How to Create Black Stories with Google Veo 3 AI (Step-by-Step Tutorial)

0:00 Man, listen. Have you ever turned on Netflix and said, "Where are all the black people?" Okay, maybe it wasn't Netflix, but it was another channel. Y'all, we need more love and basketball belly. You remember that scene? Friday juice. Why did I get married? Tyler Perry with better wigs. I'm talking baby hairs and HD lace fronts, hun. I'm talking Diary of a mad black woman. Okay. And sinners. So, what if I told you that you can produce an entire movie with one prop using artificial intelligence? Oh, you can do that with Google's V3. And I'm going to tell you how. In one hour, honey, I created this. If AI replaces cashiers, call centers, and drivers, what happens to the brothers who never got a shot past high school? This ain't just tech. It's economic eviction in real time. If AI is

0:46 learning from the internet, it's learning to ignore me. Black folks make up the culture, but we missing from the code. AI is the new gold rush, but the wealth, it's still fenced off behind white tech bros. Facial recognition thinks I'm a threat, but job automation thinks I'm disposable. AI got trust issues when it comes to black skin. You automate the jobs we do and underfund the schools we learn in. Tell me again how AI is equal opportunity. That's why I join school at night. They brought in machines to help us. Next thing I know, 10 of my people were gone. Now they saying AI going to help. Help who? Shout out to all the moms out there cuz I'm doing this with a baby on my lap. Let's get into it. All right y'all. So today we're going to be using Google's V3, which is a video content generation tool

1:31 that allows for you to take a text prop and turn it into a cinematic video. Okay, make sure you visit this site right here, [Music] deepmind.google/models/vo. You can also Google search it, but make sure you come to this homepage right here. And then I want you to type and flow. Then I want you to click create with flow and then select new project number one. Let's storyboard our cinema. All right. So first y'all, we're going into chat GPT and you're going to type this prompt right here. Give me five content ideas for a short cinematic video which will not feature the same character but will have an overarching theme. The clips will be four to eight seconds long and will be used to create my first

2:16 cinematic video. give me ideas where I could produce a video with five clips that appeal to black people using AI. Specifically, how AI is primed to replace jobs held by black people. Now, you don't have to choose my same content idea, but you do want to be specific. So, are you creating a video that appeals to black mothers? Maybe you're creating a video to appeal to H.B.CU students. Get super specific and think about a theme that could be displayed in five different clips. Okay, so here's what it gave me. A theme of vanishing work, right? The disappearing cashier. When's the last time you went to the store and somebody, an actual human being, actually helped you? No, you used that selfch checkout. And how much did that store pay you for doing that? $0

3:01 and0 and you was over there happily putting your stuff in the bag. You got three items, two items ended up being paid for. Anyway, that's none of my business, right? The call center that don't call back. We all done talked to a robot. Press three, four, five, six. From janitor to algorithm Uber no more. Y'all know those self-driving cars where they want you to get in the back seat and let the steering wheel take you wherever the winds go, wherever the winds flow. And then robots that don't tip. If you read through your idea and you don't like it, here's a follow-up prompt to try. Just type give me five more ideas using the same format. Two, let's create our characters. I want you to start envisioning your characters. Okay, so Chad GPT is already going to start you off. You see here it tells me a black waitress laughs with her table

3:48 writing the order with flare. In the next scene, a robot slides food across the tray while the human waitress sits alone in the breakroom. So, it's already giving me a description here. Uber no more. A young black man waits by his car for a ride share passenger. In the next shot, in the same street, the same passenger, but now we see a driverless vehicle that passes him. So chat GPT is going to start you off with general descriptions, but I'm going to teach you how to make it that much better. And now what I want you to do is get specific. So you see here, instead of saying a black barber, I'm saying a 35year-old Africanamean barber. So you may want to mention the age, mention the race. If there's a specific, you know how people be saying specific, specific, specific,

4:34 specific. If there's a specific hairstyle that you want, then this would be where you would put it. So, here you see I said a black high school teenage female with braids, right? And in my next clip, I said a teenage male with a red hoodie. Be specific about what they're wearing, right? If you're talking about a waitress, make sure they're using a uniform. And here's another example. A 24 yearear-old black woman with a big pink afro. I wanted to make it creative. She's wearing a hoodie. I also throw in there that she's wearing headphones and she's also holding a game controller. So, I want you to be extremely detailed in your response because you want your characters to look as though they live in real life, right? We already know that we are not copy paste. We are God's original. Let's select our camera

5:21 angles. One thing that I love about AI is that when you marry that thing with human intelligence, it births a creative bay. Okay? So, don't feel challenged if you're thinking, "How does my character look? I need ideas." That's what chat GPT is for. And other large language models like Gemini. So, I want you to type this prompt here. I'm choosing the cinematic video idea. And for me, it was vanishing work, a visual eulogy to black labor in the age of AI. Then, I typed for each clip, I want you to describe my character in detail and recommend a camera angle and voice tonality. So now chat GPT is going to take those same clips and it's giving you more detail. Right? I want you to pay special attention to the camera angles. Okay? We

6:07 don't want just the same camera angle where the camera is pointed at the character's face and then they're speaking directly to it. So some popular camera angles I'm going to show you in action. Here is an overthe-shoulder shot. So granny over here trying to find a new grandpa. Okay. She on Tinder. Y'all have a wild imagination. But you see, we don't see her face, but the camera is over her shoulder showing that pearl earring that's probably 104 years old. Let's keep going. So, this shot right here is a low angle, right? So, we're looking up at the double chin. This lady doesn't have one, but if it was here, the camera would catch every single roll. All right, so it's a low angle looking up. Okay, here's how it looks in action. So, our character, she's walking, she's talking. We see the gold grill. We see the gold hips. We see

6:54 the tats, honey. And it's all low. Okay. So, next up here, we have a wide shot. So, we want to see his entire background. We see a old couch. We see a television. We see that he's on a dirty floor. He's crying. He's praying. He is heartruck because he texted, "Hey, big hand." to Lkesha. And Lkesha done moved on. Hot girl summer, come on. 2025. Okay, so this next angle is one of my favorites. Okay, it's called a tracking shot. And what it does is it follows the character around, right? Like that ex-boyfriend, ex-girlfriend that just is obsessed with you. Call the police. Okay, so anyway, a tracking shot. Okay, it's going to follow you everywhere you going, everywhere your character is going. But you see the shot diversity here. This is a split screen. Okay. So,

7:40 we have the girl. She's on the left. She's playing a video game. But then on the right hand side, we can also see the screen in which she's playing. So, choose one of those. And we're going to put that in a prompt in just a second. Boy, let's select our voice tonality. You don't want all of your characters to sound the same. Okay? So, if your character is sad, we want to make sure to include that in the prompt that we're building. So, make sure that you're thinking about for each content clip, what is the voice tonality? Let's listen to some of the voices that are part of the clips that I generated. You automate the jobs we do and underfund the schools we learn in. Tell me again how AI is equal opportunity. That's why I joined school at night. Okay, girl. You see how

8:25 she's speaking with passion and conviction. You better go on head. All right. So, let's listen to our 24year-old black gamer. If AI is learning from the internet, it's learning to ignore me. Black folks make up the culture, but we missing from the code. Oh yeah, that was a bot. So, let's hear from Rosco. They brought in machines to help us. Next thing I know, 10 of my people were gone. Now, they saying AI going to help. Help who? Hear that? Again, we have a voice. He's frustrated, right? We see the robotic arms moving behind him. So, I want you to think about voice tonality. And when we construct our prompt, when we finally get over to Google's VO to start creating, I want that to be part of your prompt. Okay, we have reached the final step before we finally produce our

9:11 video. Okay, I want you to go ahead and put in this prompt. For each of my clips, give me one to two sentences that are relevant to each clip and that my character will speak. Ensure all the sentences that they say are compelling, dramatic, but also factually accurate and match the scene. Use the clips from the previous prompt. Incorporate the spoken text into the prompt using quotation marks. Describe the character, the scene, the camera angle, voice tonality, and finally the written text. And here's what I got. All right, so for clip one, it's describe my character. It's describe the scene. So Kesha is scanning groceries in the neighborhood corner store. It's giving me my camera angle, which is the overthe-shoulder dolly shot. It's giving me voice tonality, which is warm, nostalgic, and titched with melancholy. And I also have

9:58 the line that is spoken. I worked here for six years. Now I trained the machine that took my job. And now you have the entire prompt with all of that information right here. Okay, it's a prompt that we can copy and carry over to Google V3. All right, let's do it. Now we head over to V3. Okay, you see how I've copied this entire text? I'm going to select text to video. Then I'm going to go over here to these three bars here that say settings and I'm going to click on settings. If you want a video that doesn't have any spoken word, that's totally fine. You can take those words out of the quotation marks and you will just have your character moving. Okay? You can create the voice over yourself in software like Cap Cut.

10:45 I normally leave my outputs at one because it does cost you credits to generate anywhere between 100 credits and 150 credits. And then you're going to go to what you want the quality to be. So your V2 options are going to have your characters generated without them speaking. It will still be a compelling clip and I'll show you an example in just a second. But if you want your character to speak, you're going to scroll down to highest quality experimental audio. So go ahead and select that. And once you've selected that, then you're just going to press generate. While it is generating, I want to explain a couple things to you. This is not free software. It's not free. It will cost you. When I first started using Google V3, I was on the AI Pro plan. And I could generate videos, but I

11:32 could only do it with V2. All right? So, that means that I had videos, but they were generated without sound. Here is an example. Here are some of the features that you get with V2. But if you're generating and you're like, "Oh my gosh, all the videos that I see on social media, the characters are speaking, it's because they have this plan, and that is Google AI Ultra. Starts off at $125 a month, and that's for the first three months. Next up is dialogue. What will our character say?" So, I decided to go with a different prompt, and this one is about a janitor who's mopping a floor, and he's in competition with a robo vacuum. Okay, so here's the clip.

12:18 I kept these halls clean for 30 years. They replaced me with a vacuum on wheels. In this clip, you see that he doesn't speak, but you hear a voice over. And here's the reason why. In my prompt, it says, "Deep male voice speaks." So, I decided to regenerate the clip. I changed the prompt to say this. Camera tilts up to reveal a solemn face. the African-American janitor says. So you always want to make sure that if you want your character to speak that you say who your character is, you type the word says, and you make sure that the text is in quotation marks before you generate. And then that's the way that you can hear the character speak every single time from their mouth as a first person narrator and not a third person

13:04 narrator. And here's my clip in final format. I kept these halls clean for 30 years. They replaced me with a vacuum on wheels. And there you have it, folks. Drop a comment if you want access to my entire prompt pack, honey. Just drop a comment that says, "I'm ready." Go ahead and click like and subscribe for more content like