Claude Option 1: Use a Project (best method) 1. Go to claude.ai and create a New Project 2. Inside the project, open Project Instructions 3. Paste your caption structure there, something like: "Always write captions following this structure: [Hook line] + [2-3 lines of value/story] + [CTA]. Tone: conversational, no hashtag spam, end with a question. Never repeat the same opening." 4. Every conversation inside that project will follow this automatically. Option 2: Start each chat with a style prompt Paste a "style brief" at the top of the conversation once, then just ask for captions normally after. ChatGPT Option 1: Custom Instructions (always on) 1. Click your profile icon > Settings > Personalization > Custom Instructions 2. In the "How should ChatGPT respond?" box, paste your caption structure/rules 3. This applies to every conversation automatically, no need to re-paste Option 2: Create a GPT (most powerful) 1. Go to chatgpt.com/gpts > Create 2. In the instructions, define the exact caption structure, tone, and rules 3. Save it and use that GPT every time you need captions 4. You can even give it your brand voice examples Option 3: Start a long thread and pin it Keep one ongoing ChatGPT conversation for captions. The context stays alive and it "learns" your style as you go. Not ideal long-term but works for short campaigns. The Real Power Move (for both) Give it 3-5 example captions you've already approved and say: "These are captions I love. Extract the structure, tone, and style from them. Use this as your framework going forward, but write every caption fresh." This way it reverse-engineers your style rather than you having to describe it.

Hey fam — I just dropped a reel on @ai.withmb that's blowing up, and I want to break down the EXACT workflow so you can replicate it yourself. The reel is a 31-second cinematic Burger King ad that looks like it cost $50K to produce. It cost $0. No camera. No crew. No studio. Just AI tools and a solid prompt. View the details here

I designed a single text-to-image prompt with 7 hidden stress tests baked in. Then I ran it through Gemini, GPT Image 1.5, Midjourney, Qwen, Wan, Seedream, Flux, Z-Image, and Uni-1. The prompt asked for an elderly Japanese calligrapher painting the kanji character 勇 (courage) in a candlelit tatami room with very specific requirements: correct shadow direction from a defined light source, three wall scrolls with different kanji, a tortoiseshell cat sleeping on a red cushion, a snowy garden through an open window, and medium format film aesthetics. Every element was chosen to test something most AI models get wrong. What I was actually testing: Specific kanji rendering - Can it produce a real Japanese character, not gibberish? Shadow physics - Candle on the left means shadow falls right. Basic physics that most models ignore. Object counting + variation - Three scrolls, each with a different character. Breed-specific animal - Tortoiseshell pattern, sleeping pose, correct placement on the floor. Reflections on curved surfaces - Candlelight on a ceramic tea cup. Dual-temperature lighting - Warm tungsten interior vs. cool blue moonlit exterior in the same frame. Film simulation - Shallow depth of field with medium format bokeh characteristics. The Results (ranked) 1. Gemini (Nano Banana Pro) - 9.0/10 The champion, and it won by doing something no other model managed: zero catastrophic failures. Three perfect scroll kanji (道, 夢, 志 meaning way, dream, ambition), correct shadow on the shoji screen, sleeping tortoiseshell cat on the floor on a red cushion, visible candlelight reflection on the tea cup, and the most natural dual-temperature lighting of any model. The only misses were the main kanji on the paper (not quite 勇) and slightly ambiguous gender. 2 (tied). Uni-1 - 8.7/10 Uni-1 debuted in the shootout and immediately tied for second place. Three real distinct kanji on the scrolls (静, 誠, 道 meaning tranquility, sincerity, way), tabby cat sleeping curled on a red cushion on the floor, correct shadow direction on the shoji screen, and an authentic indigo patterned kimono. The calligrapher clearly reads as female. Only real miss was the main kanji on the paper not being clearly 勇.

Two years ago I started this community with a simple idea: the internet is overcomplicated and most people trying to navigate it are doing so alone. Then life happened. The agency grew. Personal stuff got in the way. And this place sat untouched. But here's what didn't stop: the online world kept getting more chaotic. AI showed up and made everything louder. And the people I work with every day — business owners, marketers, entrepreneurs — kept asking the same questions they were asking two years ago, just with more urgency. So I'm back. And I'm more focused than before. Here's what Online Simplified by MB is going to be: A place where we cut through the noise together. AI tools, digital marketing, SEO, paid media, what's actually working in 2026 — broken down in plain language, with no agenda other than helping you stay ahead. No courses to sell you right now. No pitch at the end of this post. Just consistent, useful content from someone who lives in this space every day. If you're still here after two years of silence — thank you. That says something. Drop a comment below and tell me you're still around. And tell me the one online topic you want me to tackle first. That's what I'll simplify next. -- MB