220 | The AI world is going Bananas - How to create mind-blowing visual assets for any need in minutes with Google's new Nano Banana AI image generator/editor

Is this the beginning of the end for Photoshop, stock photography, and traditional ad shoots?

In this episode of Leveraging AI,  Isar Meitis dives headfirst into the wildly capable new Gemini 2.5 Flash image generation tool, also called “Nano Banana.” 

Here’s the twist: It's not just about creating images. It's about building scalable, repeatable, hyper-custom content pipelines that save weeks of work and thousands in budget with zero design skills required.

In this session, you'll discover:

• What exactly is Gemini 2.5 Flash (Nano Banana) and how it differs from ChatGPT or Midjourney
• How to create realistic, brand-consistent images from nothing more than a sketch and a sentence
• Why image consistency, layering, and template reusability are a game-changer for marketing teams
• How to transform product shots, team headshots, and social ad campaigns in minutes
• The “dangerously easy” way anyone can now create deepfakes and why that’s a double-edged sword
• Real-world, business-relevant use cases: from eCommerce to real estate, design, ad testing, and product mockups
• How to scale image creation with workflows using tools like Weavy and upscaling solutions
• Why this is the beginning of a major disruption in content production and how you can stay ahead
• What’s still missing: resolution limits, lack of layers, and where AI tooling must evolve next

About Leveraging AI

• The Ultimate AI Course for Business People: https://multiplai.ai/ai-course/
• YouTube Full Episodes: https://www.youtube.com/@Multiplai_AI/
• Connect with Isar Meitis: https://www.linkedin.com/in/isarmeitis/
• Join our Live Sessions, AI Hangouts and newsletter: https://services.multiplai.ai/events

If you’ve enjoyed or benefited from some of the insights of this episode, leave us a five-star review on your favorite podcast platform, and let us know what you learned, found helpful, or liked most about this show!

Transcript

GMT20250901-193429_Recording_avo_1280x720: 0:00

Hello and welcome to the Leveraging AI Podcast, a podcast that shares practical, ethical ways to leverage AI to improve efficiency, grow your business, and advance your career. This is Isar Meitis, your host, and in the past week and a half the world, or at least the AI world, has gone bananas. And the reason it has gone bananas it's because Google has introduced Gemini 2.5 flash image generation tool, which they initially launched on the LM CS arena under the brand name nano banana. So everybody was going crazy with what this model can do. Now we know it's Gemini 2.5 flash image generation capability, and this model can do incredible things. Now, after doing a very short review of what it can do on this past weekend's news episode, I decided I have to dive in and actually show you different use cases that are absolutely mind blowing. So let's start with what are the capabilities of this model and why is it so exciting? And then we're gonna dive into the actual details and examples that I, some of them are mine that I'm gonna show you exactly how they're done. Some of them are just outputs of multiple people from the web that are sharing their results. mAny of them are way higher than what I know how to do, or I had the time to do in this particular case. But let's get started with capabilities. So the first and most important aspect of this model is that it's incredible at keeping consistency of everything on the image. So it can keep the consistency of the background, it can keep the consistency of object, it can keep the consistency of faces, people, modes, interactions, themes, styles, literally anything you want, you can take from an image and apply to another, merge, two images together and so on. The other thing, talking about merging things together, it is very good at multi-image fusion. I will show you multiple examples of this. You can combine different objects, you can combine objects and scenes and background. You can combine literally anything you can imagine. Now, the other thing that you can do is you can do local edits. Either adding, removing, changing, more or less anything in an existing image. This image could be generated by AI or generated by an actual image, meaning some picture you took of something, it doesn't really matter. You can make changes to it. When I say make changes, you can remove people from an image that somebody photobombed your picture or you have a picture of yourself with the alpha tower in the background, and there's a lot of other people there, but you don't want them in the image. You can have them all removed. You can, by the way, do the other way around. You can take a picture of your family and then add yourself to the image that even though you're the one that worked behind the camera, taking the picture, you can upload a picture of yourself with or without the same exact stuff that you were wearing that day, and then ask Gemini to add you to that image. The other thing that you can do is you can create a template and then use it consistently across multiple examples. So this is good for creating cards, badges, mockups of ads, et cetera. Literally any template that you want, you can use as a template and then reuse it again and again and again in multiple options and different implementations. Again, I'll show you several different examples of that. World knowledge, image reasoning is huge. There's two aspects of that. One, it is because it's a thinking multimodal AI model and not just an image generator. It understands the world, which means on one hand it understands physics and how things interact with one another. So when you ask it to create an image of something, it'll make sense in the physical world unless you ask it not to. And the other aspect of this is that it understands business and flow charts and so on. So if you are now in a room, in a conversation and you have a whiteboard and you're sketching and drawing things, you can take a picture of that. Let's say there's a flow chart diagram and you can ask it to design the business flow and show it in a graphical way. It can do that because it understands everything that you have created and you can recreate it in a more accurate way. You can also, because it's a large language model, you can continue to have a conversation with it to continue to edit and manipulate the image step by step by step, which you obviously cannot do if you are using a tool like Me Journey or just a standard image generator. Now, in theory, you could do that because you can take a screenshot of the image, enter it as a starting point, and then try to edit from there. But it's clunky and it's definitely not as coherent and cohesive as it is doing it in nano banana. From this perspective, it's very similar to using the new-ish image generation model from ChatGPT that was released earlier this year and again took everybody by a storm. This model is also really good at generating text on images, but because it's also very good in consistency, it can apply your logo or anybody else's logo for that matter, to either a banner that you're creating or put it on top of whatever object that is in the image because it understands, again, physics. So if there's a cup and you put your logo on the cup, it will render it as if it's in a circular way, wrapping around the cup, et cetera, et cetera. T-shirts, glasses, anything you wanna put this on will be generated in the correct graphics of the logo, but also in the correct 3D application where you want to apply this. Now, as I mentioned, you can also transfer style. So you can take a style from an image, a style from a type of cloth, a style from literally anything and apply to anything else. It is very good at maintaining pose and applying wardrobe changes. So if you want to see how something fits on you, you can take a picture of you, take a picture of that piece of clothing or accessory and you can apply it to you and see how exactly it is going to look like. I will show you multiple examples of this as well, and you can play around with this with different things. You can re colorize and improve any kind of image. So let's say you have a beautiful neighborhood photo and you want to change the color of the buildings from blue to red. You can do that, but more importantly, you can take old photos that are either black and white and grainy, or not even black or white, but just are in bad shape and use nano bananas to make it significantly better. There are multiple examples online right now for that, and it's absolutely mind blowing. It is on one hand, not as fast as some of the really fast image generation models, like the faster mode in mid journey, but it is significantly faster than creating images in chat pt, which is the other option we have if you want to have a conversation and have a real world understanding and include that as part of a bigger process. What you do need to be aware of is that when you're creating images with. Nano bananas. It will put two kind of watermarks on the image. One is a Gemini logo on the bottom right corner, and the other one is a synth id, invisible watermark, which is basically a hidden pattern into the actual pixels that Google tools know how identify as tools that were created by ai. I don't necessarily agree with the visual watermark. I think it's kind of stupid, especially that you can crop it out in about two seconds. But I love the fact that they finally, including synth ID in the image generation aspect of it. I really hope that the world will come to an agreement on a standard. I don't care if it's synth ID or anything else of image and video generation tools that generate stuff with AI that you'll be able to, without any effort, hopefully with just a plugin on your computer to know every image on every video, whether they're generated with AI or not. So I really hope we're gonna go in that direction. Now, let's start looking at multiple examples and use cases, and I'm going to share my screen and I'm going to explain everything that is on the screen because I know you're right now either walking your dog or driving a car or jogging on a treadmill or whatever it is that you're doing right now, listening to this podcast. So I will explain everything that is on the screen, but because this particular episode is extremely visual, I suggest that afterwards when you have time, click on the link in the show notes of this episode, go to our YouTube channel and watch the YouTube segments at least jump to the stuff that you're really interested in so you don't have to watch the whole thing again. That doesn't make any sense, but at least you can jump to the parts that you wanna see exactly what I was demonstrating. Okay, so let me share my screen and let's dive into a bunch of examples.

GMT20250901-210806_Recording_avo_1280x720: 7:50

So let's start by looking where to get to this new magical model that you can use with inside of Gemini. So first of all, you just log into your Gemini account and you can see I have 2.5 pro up here, in the model selector. I actually don't need that. All you have to do is go to where your prompt is. There's a section for tools. It has all the previous options that were there, like deep research, create videos with video canvas and guided learning. But you can see there's a new option called Create Images with a little banana image next to it. You click on that. And then you can start creating new images. To be fair, you don't even have to do that if you ask it to create an image or if you drag an image in there for editing. We're gonna talk about that later on. It will know what it needs to do, and it will know how to run this model either way. But if you wanna be safe, you can choose it from here. The other option to get to it is through google AI Studio. So if you are in ai, studio.Google.ai, there's all these amazing options like streaming live and so on. But if you go to the regular chat, there's a button here that says Try Nano banana. And you click on that and now you can literally write your prompt, edit your images, upload stuff, whatever you want do, and run it from here. The biggest difference is if you're running it in Google AI Studio, you are paying for tokens. So you're paying a really small amount of money every time you run any command in here in AI Studio in general, including creating and editing images. And it is included with your license if you're running it in your regular Gemini account. So my suggestion to you is start in the regular Gemini account and if you run out of credits and you need more, then go to the air studio and just keep on running it over there and over there you can run it forever because you're paying for every token.

GMT20250901-193429_Recording_avo_1280x720: 9:14

The first few examples I wanna show you come from Google themselves. It's showing different implementation that they have shared with the world. The reason it's cool to look at is because first of all, it's Google themselves. They know exactly what the model can do. But the other thing is it has a lot of explanations of how to actually use the models effectively. So the first one is from their developers.google blog.com, and we will share the results to that. And this article is called How to Prompt Gemini 2.5 slash Image Generation for the Best Results. And if I scroll down here. You'll see. How does this look like? So you see a beautiful image of an all Chinese guy holding a piece of pottery that he has created. The image looks completely photorealistic. He's in his studio. The background looks completely real. The lighting is perfect, and they're showing you a few things. They're showing you what is the prompt, and they're showing you what is the template of the prompt. So I will give you an example about this one. The prompt itself is a photorealistic closeup portrait of an elderly Japanese ceramicist with deep sun edged wrinkles and a warm knowing smile. He's carefully inspecting a freshly glazed T-ball.\ The setting is his rustic sun drench workshop. The scene is illuminated by soft golden hour light streaming through the window, highlighting the fine texture of the clay captured with an 85 millimeter portrait lens, resulting in a soft, blurred background in parenthesis, bouquet. The overall mood is serene and masterful. Vertical portrait orientation. And again, the overall image is perfectly coherent with a very long and highly detailed prompt, and they're giving it as a template. So the template is a photo realistic. And then what is the shot type of, what is the subject next action or expression. Next set in whatever environment the scene is illuminated by liking, description, creating a mood. Atmosphere capturing with a what type of lens details. And you can skip that if you don't know that. You can just explain what kind of look you want for the image emphasizing, and then key texture and details. The image should be in whatever aspect ratio you want, so you can apply this to literally anything you need. So this is just one example. A very different example is there's a cool red panda sticker that looks perfect as a sticker. It's 2D. It's very, very different than the previous image. And here the example prompt is a kawai style sticker, a happy red panda, wearing a tiny bamboo hat. It is munching on a green bamboo leaf. The design feature, bold clean outlines, simple cell shading and vibrant color palette. The background must be white. And it looks like a perfect sticker that is aligned exactly with that. And again, they have the template, which I'm not going to read on how to create stickers, and then they give even more and more examples on how to create logos and how to create a perfect image of a product photo shoot. In this particular case, it's a black mug on a granite style kitchen countertop with fumes coming out of it. And then they have a minimalistic view of a white background with a delicate red maple leaf in the corner. And again, the example prompt is a minimalist composition featuring a single delicate red maple leaf positioned in the bottom right of the frame, et cetera, et cetera. So very different from all the other ones. Another great example is a very dark noir art style comic. Panel out of a comic sketch and it is perfect. It has this guy in a long, dark coat with a hat. It looks very shady in like an alley. There's a place in the back that is called the Black cat. It is raining, but it's all in like these black and white sketches. It looks absolutely incredible. And again, same kind of thing. They're giving you the exact prompt and the exact template to create the same kind of thing. Then they're showing you that you can edit images. So the first example they give is there's an image of a cat on the left, and the prompt is very, very simple. Using the provided image of my cat, please add a small knitted wizard hat on its head. Make it look like it's sitting comfortably and matches the soft lighting of the photo. And the second image looks exactly like the first image, only the cat has a little needed. Wizard hat on its head. His ears are seeking out and it's really, really cute. But everything else in the image stays exactly the same. The next example is they're showing a beautiful contemporary living room with a really big blue sofa. And in this particular case, you're using the prompt using the provided image of a living room change only the blue sofa to be a vintage brown leather Chesterfield sofa. Keep the rest of the room, including the pillows on the sofa and the lighting unchanged and it works. Now it's. Again, absolutely amazing. Everything in the room stays exactly the same. There are multiple details in this room. There are lights in the ceiling. There are shelves on the right with multiple Tcho keys on them in this contemporary room, and nothing else changes, including the view from the window other than the solf itself. This is nothing short of magical, and they're showing multiple examples. Another really interesting one that I actually did an experiment of my own, and actually took it a little further, is what I told you before, is the ability to dress somebody. So they took two inputs. They took an input of a dress, just a flat image of the dress without anybody. It's just the dress itself and an image of a model that is standing in a studio wearing black pants and a white sweater. And here the prompt is create a professional e-commerce fashion photo. Take the blue floral dress from the first image Let the woman from the second image wear it, generate a realistic, full body shot of the woman wearing the dress with a lighting and a shadow adjusted to match the outdoor environment, and the image is absolutely perfect. You can see the same Molly, it's the same face, same hairstyle, a different pose. It looks as if she's walking quietly in a little alley in Europe. The lighting is perfect. It's absolutely amazing. And she's wearing the exact dress from the flat image, so you can use this either if you're selling products or if you want to try on products, you can do that as well. Now I wanna show you my little attempt on the same kind of thing. And I took it to another extreme. I thought about, okay, if I wanna sell a dress and I wanna do a cool campaign with it, how about dressing famous women's statues with that dress? So I have two pictures here. One is a picture of the statue of the woman in front of the ferry building in San Francisco. And I have a picture of a dress and the picture of that statue of the woman in San Francisco. And literally all I wrote is dress the statue with the dress in the other picture really, really short. And I got the exact output that I wanted. You can see the statue, the ferry building stays the same. The people walking around and sitting next to the statue is the same. Everything is the same, but the statue is wearing the dress. Then I wanted to get fancy. I uploaded a picture of the Statue of Liberty, which is kind of strange because it's already dressed in a gown or a dress or a toga. I'm not sure exactly what she's wearing. But I asked the same question. I uploaded the dress again. I said, make the Statue of Liberty. Wear the attached dress instead of its current cloth. What it did is actually really, really cool. It wrapped the Statue of Liberty with the same fabric and kinda like applied it to the current thing the Statue of Liberty is wearing. This is really, really clever because I wasn't exactly sure how this is gonna work out. And it actually worked out very interestingly. So this is just my example of that. So let's look at a different set of Google examples. So in this particular case, they're showing different sets of double photos and how they can be edited in order to apply new aspects or new anything to the same images. They're all really, really stunning. So let's look at them one by one. The first one has a input image of a girl with her face really dirty in a spacesuit, and the second image is a closer closeup of her without the helmet. So just taking the helmet off. The next one is taking a side shot of a redhead girl in a truck, driving in an open terrain space somewhere, and then it applies different backgrounds to the same picture, so snowy mountains, et cetera, whatever you want. The girl stays the same, her gaze, the lighting, everything stays the same. The scenery outside changes. Another example is that they have. Is taking two little funny little cartoonish characters and then placing them in a recording studio and they're actually operating the controls as they're recording somebody else. One of them have headphones on and so on, and they have the prompts to each and every one of those. I'm just sharing with you, you what you can do. And there's really amazing images of different kinds of photos. They have a photo of a girl on the left and then multiple images of her in Polaroid wearing different things in different hail styles on the right. Again, absolutely accurate and amazing. And then an underwater scene and. Literally anything you can imagine. Here's another example of merging two images, so a professional photography image of a woman and then a guy, and then the joint photo. They are hugging and looking very tenderly at each other in a street in the snow with bouquet lighting in the background. It's a perfect image, but it's exactly the same people from the two other images. So these are just different examples of what you can do with this. The next one is really, really cool. They have a picture of a young guy, kind like looking towards the camera, and then they have different professions for the same guy all created just with proms. So in the first one, he's wearing doctor scrubs and he has stethoscope on his neck. In the second one, he's a teacher in a classroom. In the third one, he's working on a clay statue, and in the fourth one, he's preparing dough in a bakery. All of them look exactly like the guy only in completely different settings with completely different poses, but it is very clearly the same guy again, using consistency. So these are these kind of examples. Then they have an example of how to turn a dog that is an actual real dog into a game character in a game style that looks like Mario. Then they're showing another cartoon character. In this particular case, it's a little dinosaur that looks like it's made out of plastic, and then they're dressing it up in multiple different kind of uniforms as a superhero, as a pirate, as a cheek and so on. Absolutely incredible. Then they're showing how you can combine different styles and aspects of different photos. So they're taking a photo of a woman wearing an orange piece of clothing and a photo of a piece of glass floating in midair, and they're showing the woman the output is the woman floating in this bubble of glass in midair, and it's very impressive. They took a picture of whales jumping in the water and a picture of a snowy mountain with the lighting on top in the sunset and combine the two where you can see the mountain in the background as the whales are jumping in the ocean. They took a picture of a fork and a picture of spaghetti and created a fork made out of spaghetti just with a single prompt and so on and so forth. Really, really amazing. And then they're showing how you can clean images and make changes to images. So there's an image of an image of a building with very distinct color patterns on it. In several different segments of the pattern are in blue, and the prompt is literally change all the blue to green and you get the same exact image, with the same model standing in the middle, yet wearing her yellow cloth just with a green background instead of blue. They have an image of a guy walking towards a truck, probably somewhere in the Midwest. There's a crazy tornado in the background. And in the second image, it's a beautiful evening with no bad weather whatsoever. But everything else is exactly the same. They have an image of a house where the next four photos are the same exact house, but in four different seasons. And you can very clearly see how the house looks like when it's covered in snow, how it is when the leaves are changing, and so on and so forth. The next example that they're giving is how to push design and use existing ideas in order to apply them to something else. So I have a picture of a butterfly, and they're creating a picture of a model with a butterfly dress that is completely inspired by the pattern on the butterfly's wings. They have a picture of a living room that is very nicely done, and then they're adding a color palette, literally just five different colors, and they're applying it to the cushions on the sofa to give some colors to the room. Again, everything else stays the same. Then they're taking a face of a person and applying it as a cartoonish style on the side of a cereal box and so on and so forth. Taking something from one image and applying it to a different image in multiple aspects. Let's look at a few more examples of things that I did. So I found this thing on x.com. And I hope this is the original I was really trying to find, because these examples now are everywhere. I was trying to find the original creator so I can give them credit. So in this particular case, it's Zoe, I hope I'm pronouncing his name correctly. And his handle in X is Zoe. Z. Z Zo. So Zo, Z, Z. And what he did is he took a sketch of a figure. The way you start, when you start creating a painting of something, especially when you're trying to create an image of a superhero or something like that. sO just the outlines of the body In a very specific pose, and he took an image of a Japanese model and he asked to apply the model to the pose in the sketched character, and he did a perfect job. It is absolutely mind blowing because the girl is the same exact girl. She's wearing the same exact stuff she's wearing in the image and she's standing or kind of like crouching in the same exact pose as the sketched figure is sitting at. The other really incredible thing is that the, the original image of the model, you cannot see the bottom of her feet, so it had to guess what she's wearing. Give it the same kind of style of the other stuff that she's wearing. And it's actually done a very good job of doing this. And I was interested was that, how many attempts it took to create that? So what I did is I took the image on the left, basically the sketched outline of the superhero pose and together with the image of the girl on the top right. And I asked Gemini and I dropped it into Gemini and I wrote the following prompt, remove the image of the girl from the top right corner and then. We already learned that that's doable. I can remove stuff from the image. So I got a image of just the sketched outline of the superhero pose, and then I uploaded a selfie image of more or less just my head and shoulders. That's it. And I said, now create a photo realistic studio photo shoot of the guy in the image, in the same exact pose as the action figure sketch white background. And I got a perfect pose of me. And again, you had to guess everything that I'm wearing because all you see is my shoulders with an off white sweater. So it applied the sweater as, as if I'm wearing it tied around my neck. And it gave me very stylish clothing that I will never actually wear because I'm not a stylish kind of guy. I wear t-shirts and shorts most of the time, but the image is perfect and it's totally me, like the face is me. The pose could have been me if would've been that flexible. Uh, but it's very, very impressive. And then I said, okay, now dress him up as Spider-Man, but without the mask. So now I have the same exact image because it's a superhero kind of pose with me instead of wearing the clothes I was wearing before or that I'd invented for me dressed as Spider-Man, but it's still my face, still the same pose. And then I said, okay, place him on top of a skyscraper at night in a medium shot. And it did that as well. It Even placed the light from the moon in a very specific angle that drops shadow from the pose that I'm standing on the rooftop. And then you can go on and on and on and build an entire superhero story from that one original shot. So it's one line of text and these magical outputs just come out. Let's look at another example from X. This creator from X created a sketch on a piece of paper and took a picture of it. The sketch is of a model standing and stretching up. It is the kind of sketch you would see from clothing designers create when they start thinking and sketching the ideas for their next clothes. So it's this girl who's stretching up and you can see that she's wearing Calvin Klein underwear and you can see what kind of shirt she's wearing, but you can't really see the girl. And with one prompt, he created an image of an actual model in the same exact pose, wearing the exact clothe that he was imagining in his sketch. This is absolutely insane. It's going from an idea to paper, old school way, going from that to a photo shoot, fully realistic image of the model in the same exact scenario. This is absolutely mind blowing. Now let's look at combining images. I was trying something very cool here. So what I did is I looked online for standard images of different ingredients. I found most of them on websites of different grocery shops. So I have an image of a avocado sliced. I have an image of corn. I have an image of sour cream. I have an image of chicken thighs wrapped in a frozen chicken thigh box. I have cilantro, I have tomatoes, and I have tortillas. Each and every one of them is a standalone image that I got from the web with no background of each and every one of these ingredients. And then all I wrote is create a promotional social media image of a dish I can make from all these ingredients. And it created this beautiful, incredible image of a modern kitchen that is blurry in the background. There's a gray countertop on top of that, there is a wooden cutting board. And on top of that, there is this perfect deliciously looking chicken wrap. And the ad says, dinner solved, exclamation point, delicious chicken wraps made with fresh ingredients. It is absolutely amazing. Then I ask it to change the aspect ratio of the image, and it failed. And I must admit, I tried changing the aspect ratio of images on Gemini multiple times, and it fails every single time. Now it might be me personally, I've seen a lot of other people do it. When I asked Gemini, can you do that? It said, yes, absolutely. But every time I tried it, it failed. I tried it in multiple scenarios with multiple images. It never worked for me. I assumed this is something they will solve soon. Then I was trying to create an even funnier ad. So I was playing around with different Yeti images in the end, what I got to is there's a Yeti in a bus with multiple other people next to the Yeti. Everybody's eating chicken wraps. The Yeti is all dirty with the sour cream dripping all over his fur. And the ad is saying unapologetically delicious, and on the bottom, in smaller front it says chicken wrap. So tasty. You do not care. Getting messy in public, it is a great funny ad that I went from ingredients to an ad I can probably use in about six to seven minutes. This would've taken the old way, months, or at least weeks to create between coming up with the ideas, doing the shoot of the actual wrap, et cetera, et cetera, et cetera. I can create now a whole series of ads based on the same style, based on the same comedy concepts, and just keep on creating them right here in the same chat. Another example of using multiple images to achieve a goal is I took a selfie of myself with a very colorful background, really messy, and I took a screenshot outta my brand guidelines of a transition from dark blue to light blue that I use in different aspects of my products and my websites and so on. And I took a screenshot of my logo and I asked the AI to create a professional headshot of this person. Keep the image aspect ratio of the person as is, change the background to the color transition in the same color of blue as your attachment on the top left corner, add a small version of the logo that is attached. The person should be wearing a white buttoned up shirt with a blazer and no tie. And other than the aspect ratio aspect of it, which again we already covered it created a beautiful, amazing professional photo shoot of myself with the transitioning color behind me with the logo. Perfectly done on the top left corner. This is amazing. This is absolutely incredible. What I did then is I cropped just the section of myself, because again, I tried multiple times to get the aspect ratio right. And it didn't, so I just cropped the area where I appear and then I went to the next step. I decided to use it as a template. So I uploaded the cropped image to another new chat and I uploaded an image of my son just before getting a haircut. So his hair is really long and he is sitting in a barber shop looking at the camera, kind of like a before and after image. And what I asked the AI to do is I said, use the professional headshot of the bald guy as a template, apply the same exact background and similar clothing style to the kid. And now I have a picture of my son wearing a buttoned up shirt in a blazer with the same long, funny hair that he had before he took the haircut, but with the same exact transitioning blue in the background. Why is that important for business? You can create professional looking headshots of everybody in your company, in the same exact style, with the same exact background, with your logo in a few minutes. Just by having images of your employees, any employee, doesn't matter what the image is, and you can recreate it without having to bring a professional photographer. Set out the background, have cameras, edit that, let people select for multiple options. All of that can be done on a computer in just a few minutes. Another example, this one is from a user called Choi on X. Again, his handel is ais ai and he took a surveillance camera image of a woman that is looking to the left, and all he asked is in the prompt to have her look straight into the camera, and she does. Again, everything else stays the same. The corridor, the lighting, the clothing that you, the woman is wearing, everything is the same. Why am I showing you this? I'm showing this to see how easy it becomes to create deep fakes. You can literally create any scenario you want with any person's face in any surrounding, with just a simple prompt and images of the person and the surroundings. Now I want to talk to you about the next level that Google already provided us. So in addition to all these incredible capabilities, I'm gonna show you some more advanced implementations. Later on, I wanna show you another amazing tool that they created. So Google created a solution in Google AI Studio that allows you to create apps using this new Gemini capability. So you can use existing apps that they're already sharing or create your own apps that you wanna create just by explaining them and prompting exactly what the app will do. So they're shared several different examples of what these apps do, and I'm gonna show you some of them. The first one allows you to edit images. It's called AI powered photo editing. Simplified retouch photo, apply creative filters or make professional adjustments using a simple text prompts, no complex tools needed. So this is a tool that is called Pix Shop that they've built with their own app building tools so you can create your own tools that do different things, that allows you to edit images in a professional way. The next version is actually really, really cool. It's called Past Forward. It allows you to upload an image of you or anybody else and it creates you in different decades and it creates this kind of like as Polaroid images right at the bottom of the image. You will see which decade it's from and it's absolutely cool. Like my first top left image is from the 1950s and it has this SIA color and I'm wearing these jacket and tie old school, including the images in the background. There's like an old sofa and wood plank panels on the wall it looks like from the fifties, but it definitely looks like me. I actually really like my seventies version of me. It has really long lush hair, kind of like Miami Vice style. And I'm wearing a patterned shirt with a really large collar and I look really, really cool. The original image comes from NASDAQ and it has several different company logos in the background. So what it did, which is really, really incredible, it understood where I am. It understood what's in the background and in the seventies image it has companies from that era. So it has IBM, and Coca-Cola and Intel versus the companies that are there right now in the original image that I uploaded and so on and so forth. It's really cool. It's really fun to play with and the way it works, you can actually click and see the prompting that it's doing in order to create all these images. So. It's just very cool example. The next one is a lot more practical. It's called Home Canvas. And what they're showing in home canvas is the ability to take a image of a room, an actual room that you have in your office, in your home, or wherever else, and take a product image. In this particular case, it is showing a light stand, light source. And you can literally drag it and tell it where to place it. I'm gonna place it on the little nightstand next to the bed and it's gonna think about it and it's gonna create that exact light on the bedstand. And you'll see in a second exactly what it does. So it's thinking about it, it's telling you what it's doing, and now it's placed it there and you can actually even see the light. So it's like a yellowish light, and you can see exactly how it's gonna look like, how it reflects on the bed, and the bed stand, and the floor and the shadows. Everything is realistic, literally by just dragging and dropping the thing in. Now, I must admit, I try this on multiple different scenarios with multiple different products, and it works about 25% of the time and not a hundred percent of the time, but every time that it does work, it's absolutely magical. The other cool thing is that it knows how to change the orientation of the object to be relevant to where you placed it in the room. So again, it understands the 3D nature of things. It can anticipate how they're gonna look like from different directions, and it can apply that in the new scenario. More on that in the end. Don't leave before we get to the end, because if you're selling any kind of products. Or you're doing promotion for any kind of products, you want to see how you can do this at scale for your products. And we're gonna show that as the grand finale of this episode. THe next few examples are people taking it to a completely different level. The first one that really blew my mind comes from Alex Rasco. His handle is at at x. And what he created is he created a video of famous pictures like Van Gogh, self portrait, and the Mona Lisa and so on. And each and every one of them on his turn is turning into an actual person that is walking in Central Park in New York, and they're meeting each other and becoming very friendly and taking selfies. And it's nothing short of magic, both in means of creativity, but also the actual implementation of this. And then he adds more and more photos of more and more famous pictures, and they're turning. Into actual humans and they're in Central Park and they're meeting all the other people and they keep on engaging with one another. Again, brilliant, brilliant creativity. Absolutely stunning. And it's showing you the ability to take the fact that it knows how to keep consistent characters and turn this into an actual video with a storyline with really unique, amazing capabilities. This is obviously not somebody who's doing this the first time, but the fact that it's possible is absolutely mind blowing. Let's look at another example from a different creator. This creator is called Framer with the handle at zero X framer, and he created this unbelievable anime, multiple scenes, about 30 to 45. Second video of a chase of this younger kid driving his souped up, upgraded motorcycle with the police, chasing him around, and the police car gets into an accident on the bridge, almost falling off the bridge. And it looks like a professionally done manga video. It is incredible. Like I'm sure my kids, if there was a full episode, would watch this thing. How has this achieved the same exact thing? You can create the same exact character with complete consistency, the same scene, the same street, the same police car, the same police officer, the same bridge in multiple angles using nano banana. Then drop it into your favorite video generation tool and then create this. In this particular case, he created it with C Dance 1.0, which is a video generation tool that is getting a lot of love online right now. But he's saying he created this entire video in less than two hours, which is mind blowing again. The old way would've taken a team of multiple people several days to create this short film. The next example that I wanna show you that goes into how crazy the deep fakes of this is. This comes from a guy's blog in India and he's showing how he's adding different people, famous people into his selfies. In this particular one, it's a selfie of him together with Brad Pitt. It is completely realistic. There is no way of telling that it's not actually Brad Pitt next to him, and what you can see he also have a selfie with a known Indian celebrity, which I don't know, but the original photo is just him, obviously in the selfie. The background is the same background. The lighting is the same lighting, and you can add whoever you want next to you. This is really, really scary, but this is reality. This is what we have right now in the hands of anybody who has access to Gemini, which is anybody who has access to the internet. And again, to make this crazier, you can drop this into any video generation tool and have them have a conversation in front of the camera as if someone is filming them or it's like a selfie film of them talking. This is totally doable in a matter of minutes to anybody you want in the world. The next one is a really simple and yet really cool and helpful use case that I actually used this past week for a presentation, so one of the things I was teaching in one of the workshops last week is how to do workflow automation together with ai. And I showed examples in Make, and I showed examples in N8N. Now, when I create my presentations, all my presentation slides have a black background. This is just a style that I like and this shows up in all the different presentations that I give. The problem is that Make has this kind of off-white background. So you have one option is to actually screenshot the thing out of make, drop it into your slide, but then you have a ugly off-white rectangle in the middle of your slide. That's the top thing that you're seeing on the screen right now. The other option is to take it to a background remover. Some of the background removers today are actually not bad. I think the best one right now is built into Canva. So I used, for the second option, I used the Canva background remover. There are two problems with that. One is that the text is almost black. It's like a very dark gray. So when you remove the background, the text that is again, very dark gray, almost doesn't show up against the black background of my slides. The other problem is for most of the different nodes in clay, the inner image, basically that's telling you what kind of node it is, is in the same color as the background. So when you remove the background, it turns black and it's sometimes very hard to see what it is. So then I just took the screenshot that I took originally that has the white background and the dark text, and I dropped it into Gemini and I literally wrote a prompt that says, please change the background off this screenshot to black and change the font. Off the text to white. That's it. And I got exactly the outcome that I wanted. The image stayed exactly the same, the text stayed exactly the same, and I could now paste it into my slides. So this is a very practical day-to-day kind of use case. That was very hard to do before unless you have access to Photoshop and you're a Photoshop whiz. And then still, this is a faster way to do this. The next example is an exterior design exercise that I've done. So I took an image of this really old falling apart balcony, and you can see the yard and you can see, kind of like, see it's a weird angle, but you can kind of see the doors, and the house. You can see the light next to the door, but it's an old porch that is falling apart. Then you can see old wooden planks on the floor of the porch, and they're all. And they're all very, very old. The color is mostly not there. The, some of the pieces of the fence are missing or not connected well. And all I did is I wrote the following prompt, create an image suggesting how an upgraded version of this porch can look like. Do not add furniture, but make it look new, inviting and contemporary style. And then it created this beautiful version of the same exact porch. Now what you can see again, the background stays the same. The trees stay the same, the pool in the backyard stays the same. Everything stays the same. Just looks like a brand new porch that if I was a designer or a general contractor, I could have probably sold this to the owner of that porch. But then I did something else. I asked it to change the angle so I can see more of the porch, and it was able to change the angle of the camera. Now think about how crazy this is. It literally has to invent things that are not in the image, but understand exactly how it's gonna look like. I'm gonna mention a few very specific details that are very interesting. It took the light that in the original image shows up on one side of the doors. It cannot see the rest of the doors, so it added a, another one on the right side of the doors because it makes sense. It saw that in the middle of the wall in a specific distance, there is a outdoor power plug with a cover on it. It added another one on this, on the opposite side of the other door. It also, because of the new angle and also, which is absolutely incredible because it created glass panels in the railing. Because we now have a new angle, it created reflections on the glass of how the rest of the porch looks like. This is really, really next level because you can see the yard through the reflection. This looks completely realistic, and yet it made a lot of that up. Which takes us to the incredible ability of this tool to change angles on things that it's seeing and create new variations from different aspects of the same scene. Now, this could be this porch. This could be a product that you're selling. So if you're selling a product and you wanna have multiple shots of the same product from multiple angles, you don't need a professional photo shoot, you can upload one or two versions of the product. It'll remove the background on its own, and then you can generate any angles of the product as you wish. But to put this on steroids, I will go to an incredible example from my good friend, an incredible visual AI creator, Rory Flynn. So Rory took this new tool and used it in Weavy. Those of you who don't know Weavy, Weavy is an incredible, incredibly powerful visual generation tool that allows you to create images, combine them with prompts, combine them with different templates, and create an entire workflow out of that template. So what he did here is he took an image of a Mercedes racing car, but that's it. That's the only visual input. And then he added different prompts and different steps that basically tell it to create different angles of the image. So what you can see here is that it's taking the array of different angles. You can see an array here and you can see that it's suggesting different angles. It takes all of that in. It takes the input of the car, and it is creating all these angles. So now, instead of just a diagonal 45 degrees from the front view of the car, we have a rear view of the car. We have a side view of the car, and we have a 45 degree view of the car as well. Now to make this even crazier, one of the biggest still existing disadvantages of nano banana, also known as Gemini 2.5 flash image generation tool, is that it, the quality of its images is still low resolution. So if we click on this image and make it bigger. You can see that the angle is perfect. The details of the car is absolutely perfect, but the quality of the image is a little grainy. However, the next step, because this is, allows you to create multiple steps, the next step is using, recraft upscale. So the next image, which is the same exact image only upscaled, I can zoom in and zoom in and zoom in, and zoom in and zoom in, and it still stays perfectly clear. So using a tool like Wvi or just using any external upscale or allows you to solve the scaling problem, but then he added something else for each and every one of the different angles of the results, he added additional steps that provide different levels of zoom and different, more ad like angles of the car. So now you can see multiple images of the car in different angles with closeups from different directions that. Each and every one of them looks like out of a professional shoot off the car for ads, for whatever you want. And all of this is created with now a single template. What does that mean? A single template? It means that if you drop a different kind of image into the same thing in this particular case, an old red BMW, you get all these assets, all these outputs, including the high-res images, including all the zoomed in ones, including every one of the things that we've seen before with the Mercedes racing car for this new car. And you can do this as many times as you want with other different images, like this semi-transparent view of this futuristic vehicle. Same exact result. Why am I showing you all of this if you are manufacturing or promoting and selling any kind of physical goods, you can use a tool like VY together, combined with Gemini 2.5 flash combined with an upscale to create any asset you want from any angle with zero time. Because once you create the template, you can drop any new image in there and create all these assets. So what does all of this mean? Well, some of these capabilities existed when the new Chate image generation model came out when everybody was starting to create manga images of themselves, Ghibli style kind of images, and so on and so forth, myself included. And that was the first time we really had a multimodal AI model that you can have a conversation with that understands what you're asking, that understands context, that can have a step-by-step workflow with AI images. And it took that to a completely different level in realism, completely different level in its ability to change angles and directions, keep consistency of characters and objects and backgrounds and styles and so on. And it is right now just an incredible tool to use for more or less any image generation and or image editing that you want. So I will ask a few questions already asked in the previous news episode, but I think if you haven't listened to that, and even if you have it will be a good reminder. So is this the death of Photoshop in marketing or in any other aspect, right? Is this the death of other editing tools? Is this the death of TV and movie studios, et cetera, et cetera. Again, I showed you multiple examples on what this tool can do, and this is just the first two weeks that it's been out. Think about what happens when it gets upgraded. So my answer is not yet, and I think the focus here is on yet, but it's a big step in the direction of putting all the things that I mentioned earlier of very serious risk. So what are the two main issues that I see for this tool right now? What I believe and hope, at least from my personal perspective, would be the next steps. The first one is resolution. This tool and these tools in general generate very low resolution results. I don't think it's a technological problem. I think it's a compute limitation problem. I think if they would allow everybody to create images that are eight K in resolution, then they will run out of GPUs very, very quickly because just processing that just takes significantly more compute, which means once they have more bandwidth and or improve the algorithms that do this, you'll be able to generate more or less any resolution you want. Now, as I mentioned, there's already upscales that exist today that can solve this problem, at least to an extent, which means you can take the output of nano bananas, put it in an upscale, and then get a much higher resolution image of the same image that you had before. The next aspect, which I think is the biggest thing, comes back to tooling. And as we've seen with the recent six months of releases from all the large labs, a big part of what they released was new tooling, new ways to engage with the models, whether it's canvas for text and code editing, et cetera. It's Projects in ChatGPT where you can create these mini universes of context, et cetera, et cetera. Multiple tools that we can use to just have a better interaction with the ai. So I think what it's missing on the visual side is layers, right? Think about even very basic. Even very basic image editing tools, whether you go to Canva or even PowerPoint has layers. You can move things in front or behind different other objects, and then you can grab a layer and move it around and resize it and rotate it and tilt it in some more advanced tools. And you can do all of that, in all these tools. I think once Cha GPT and Gemini adds layer control to their tools, and I'll be really surprised if it's not happening sometime soon. That is game over to most professional to most image editing platforms out there because it's just better and it's free or almost free because it is included with your subscription. If you get to the limits of what you're doing, you can always use it through the API and pay for tokens, which is still gonna be significantly cheaper than any other way of doing this. So bottom line, nano bananas, also known as GPT 2.5. Flash is a very powerful, capable tool for image generation, image editing, and any kind of creative work that you want to do. What I recommend is, as I recommend for everything, instead of just playing with it just for fun, think about actual use cases. These use cases could be personal use cases. You wanna renovate your house, you wanna see how a sofa will look in your living room. You wanna create a really cool sticker for your kids, for their scouts. You want to create a night story for your kids, whatever the case may be, you can do that with this model. This obviously can apply to business needs that we mentioned a lot in this episode, so find an actual use case, business or personal, and then just go experiment. Go get some inspiration online. Again, there's no millions of examples of how people have used it, and just try to do your own thing, just like I showed you today, and then take it from there and decide how that applies and how it can help you in your day to day life. I hope you found this episode helpful. As I mentioned, you can go check out just the components or everything on our YouTube channel. There's gonna be a link in the show notes. I'm also going to put a link. I'm also gonna put the credits in the show notes to every single person that we have used their work in this episode. So I want to thank them even though I don't know them. And as I mentioned in the previous episode, we just relaunched the self-paced version of the AI Business Transformation course. We're finalizing the edits by mid-September, so if you join the course, you will get the latest and greatest of the course that I am still teaching. That started in August and is finishing in September. You'll be able to take it as a self pace lesson by lesson at your own time. I don't think it's as good as taking the course with me. I think having a human instructor and other people to engage with during the course is better, but the next live course might happen only in 2026 because I'm booked solid with company specific training between now and then. So if you wanna take the course, that is a great way to do this right now, and then later on, if you want to join the updated live version sometime in 2026, you are more than welcome to do that. That's it for today. Have an amazing rest of your day. Keep experimenting, share with the world what you find so we can all learn and have a great rest of your day.