Prompt Engineering for AI Content Creation

1. INTRODUCTION: Hi, everyone. Thanks for joining this class. Have you ever asked an AI to create an image or video and got something totally different from what you wanted? That's what we're going to fix. In this class, you learn how to write better prompt. The instructions you give to AI tool, you will learn how small changes in your words can make a huge difference in your result. By the end of this class, you learn exactly what to type to get what you actually want in your result. Let's mean, no more guessing and no more endless retries. So let's get started, and let's turn your ideas into reality. 2. Basic Level: Be Specific: Hi, everyone, and welcome to this course. Now, as you can see, we have this website, Hey, you can generate unlimited images for completely free. Labs dot Google slash Ax SlashTolsS WIS. Let's do prompt Engineering. Let's start from the base. First, you have to be specific about what you want to generate. Example, if you want to generate a card, don't just type a card. As you can see, we have this example where I just type a car and it could be any card and it could be in any location. We have the second example. Now, if you have a specific car model, you can type that if you want to place a card onto a specific location. Now we have this example. This is a good prompt engineering exam. Vintage 90 60 red convertible driving along a coastal highway. Now, if you notice we put this detail in the prom, that is why we are getting the specific result. The more details you add, the better the result we have another example. If we type a building, you will get random building. We have second example. But if you have a specific design and structure about the building and the scene, you will get your specific result. In this example, we have a futuristic glassy scraper reflecting the sunset in downtown in New York. This is my second example. If you want to generate specific food, don't type food, you will get random. This is my second generation. And if you type specific thing about the food, we have this prompt, and as you can see, we have our specific result. This is my second generation of this prompt. If you want to generate some monster, don't just type a monster. You will get random results like we have this. But if you have some details in your head and you want to generate that, type those details. A giant raw column with glowing blue crystals growing on its back, and as you can see, we have what we typed in the prompt. This is my second generation have this example where I just type a shoe and we got a random shoe in random color in random location. This is the second generation of the same prompt and this is a detailed prompt result, a worn out leather hiking boot covered in mud. As you can see, this is crazy. We have so much detail in the shot. This is my second generation we have this example a spaceship. As you can see, I did not type any specific thing about the ship. That is why we got this result. We have another example of the same prompt. Now with this prompt, a sleek sliver alien spacecraft landing in a desert at night, and this is the result that we got. This is my second generation. We have this final example, a flower, you will get random flower. With the same prom as you can see, it changed its result, but with a prompt a single white rose, with the dew drops in a crystal vase, check this out. Now even the second image, we have what we type. As you can see, guys, with some mediocre prompt, you will get mediocre result. But with the detailed prompt, you will get better result. B specific. 3. Basic Level: Use Adjectives: Welcome back, everyone. Now use adjectives in basic prompt engineering. So what are adjectives? Like, you have to define the texture, you have to define the mode of your scene. Basically, you have to define what you see. So we have these examples. A forest, you will get random forest, random location in the forest, and random lighting. This is the second example with the same if you define the objective of your scene, you will get detailed results. In this example, as you can see, I type this detail airy misty dark forest with twisted trees. As you can see, we got what we typed in the prompt. This is my second generation of the same prompt. Now we have the next example where I just type robot, I got this robot shot. If you describe the texture, for example, a rusty ancient steam powered robot covered in a moss, you will get this type of detail next we have this example where we have a simple word clouds. We have this image and we have the second image of the same prompt. Now, with details with some adjectives, we have this result, fluffy cotton candy, pink, soft clouds at sunset. As you can see, I did not describe any details about the shot, but with some few words, you will get better result. This is my second example of this pro we have this example where I just type a simple room. We got this room image, even with the room word. I think this image looks great. This is my second example of the same word, but it's not about the image quality or the image scene composition. It's about what you want. So for example, if you want this type of room and you get this room image with some few words, you can use that. But if you want some details, if you want some specific things in the room, you have to type that. So in this example, I have cluttered cozy warm library with a fireplace. And as you can see, this is a better looking image. This is my second image. Now we have this image. We just type simple word Ocean. This is the second image of the same word. Now, with the turbulent, dark, terrifying ocean waves during a storm, you will get this type of result. Check this out. This is my second image. Now we have the final example, where I type texture. I have these random texture. But if you put some details about the texture that you want, in this example, I have rough weather cracked stone wall texture. What we got. So in order to get better result, you have to define your adjective like texture, mood, et cetera. 4. Basic Level: Include Style: Welcome back, everyone. Now include style in your prompt. For example, how you want your image to look. You can type like vintage style, retro style, photo realistic style, or even three D. You can define any style that you want your image to look according to your style. We have this example. We type toy monster, three D render, piza style, and real engine five. These are my style and check this out. This is the image that we got. This is my second example of the same now next we have this example where I type architectural drawing of a bridge, rough pencil sketGraphte. This is, by the way, my style of this image. Without this style, you will get a drawing of a bridge. Now we have this example with the same prompt. Next, we have this example, a car driving in a grid sun synth wave, retro wave 80s style. As you can see, I have defined this style. That is why we got this type of result. Is my second example of the same prom. Next, we have this image type close up of I micro photography, I pair realistic. So this is my style. This is my second image. Next, we got this image, a line made of paper or gaming style, papercraft art. This is by the way, my style. This is my second image. Next, we got this high school boy anime style. This is my second image with the same style. Now, if you tweak your style, like for example, this is the same style, and if you change the timeline of the style, you will get different results. So for example, this is the anime style high school boy. With the same prompt, I describe 80s anime style and check this out. We have different result. So this is a modern anime. This is the 80s anime. This is another example of 80s anime style. Now we have the 90s style anime. Check this out, so you can define any style that you want. 5. Basic Level: Avoid Confusion: Welcome back everyone. Now in basic prompt engineering, you should avoid confusion. Don't use unclear word, instead of cool, say, cold and icy or modern ends, let me show you what I mean. We have this example where I just type a bat and as you can see, we have this bat image. But what if you want to generate a baseball bat? That is why I say avoid confusion. So you have to type a baseball bat leaning against a wall or you can type simply baseball bat. This is my second example of the same prompt. Now we got this example. I want a simple desert. So I say no trees. Let's say you want to say no trees and you want a desert, but you are wrong. You will get these type of images. This is the second image with the same prompt, so you have to type desert. A ban desert landscape, empty sand dunes and check this out. This is a crazy clear lizard. So avoid confusion, use clear word. This is the second example of the same next, we have a glass of water on a computer. Check this out. The AI generated what I ask it to generate. This is the second image of the same prompt, and now we have a clear prompt result, a glass of water sitting next to a laptop computer. And check this out. We have this glass next to this laptop. This is the second example of the same prompt. Next, we have cranes. As you can see, we have these random cranes. This is the second image of the same prompt, and now we have construction cranes, building a skyscraper. Check this out, this is the second example and final a flying car, check this out. We have the second example and with some details a futuristic car hovering in air above the ground. We have what we ask it to generate. This is my second image of the same prompt. Avoid confusion. B clear, use clearword to generate your image. 6. Basic Level: Keep It Short: Welcome back everyone. Don't make long prompts because most of the cases long prompt does not work as good as a short prompt. Keep it short. Make your prompt two to three clear sentence. Sometimes two to three clear sentence work better than long rambling text. We have this example, a landscape with mountains and a river as you can see with few words, we have this image. We have the second image. But if you type some keywords like mountain landscape, river, pine trees day, we did not put any other details than these keywords and check this out. We have a better looking image than these two. We have the same image with the same prompt. Now we have this example, make a logo for a coffee shop that looks modern. This is my second example. And with some few keywords, coffee shop logo, minimalist vector and modern, check this out. We have plain looking vector logo. We got a robot that looks like it is made of gold and silver. This is the second example and this is an example of keywords. Robot, gold and silver plating, metallic, shiny. This is the second example. Next, we got a tasty burger with cheese and lettuce and tomatoes. This is the second generation of the same pro. And we have cheese burger, lettuce, tomato, food photography, delicious. And check this out. We have a better looking image than before. So in this image, I did not ask for the ketchup or the fries, but it generated anyway. So in this example, as you can see, we don't have any other dish than this cheese burger. And we have this example an astronaut walking on surface of the moon. This is my second image. This is not a bad looking image. I'm not saying that. I'm just saying that you can make your prompt even shorter and simpler than this prompt. So astronaut moon surface, space, NASA suit and check this out. With few words, we are getting this result. Keep your prompt short and easy for AI to understand what you want to generate. Next, we got a dragon breathing fire in a dungeon. This is my second generation of the same prompt. Check this out dragon fire breath, dungeon, fantasy. So with four words, we are getting this kind of results. This is my second image of the same prompt. And finally, we have a picture of a city in the future with flying cars. Check this out. And with a simple and clear prompt, future city flying card, Sci Fi skyscrapers, we are getting this kind of crazy results. So keep your prompt short. Don't make long prom because long prom don't always work. 7. Intermediate Level: Describe the Shot: Welcome back everyone. Now, tell the AI what type of shot you want in your prompt. If you want a specific camera shot camera angle, you have to tell AI what type of shot. We have some examples. First, we have the shot like for closeup. For extreme close up, you can type macro. For white shot, you can type wide angle. If you want a draw shot or you want bud eyes view, then you can type but eyes view. Let me show you some examples of how you can generate images with specific camera angles or shot. We have this example, macro shot of a leaf and as you can see, I'm describing my angle macro. That is why we are getting this type of result. This is my second example of the same prompt and we have the white angle of a leaf. As you can see, I'm using the same subject, which is the leaf or you can say object. I'm just changing my camera angle. With a white angle, this is how the leaf look like. Now we have the low angle. This is by the way, looking up angle. That is why we called it low angle and we have this result. Next, we have the low angle again of a wooden house inside a forest. This is my second example of the same prom. Next for drawn or aerial shot, you can type bud eyes view of a wooden house inside a forest. With the same prompt, I'm just changing my angle and as you can see, guys, the image is looking different than before. We have the white angle view of the same house in the forest. Check this out. Now for the three D perspective shot, you can type isometric view. The same prompt, I'm just changing my camera view or angle and we are getting different results. Next, we have the shallow depth of field of a wooden house inside the forest, Teck this out and final, if you don't describe any shot, if you just simply type a house in a forest, you will get different camera angles and different shallow depth of field. If you want your shot to look interesting, then describe your camera angle or camera shot. 8. Intermediate Level: Add Lighting: Welcome back, everyone. So if you want to make your prompt, make your images look interesting, you should type the detail of the lighting condition of that scene. So, for example, say how the light should look in this scene. So let me show you some example. Now, we have this golden R and apple on the ground, and as you can see, I'm describing my lighting. That is why we are getting this type of result. With the second image with the same prom, this is the result that we got. Now in this prompt, I'm typing rim lighting with the same prompt, I'm just changing my lighting condition and as you can see the apple is now looking different than before. This is the second image with the same prompt. Now we have neon apple on a ground and check this. We have neon apple on a ground. Next we got cinematic lighting and apple on the ground and check this out, we have this cinematic lighting. You can also tell the AI where the light is casting. For example, you can type on the left side from the right side from above and from the low angle type. You can type the lighting direction as well. Next, we have cyberpunk lighting and with the same prompt apple on the ground, this is what we got. This is my second. As you can see, we have soft box lighting and apple on the ground and check this out. We have this soft box right here. Next, we have candlelight and apple on the ground, and we have the candle apple on the ground. Finally, we have natural light and apple on ground and check this out, we are getting the natural light. We are getting what we type in the prom. So if you want your shot to look interesting, you can type the lighting condition of your scene. This way, you will get a more clear and better results. 9. Intermediate Level: Set the Mood with Colors: Welcome back everyone. So to make it to the next level, you can type the mode and the color tones. Let me show you what I mean. We have this problem abandoned car inside and abandoned mention, pastel palette and check the st we are getting this result. Now, if you change the mode like monochromic, abandoned and check this ut, we just type monochromic and we are getting this type of result. We are changing in the mode, the color tones of the image. Next, we have the vibrant and I type this vibrant keyword because it described the mode, the color tones of the image. And we have the same example with a second image. Next, we have this muted. This is, by the way, my color tones, and have a look. We have the same prompt result but with different color tones. This is my second generation of the same prompt. We have the CPA tone and have a look. We have this CPA tone effect on top of this image, have a look. Even if you are describing your color tones according to the time, it will also change the models, the mentioned style of your image as well. In this prom, as you can see, I type CPA tone and CBA tone basically is the black and white decade color tone. That is why it generated this old model in old mentioned. If we look at the previous, we are getting some modern car in the images as well. With the CPA tone, we are getting the old model card. Next, we have the synth wave, and we are getting the modern card because the synth wave is like the modern color tone and have a look. Next, we have the earth tones, and we are getting this type of result. This is the second image. We have black and white noir color tone effect, and this is the image that we got from it. Have a look. Next, we have warm tones and check this out. This is my second image. In this example, I have cool tones and check this out. We have this simple abandoned card inside and abandoned mention. Without any tones, this is how it look. Have a look. Now you can also combine multiple colored tones, and then you will get interesting result. Let me show you. So in the final prompt, I have abandoned car inside and abandoned mention, and this is where my prompt of the colored tones begin warm color palette, CPR tones, cozy autumn atmosphere. Orange and brown hues. If you don't know about the hues, it's like the color. You can swap the colors using the hues. That is why I say orange and brown. This is my second example of the same prompt. If you want to make your images more interesting, than describe the color tones. 10. Intermediate Level: Reference Art Style: Welcome back everyone. To make it your image generation to the next level, you can reference a famous art style. So as you can see, guys, we have this famous art style like we have the cyberpunk anime, oil painting, low poly unveil engine, water, color, et cetera, we have pixel art. These are the famous art style. You can reference this in your image and you will get this style image result in your image. So now we have this example where I type sunflower in a field and this is my reference style. This is by the way, the second image of the same prompt. Now with a different style with the cyberpunk anime style, this is what we now with the low poly style with the same prompt, we are getting different results. Next, we have the unreal engine five result. As you can see, guys, this is a fantastic looking image. Next, we have the steam punk style, and as you can see, guys, we have the mechanic on top of the sunflower. We have the second image of the same prompt. Next, we have the pop art. This is by the way, my style, and this is my second image of the same prompt. In this prompt, I have the pencil sketch. That is why we are getting this type of result. Is second image. If you want to generate pixel art style, you can type that in your prompt and you will get pixel style result. This is second image. Now in the pixel art, we have eight bit or 16 bit pixel. I type pixel art 16 bit. That is why we are getting different pixel art result than before. This is, again, the 16 bit style. And with the eight bit style, we have this. So if you notice these three style is different than each other because we are defining even more details of the style like 16 bit or eight bit. And we have this example, the robot, you can define you can mix multiple style like a robot, oil painting style, thick impasto, burst stroke, colorful and abstract strictly traditional art vibe and have a. So you can reference famous studios like Pixar unreal Engine, five et cetera, in your images, and you will get those style images in your image. 11. Advanced Level: Use Camera Language: Welcome back everyone. Now, if you want to take your image even to the next level than before, then you should use camera language. So what is camera language? Basically, the goal is to describe AI what you see and describe how the camera captures. This involves lens, the stock and the setting of your camera. So let me show you some few examples so you can better understand what now in this example, we have full moon behind a building, and as you can see, we have a great looking building, great looking image right here. Now, with my tally photo shot 600 MM lens perspective compression, as you can see, we are getting different results. This is my second now climbing a mountain view and as you can see, we have this random view, and now first person POE GoPro footage, this is what we got. We are using camera language. We are describing how the camera captures the shot. Next, we got this movie scene of a coffee cup and as you can see, we have these images. Now cinematic still of a coffee cup in the rain shot on 35 MM anamorphic lens player, and we got this aspect ratio moody blue lighting film noi atmosphere. This out. We are using the camera languages. We are describing what camera used to capture this image. That is why we are getting this type of clear and criticized result. Next, we have the picture of a forest from above. This is the same prompt example, and check this out top down ideal drawn shot of a pine forest in winter. Geometric composition, flat, lighting, four k resolution, DGI Mvic style. So we are even describing the camera style that used to capture that have. We have the second example of the same we got apple falling midair. This is my second image of the same prompt and have a look. We have apple falling mid air, short on de portra 400 film grain, slide leaks, warm vintage tones, muted colors, disposable camera aesthetic. We have this example. Next, we have this fisheye lens, barrel distortion, veneeating hart sunlight. So we are describing the camera languages. We are describing the camera settings, the camera type, and the camera lens to generate the image. Next, we have a race car going fast. So by the way, if you want to generate car going very fast, instead of typing going very fast or fast, you should type high speed photograph because this is what it called, you should type high speed photography and you should type these details as well. You can even describe your shutter speed. This is by the way, the second image of the same we have this example, a photo of a car in the street. This is another photo of a car in a street, and we have portrait photography, 65 prime lens F 1.8 book. So this is, by the way, your aperture in your camera. I know that because I'm using my camera and I know everything about camera lenses and camera aperture, shutter speed act. So you can use camera language to generate your images, and you can take your image generation to the next level. Next, we have this example again, with the same prompt, we have these big mountain landscape. This is the second image. And with ultra white angle shot of a Swiss Elf 16 MM lens, this is what we got. Have a look. And final, a close picture of an eye. This is the second example and with extreme microphotography. So for extreme closeup instead of typing extreme close up, you should type micro photograph. Of a human IRS 100 MM macro lens, F 2.8 SR focus on pupil, detailed texture, intricate, pattern, studio light, and have.T way you can take your AI content gradation to the next level. 12. Advanced Level: Emphasize Important Element: Welcome back, everyone. To take your prompt engineering event to the next level, you should tell AI what matters the most. So let me show you some examples so you can better understand what I mean. Now, as you can see, I have this yellow taxi in New York. We have this example. And if you want to focus only on the taxi, you should type black and white photography of a New York City street, selective color, a bright yellow taxi cab in the center, high contrast, monochrome so as you can see, we are taking the focus to the taxi, not the whole image. This is my second image. Now we got this robot made of glass. Check this out. This is a nice looking image. But if you want this type of glass, then you should type a robot made entirely of transparent crystal glass, light reflection, caustics, prismatic rainbows, internal gears, visible, gleaming eight K render. We are taking focus to the robot, not to the entire image. In the previous example, the image is look empty. But in this example, as you can see, guys, everything is filled with the detail. We got this example, a garden with a red rose. As you can see, guys, we are focusing on the red rose, but at the same time, we are focusing on to this area, this table. And in this example, even we are focusing in the background a little bit. Not too much, but a large green garden, huge red rose 1.5 in the intermediate foreground, focus on the rose Blur garden background. Check this out and have. We are taking focus to this rose. So emphasize important. AI what matters most. A robot holding a candle. This is the same prompt example, and check this out a dark room, robot illuminating only by a single candle flame. So we are describing that the robot is holding only the candlelight, and the whole scene is illuminating by only the candlelight. There is no source of light other than this candle. That is why we are getting this type of crazy result. Now we have this example, an old wrinkle man hand. This is another example and check this out. Extreme close up of an elderly man hand, emphasize on deep wrinkles, et cetera, blah blah, and have a look. You get the point. We are typing what matters the most. Next, we have this small robot with a big monster. This is the second image. By the way, these are the great looking image with a few words and have a look. Tiny robot silhouette standing in front of closel skyscraper sized monster. Overwhelming scale different looking up atmosphere, haze, and he. This way we are imagining the size of the robot or the monster. Finally, we have an apple. This is the second example and a single red apple floating in a void, pure white background, soft shadow, studio lighting, minimalism, focus on the object only. As you can see, guys, we are focusing only this apple, not the whole image. This is the second image. Tell AI what matters most and avoid getting some unwarranted object or subject in your images. 13. Advanced Level: Analyze What Works: Welcome back, everyone. So the final step in prompt engineering is analyze what works. Say prompts that works, study them, understand why they works, then use similar language in your new prompt. So as you can see, I have generated so many images. Now, I know the language of AI. That is why I used simple language to generate high quality images. So instead of using mushroom in grass, you should type your angle like worm's eye views from the dirt, looking up at the tall mushroom like trees. We can even type the perspective of anything. We have this second image. Next, if you want a cowboy in front of the big sunset, you can type silhouette shot, Cowboy on a horse against a giant orange setting sun, no detail visible on subject strong outline, high contrast, and have a look. We have a robot looking in the middle and if you want a glass to break, if you want a high quality result, you should type shot through a broken mirror reflecting of a robot phase, this is my second image. Next we got this looking at a beach. This is the second image. Now with some details shot through an open window frame, looking out at a tropical beach curtains flowing in foreground depth layering. You can use this type of keyword AI language to describe your shot. Next, we got this toy car in a big room, have a look, and we have high angle shot. By changing the angle, as you can see, guys, we are getting different results and this mediocre result. Now, you can get this type of result by typing double explosion photography, silhouette, et cetera, and you can type a cartoon dog or if you want it in a three D style, you can type three D style or three D animated movie style. If you want a painting style image, you can type a blurry painting of a poem. But if you want to take this style, even to the next level, you can type oil painting in the style of Cloud monic. And next, as you can see, we have all art, and we have this vibrant street art mural on a brick wall. Finally we have camping gear on the floor. Basically, I wanted all the camping gears placed in a flat surface. So in my final shot, I typed this KollingPhotography, flat lay camping gear organized neatly at right angles, overhead shot, distinct separation of item, clean background check this out. So you can analyze what works best in your prom, and then you can use that same style in your latest prompt. 14. Steal Any Prompt: Welcome back everyone. Now I'm going to show you how we can replicate any content from the Internet. Go to ling AI and this is the brand new interface of ling AI. Now go to generate and go to Image Generation. Once you go to Image Generation, you can simply click here onto this dolphin sign. Now we have this deep seek R. If you want to replicate any image, you have to upload that image from here. You can click here to upload your image, Let's replicate this. Now once your image is uploaded, you can simply click onto this button. It will upload your image, and now the deep set is now analyzing the image, and it is going to give us the as you can see, we got this prompt. I'm going to click here, use this prompt and I'm going to copy this prompt. You can also use Image 2.1 of cling A to generate the image, but I'm going to show you a free way to generate unlimited image. Go to Whisk dot google slashxlashTols slash Wix. Now here we can paste the prompt and we can click here to generate this. As you can see it generated this image which is not close to our original image which is this one. In this case, if you are getting this problem, you can simply click onto this regenerate button and it will give you a second prompt. I just did that and it gave me this prompt, so I'm going to copy this prompt, go to Whisk again and I'm going to paste my brand new this time, as you can see, we got the same prompt result as we got in the original image. Now the second method to replicate any content from the Internet is simply click here and upload your image. This time, I'm going to upload a different image. Now, once your image is uploaded, you can simply clear the prompt window, which is this one and simply click here. You can also click here and as you can see, this is the detail prompt to generate this image. I'm going to press Control A, Control C to copy this close this window and as you can see, we have the same exact result from this image. I'm going to click here, de select your subject, paste your prom, that you just copy it from here and generate this. As you can see, we got the same exact result that we have in the original image. Now, the best use case of this is that you can replicate any videos that is getting viral on the Internet. Right now, these kind of content is getting where we have some organ, some object has face on the the talk. So we can replicate these kind of content very easily. You have to clear your prompt window, click here to generate the exact same images. Now, once you have your image, as you can see, we have some error. So in this case, we can deselect, click here, and we can control A Control C to copy this prompt, close this window, paste them. By the way, you have to deselect. If you don't deselect, if you have selected your subject, it will generate your images, including this subject. So you have to deselect your subject first. Now I'm going to generate these images. Now, once you have your images, you can simply click here, refine and you can replace your subject. So I just type replace Brain with mobile, remove all other characters clear background. Now as you can see, the background is clear, but we have this error. So I'm going to type replace character with now if you are getting this kind of error where you don't get your character done, you have to close this refine mode, as you can see, we can steal somewhat from the prompt. So we have three D rendering depicted a cartoonish angry human brain character standing in the foregroun. So we can simply just copy this Control C, and we can Control A to select all the text press backspace. Now we can page this prompt, and now let's create this. And as you can see, we are getting closer. Now if you really want to replace the brain with a mobile, we have another way. So go to flow, create a new project and I. Once you go to image, you can simply click here and you can upload your image. Now let's crop and save and now I'm going to type replace brain with mobile, remove all other character, clear background. Now we have to use Nano Banana Pro. It is the most advanced image editing AI to right now. I'm going to click and as you can see, we got these images. I'm going to simply click here, open this image. We can further edit this image as well. So I'm going to type change color of the hand and face to black. So I'm going to change this body part color to black and now have a look. We just replicate the wire content and we created our own. You can also change the mobile to any other object. So for example, so I just type this symbol prompt. Now change directed to mobile Charger. Place it in the room in the foreground depth of field cinematic. You don't have to type any more detail. You can just simply type these kind of keywords to replicate your content and make it cinematic. And now, as you can see, we have this charger. Now you can fix this charger. You can type further more prompts to fix this charger, but you get the idea. So this way, how you can replicate any content from the Internet using AI. 15. How to Create AI Consistent Character: Welcome back everyone. Now, I'm going to show you how we can create consistent character. First of all, go to Whisk. You have to go to absgoog slash FX slash tools slash WIS. Now, once you got here, you can click here and in the subject, you have to upload your character. Now I'm going to upload myself right here and you can do a couple of things in Whisk. You can place this character in a different scene. For example, I'm going to click here, you can type your Prompt as well, or you can upload your scene. I'm going to type parking lot and let's generate as you can see, we have this parking lot in the scene, we have our subject. Now here's the deal. If you want to type any changes, you can do that really. But if you don't want to type any prompt, you can still manage to place your subject in this brand new scene. Simply click here onto this button, it will generate your images. If you want to generate four different aspect ratios, you can simply click here and you can select one of these aspect ratio. As you can see, it plays the subject in this brand new scene. You can also delete this scene. If you want your subject to do anything, you can also do that. So you can click here to add another subject, and in another subject, you can either upload your image or you can type any prompt. So I'm going to type burger fries, nuggets on plate. Now I want my subject to eat this fast food, and I'm going to also type a luxury hotel. Now, once you have your scene and everything else, you can simply click onto this button. You can also deselect or select anything by clicking on Togle. And as you can see, the subject is sitting with the fast food in this empty luxury hotel. I just type person eating fast food, depth of field, cinematic lighting. And as you can see, we have the cinematic lighting. We have the better view in the brand image than before. So you can improve this further if you want to. So we can also go to refine. So I just type person eating burger and have a look. So this way, you can create your consistent character. Can make your arrector do anything. You can also place your character in a car, so I'm going to delete this and I'm going to place this person in a car. I just type car front view because my person is sitting looking at the camera, so it is better to place him in front of the car and from the front view. We have the car, we have the subject. I diselect my scene this time, and this time, I type person driving car, add a close up, add a co you don't have to write a full paragraph, a full 300 page book to make your content look better. You can just type simple keywords. This way, it will better understand what you are trying to say. Now we are getting this type of error. If you are getting this type of error, the reason behind this is because we are using the image f. We are not using the Nano Banana Pro. In this case, if you are not getting the result that you want, you can simply copy your prompt, move to flow. You can create a new project if you want to. You can use your project as well. Go to images right here and you have to also click Here Creative. Now you have to upload your subject. You can crop it as well, but if you don't want to crop your subject, you can simply click A Crop and save. Now you have to click here and this time, you have to upload your car or any other object. I'm just going to simply click here to download this car. Go to Flow upload this car, and I'm going to type this simple prompt. Person driving car close up cinematic. I'm going to also use the Nano Banana Pro, which is the latest image editing model right now in the mark and have a look. The person is now driving the car. This way, you can create your consistent character. I have some other example that I want to share. Now as you can see, we have the person, we have this french fries and have a look. The person is eating the french fries. If you open an image, as you can see, the prompt detail is empty, is because I did not type here we have another examples. Now, you can also apply any scene onto your character or onto your images. It's very simple. Go to Whisk, I'm going to delete this. You can also delete the scene as well. Let's delete the scene and now we have only the subject. Now go to PinRSs and you can type Photo effect. You can select any effect that you want. Once you have your effect, you can go to Style, upload your style. Now I'm going to select this style let's empty the prompt window, and now let's click here to generate. Now this time, it is going to apply the same style onto the subject and now check this out. Isn't this amazing? We have the consistent character and we have the style. Let's apply some other style as well. So let's delete. Let's upload another style. Now we have another trick. If you are not getting your style right, you can simply click here. You can copy this detail. Now I'm going to generate the images with the style and let's toggle and unselect the style, and let's this time prompt. And now let's create this style and now have a look. Check this out. Isn't this amazing? We are getting different kind of style. The best part of using the prompt style is because as you can see, it did not change anything in my image except the lighting and the camera cinematic quality, except that it did not change anything. We have the Adina background. We have this light, the mic is covered with some plastic bag, and as you can see, everything else remains the. Let's delete this. This time, I'm going to use this style, so we can empty the prompt window as well. Let's create with the style and let's create one without the style selected. And now check this out. With slide selected, we have this result, and without slides selected with the prompt only we got this. I'm going to empty my prompt window. Let's generate this. Let's copy this, unselect this, paste the prompt and let's. Now in some cases, you will get this kind of error. So it's better to unselect your style and paste the style prompt in your prompt window. And this way, you will get your improved result. Let's use this with the Nano Banana Pro let's see what it's going to bring us on the table and have a look. This is how we can create consistent character using AI.