is because they've announced they're going to be working with Microsoft and Google and others to make chips specifically for these AI functionalities that allow this massive processing power. How do you think about the technology behind that and what Nvidia is doing?

Jonathan Green: 34:36

Oh, the phone in your pocket right now is smarter than all the computers that put a spaceship on the moon in the 1960s. So we see technology jump so fast. I believe that AIs will jump to mobile phones within two years. The need for these super powerful processing cards. With how fast our computers and phones are getting more and more technology, I think it will move in that direction, which means that, for I think that for a short time Nvidia will be dominant, but then the fact that every chip is so powerful means that it will spread around. I don't think that's going to last.

Eric Eden: 35:12

Okay, interesting Any other sort of final thoughts about marketers and business leaders using AI in today's market that you wanted to share, that we didn't cover anything else that you think people should really know, that you think is important or remarkable?

Jonathan Green: 35:34

I think that the most important thing is to use AI tools in a way that is best beneficial for you. The first thing I work with. Every single person I talk to has a slightly different use case. You don't need to copy exactly how I'm using it or how to use someone else you look up to is using it. Find a way that it will save you the most time, give you the most benefit and really use it. Always think I'm just here not to learn something cool, but to increase my efficiency, to get more time back. If you do that, you're going to get the best results.

Eric Eden: 36:03

Amazing. Thanks so much for this insight, jonathan. I appreciate it. I'm sure the listeners appreciate it. I would encourage everyone listening to share this episode with your friends. Lots of great insights here. The AI is moving very fast. If you want to know more, do go out to Amazon and get Jonathan Green's book. I will link to it in the show notes as well. It's easy to find. We appreciate you being on the show today, jonathan. Thanks for being here. Everyone, please rate, review and subscribe so we can get great guests like Jonathan on the show. Thank you, jonathan.

Jonathan Green: 36:42

Thank you for having me.