[00:02] has already crossed 40,000 subscribers. Most of the uploads are hitting millions of views. In fact, one of the recent videos, posted just 4 days ago, already reached 14 million views. Honestly, even one video like this can get your channel
[00:15] estimated earnings, this channel is making somewhere between $2,000 to $22,000. The crazy part? Creating these videos is need four simple prompts, and I've already shared all of them in the
[00:29] description. Let's get started. First, copy the channel analysis prompt and screenshot of the channel name, along with subscriber count, and upload it. Then, take another screenshot showing the videos, including views, titles, and
[00:43] thumbnails, and upload that, too. Hit enter. Chat GPT will break everything down and show you how these videos can be recreated using AI, along with a full step-by-step guide. But, you don't even need to follow all of it. Next, copy the
[00:56] topic prompt from the PDF and paste it into Chat GPT. Choose how many ideas you want. I'll go with five, and hit enter. It will generate a list of ideas. Just pick the one you like. Now, take the image prompt from the PDF, paste it into
[01:12] Chat GPT, and add your selected topic along with the number of images you want. I'm choosing five. Hit enter. Copy the first image prompt and go to Google Gemini. Select create image, paste the prompt, and add this at the end: aspect
[01:25] ratio 9:16. Generate the image and download it. Now, here's an important trick for consistency. Before generating the next image, always write, "Use the previous image as a reference." Then, paste the next prompt. This keeps the
[01:39] angle, lighting, and overall look consistent across all images. Repeat this step for all five images. Once done, copy the video prompt from the PDF and paste it into Chat GPT. It will generate transition prompts between each
[01:53] image. Take the first video prompt and go to Google Flow. Start a new project and upload all your images. Set the first image as the start frame and the prompt, select the settings, and generate. Flow will turn it into a
[02:06] realistic video within seconds. Repeat the same process for all image pairs. Finally, open any video editor, import all the clips, arrange them in order, and slightly zoom in to hide the watermark if needed. Export the video.
[02:32] short is ready to upload. You've seen how it works. Now it's your turn. The prompts are in the description. Also, if you like this video, subscribe to AI you like this video, subscribe to AI Vision for more content like this.