How to Create AI Content That Reaches Millions of Views · The Adi Bergman Method.

Most people work with AI like this: they write a prompt, get an image, throw it into a video tool, upload it to the web, and hope for the best. This can produce something pretty. It almost never creates something people want to see again. In session 20 of the AI Master Club, Adi Bergman broke down the real deal: how to create AI content that features an idea, voice, story, comedic timing, visual world, and movement that feels alive. This is not a guide on which button to click. This is a guide on how to think like a content creator in a world where everyone has access to powerful AI tools.

What you will have by the end of the guide

• A method to figure out if an idea is even worth a video
• A writing process that starts from emotion rather than text
• A working structure called the Matryoshka of documents
• A way to build a visual world with a master shot and camera angles
• A method to generate the same character in multiple emotional states
• A basic workflow for animation, lipsync, and editing
• Ready-to-copy prompts for ChatGPT, Claude, Nano Banana, Kling, and HeyGen

The problem: AI alone does not create good content

The principle: Do not start with what AI tool to open. Start with what human point people recognize within 3 seconds. The tool comes later.

The Moday method: The viewer is not waiting for you

4 conditions for a strong video opening

• A familiar concept in the first three seconds: a donut, a barbershop, an ex, a holiday, a product, a work situation
• A clear emotion: laughter, annoyance, stress, nostalgia, embarrassment, or relatability
• A promise that there is more to wait for: a first punchline that hints more punchlines are on the way
• A reason for a second view: something funny, dense, or surprising enough that the viewer returns to the start

Idea check prompt using the Moday method

I want to check if a short video idea can stop the scroll. The idea: """ {Paste the idea here} """ Analyze it based on 5 points: 1. What familiar concept to the viewer jumps out in the first three seconds? 2. What emotion is the video trying to trigger? 3. Does the viewer have real friction with this concept in life: money, time, energy, relationships, food, work, kids, status? 4. What is the first punchline or surprise that can hold them for another 3 seconds? 5. What is missing so people will want to watch it a second time? Do not write a script yet. Only a sharp and direct diagnosis.

Writing from the end: Do not start with a script

Matryoshka of documents: Adi's writing process

• Personal document: A free brain dump of everything you think and feel about the topic. No censorship and without thinking if it is good
• Ideas document: Extracting real directions for the video out of the mess
• One-sentence definition: A one-liner that defines the video. This is the compass for everything that follows
• Punchlines: Writing jokes, moments, imagery, and exaggerations before writing a script
• Logline: What happens after what. The sequence of events
• Script: Only now do you turn all the materials into dialogue or voiceover
• Production appendix: What needs to be created, filmed, animated, recorded, and edited

Prompt to create a Matryoshka of documents

I am working on a short video for social media. The initial idea: """ {Paste the idea here} """ Help me build a Matryoshka of documents: 1. Ask me 7 questions that will help me write a personal document on the topic. 2. After I answer, extract 10 possible ideas for the video. 3. For each idea, write one clear one-liner. 4. For the strongest idea, suggest 20 punchlines or comedic moments. 5. Then build a logline: a sequence of what happens after what. Important: Do not try to be funny by force. Do not write cheap wordplay. Look for human situations, embarrassment, exaggeration, conflict, and real friction.

Where AI does help in writing

Concept world expansion prompt

I am writing a video about: """ {Example: a barbershop / donuts / an ex / business owners / Shavuot holiday} """ Give me 100 concepts, situations, objects, phrases, habits, small pains, and conflicts that people know from this world. Divide the list into categories: 1. Physical objects 2. Phrases people say 3. Embarrassing moments 4. Small pains 5. Things that can be comedically exaggerated 6. Visual imagery Do not write jokes. Provide raw materials for writing.

World creation: Master shot before anything else

Master shot prompt in Nano Banana

Create a cinematic master shot for the following short video scene: Scene: {Describe the scene here in simple sentences} Requirements: - Vertical 9:16 composition - Realistic but slightly heightened cinematic look - Clear readable characters and body language - Natural lighting that matches the scene - No text, no captions, no logos, no watermark - The image should feel like the main establishing shot of the scene - Keep the environment coherent and practical for later close-ups Style reference: {If you have a reference from Pinterest or an existing image, upload it alongside the prompt}

Prompt to break down a master shot into 15 angles

This is a master shot image of a scene I want to turn into a video. Create 15 prompts for me to generate additional camera angles from the same scene: 1. Five special cinematic angles that express emotion or tension. 2. Five angles focusing on the main character in different sizes: long shot, medium shot, close-up. 3. Five angles focusing on the second character or an important object in the scene. In each prompt: - Maintain the same world, same lighting, same clothing, and same perspective relative to the master shot. - Do not change the visual style. - Do not add text or captions. - Write in English, short and clear, suitable for Nano Banana.

The same character in multiple emotional states

Prompt to generate an emotion set for a character

Using the attached character image as the exact identity reference, create the same character in the following emotional states: 1. Neutral 2. Angry 3. Worried 4. Sad 5. Excited 6. Confused 7. Suspicious 8. Embarrassed 9. Deeply moved Keep the same character identity, same clothing, same visual style, same lighting, and same background. Only change facial expression and subtle body language. No text, no captions, no logos.

Video: Transition between start image and end image

Prompt for Kling / video model with first frame and last frame

Use the first image as the starting frame and the second image as the ending frame. Create a smooth, realistic video transition where: - The oven door opens naturally - A human hand pulls the tray out - The object inside the tray is already partially visible as the oven opens - As the tray moves outward, the final object is revealed fully - Keep the camera angle, lighting, oven design, and hand motion consistent with the two reference frames - The motion should feel like real footage, not animation - No text, no captions, no extra objects

Lipsync: Why HeyGen enters the process

Short prompt for HeyGen Motion

Animate this character naturally while speaking. The mouth should move fluently and accurately with the audio. Add subtle human facial expression, eye movement, and small head motion. Keep the background stable, but allow subtle natural movement in background elements so the shot feels alive. Do not exaggerate the motion.

A small point that makes a big difference: When there are cars, people, a flame, passersby, or subtle movement in the background, the video feels more believable. The brain is used to a living world. A frame where only the mouth moves and everything else is frozen feels artificial.

Full workflow for a short AI video

• Choose a human situation with real friction
• Write a free personal document about the field
• Extract ideas and choose one one-liner
• Write punchlines before the script
• Build a logline: what happens after what
• Write a short, dense, and clear script
• Create a master shot that defines the world
• Break it down into camera angles
• Create several emotional states for each character
• Record the dialogue with the appropriate intonation
• Animate characters and lipsync in HeyGen or a parallel tool
• Combine everything in editing: pacing, fast cuts, punchlines, repetitions, music, and sound

The toolkit from the guide · All links in one place

ChatGPT

Thinking, idea development, concept world expansion, and prompt writing

An excellent working tool for the initial stages: checking an idea via the Moday method, creating concept lists from a specific content world, phrasing prompts for Nano Banana / Kling / HeyGen, and organizing the writing process before production begins.

Tip for creators: Do not ask it to write a viral video for you. Ask it for raw materials: concepts, conflicts, situations, and idea breakdowns.

The AI Master club opens its doors to you, featuring a live meeting every week, recordings, and a community. 47 NIS per month, cancel anytime.

Claude

Deep development, structure, scriptwriting, and strategic thinking

Suitable for breaking down an idea, building a nested set of documents, writing a logline, improving a script, and organizing long guides. When you need depth, wide context, and language precision, this is a very strong tool.

Tip for agents: Give it the one-liner, the punchlines, and the logline, and then ask it to improve structure and pacing rather than invent everything from scratch.

Gemini / Nano Banana

Generation of images, characters, master shots, and visual versions

Nano Banana is the image model built into Gemini. In the guide, it is used to create a master shot, characters, a set of character emotions, an opening image, a closing image, and visual references that are later fed into video tools.

Tip for agents: Use it first to create one good image that defines a world. Only afterward, break it down into angles, closeups, and emotions.

Google AI Studio

Advanced work with Gemini and Google models

A Google environment for more technical work with models, experiments, and APIs. Relevant when moving from manual prompt usage to a system that runs a process automatically.

Tip for agents: This is the place to move to when you want to connect image generation to an automated system rather than working manually every time.

KIE.ai

A single interface to run image, video, voice, and LLM models

KIE.ai was mentioned in the meeting as a convenient and relatively cheap way to access various models from one place. Instead of opening an account in each tool separately, you can run image and video models through a single interface or API.

Tip for agents: Especially useful when building automation with Claude Code or Codex and you want a single API that centralizes several models.

Kling

Animating images and creating video from a start image and an end image

A video tool suitable for creating movement between frames, for example, an image of a closed oven and an image of an open oven with an object inside. In the guide, it comes in at the stage where you already have clear visual anchors and want the model to complete the motion.

Tip for agents: Do not settle for a text prompt. Give it a strong first frame and last frame, as this dramatically increases control over the result.

HeyGen

Lip syncing and animating a speaking character from an image

A central tool when you need to take an image of a character and make it speak. You upload an image and audio, and the tool creates mouth movements, facial movements, and sometimes subtle background movement. In the guide, it enters after you have a script, a character, and recorded audio.

Tip for agents: In Hebrew, it is better to record the sentences in your own voice with the right emotion, and then use that as a basis for the lip sync.

ElevenLabs

Voice generation, voice cloning, and dubbing

An audio tool for generating voices, dubbing, narration, and voice cloning. In the guide's process, it fits after you have recorded the intonation yourself and want to replace it with a different voice, such as a character, a woman, a child, or a narrator.

Tip for agents: Keep your human intonation. The final voice can change, but the pacing and emotion are more important than the tone.

MiniMax

Voice, video, and additional creative models

An AI platform with models for voice, speech, video, and creative tools. In the meeting, it was mentioned as an alternative for voice cloning and voice generation, especially in places where you want more flexibility.

Tip for agents: It is worth testing against ElevenLabs on the same audio file and choosing based on naturalness, Hebrew support, and work speed.

Seedance

ByteDance video model for scenes, commercials, and cinematic motion

A video model intended for creating videos from a description, image, or references. In the guide, it is relevant for the stage where you want to turn a storyboard, a product image, or a written scene into a commercial or short video.

Tip for agents: Use it after you have a clear storyboard. The more organized the input, the less random the video will be.

fal.ai

Access to video models like Seedance via API

A service that allows running various models via API, including ByteDance video models. More suitable for those who want to connect the process to automation, a script, or a system that builds videos at scale.

Tip for agents: If you are building an automated system, fal.ai can be the layer that connects images, prompts, and video.

Veo

Google's video model

Google's video model for creating advanced video scenes. In the guide, it is mentioned as an option for continuing work with storyboards and images, especially when working within the Google ecosystem.

Tip for agents: Good for testing against Kling and Seedance. The same storyboard can behave differently in each model.

CapCut

Video editing, cuts, sound, subtitles, and video completion

An accessible editing tool for mobile and desktop. After all scenes and voiceovers are ready, CapCut is suitable for cuts, music, sound, subtitles, pacing adjustments, and export to Reels or TikTok.

Tip for agents: Even if all scenes were created using AI, editing is where the video becomes funny, fast, and precise.

Claude Code

Building automations and tools around the content process

Anthropic's code tool that runs from the terminal and can build features, scripts, automations, and workflow systems. In the context of the guide: you can build a system that receives an image, creates prompts, runs an API, and returns an asset package.

Tip for agents: Suitable when you already have a manual workflow that works and you want to turn it into a single button.

OpenAI Codex

A code agent for building systems, scripts, and automations

Codex is an OpenAI code agent that can read, modify, and run code. In the guide, it is relevant for the stage where you want to turn the content creation process into a system: uploading an image, creating prompts, running models, saving files, and exporting outputs.

Tip for agents: Instead of asking it to 'create a video', ask it to build the system that runs the steps you already know work.

The part that cannot be fully automated

Frequently asked questions

You do not have to be professional video editors, but you do need to understand basic pacing: where to cut, where to leave a moment, and how to arrange the punchlines. You can start with CapCut and progress from there.

You can ask it for ideas and raw materials, but according to Adi, the punchlines themselves tend to come out flat if you let the model do everything alone. A better use: ask for 100 concepts from the relevant world, and then choose and write from them.

There is no single tool. The process is more important than the tool: a human idea, script, master shot, angles, emotions, lipsync, and editing. The tools mentioned in the session include Nano Banana, Kling, HeyGen, ElevenLabs, Minimax, Seedance, and KIE.

A master shot maintains a consistent world. From it, you can break down angles, close-ups, and emotions without the video feeling like every frame came from a different project.

Adi emphasized second views. When someone returns to watch the video again, it is a sign that you succeeded in pulling them out of unconscious scrolling and making them choose your content.

Want to turn this into a practical workshop? In the AI Master club, we take such sessions and break them down into implementation: prompts, automations, exercises, and tools that you can introduce into your business or content this week.