Awesome @Leo Suarez! the 5 iterations is the part nobody's going to talk about, but it's the whole thing Everyone wants the one-shot magic prompt. It doesn't exist. What exists is showing up, looking at what came back, going "nah, closer but not there yet" and pushing again. That's not a beginner move — that's exactly how people who are actually good at this work. NotebookLM is a sleeper by the way. Free, it's Google's, and it works completely different from ChatGPT or Claude. You upload your own stuff and it stays locked to YOUR content. Doesn't wander off, doesn't hallucinate random facts you never wrote. Just takes what you gave it and makes it sharper. Try the audio overview too if you haven't. It turns your docs into a full podcast conversation. Sounds gimmicky, it's not. Completely different way to hear your own ideas back. But real talk, the win here isn't the infographic. It's that you did it the same day. Didn't save it for later, didn't add it to a list. You just went and made something. That's how this skill actually builds. You're already ahead of people who've been "meaning to learn AI" for six months.

Awesome!! I did similar with my breathwork and sound therapy.

Both are in my daily stack. Honest take from someone running multiple businesses on AI right now: I reach for Claude when the output needs to sound human : content, strategy docs, anything client-facing. It also handles long, complex conversations better without losing the thread. And for building things (code, automations, apps), Claude has been consistently sharper for me. I reach for ChatGPT when I need real-time web data, image generation, or voice interaction. The ecosystem of plugins and GPTs also gives it an edge for quick integrations. The honest truth though: your context engineering skills matter more than which model you pick. Both are powerful enough to transform a business. I also use Grok, Gemini, Ollama, Perplexity, Manus (META). None of these will do much if you treat it like Google. It's a real race again and that's the best thing that could happen for everyone building with AI. Thanks for sharing @Andrew Kuhlmann, onto Day 2!

Love this concept @Aliza Juliette Bancoff, and for anyone new to this, you don't need ChatGPT specifically. This works on Claude, Gemini, Perplexity, any of them. The magic isn't the platform, it's the context you give it. Below is a simple version anyone can do right now: 1. Pick 3–5 people whose thinking you want in the room. Mix perspectives > a strategist, a creative, a numbers person, someone from your industry. 2. Open any AI chat and set the scene: "You are my advisory board. I'm going to describe a decision I'm facing. I want each of you to respond in character with your perspective." Then list each advisor with a one-line description of why they're at the table. 3. Give it a real decision you're wrestling with. Not a hypothetical, something you actually need to move on. 4. The key most people miss: push back on the responses. Say "Tony, that's too generic — what would you actually tell me to do Monday morning?" or "Kim, how would you position this for attention, specifically?" That second and third round is where it gets useful. The reason this works so well is context engineering, you're not just asking a question, you're shaping the lens the AI thinks through. The more specific you make each advisor's role, the sharper the output gets. I've been in consulting for 25+ years and honestly this mirrors how real advisory boards function you don't want five people who think the same way. You want productive tension between different frameworks. Same principle applies here. Have a great Day 2!

@Aida Olmos si necesitas ayudar me avisas. :)

Excited to be back for the 2nd AI Advantage. The noise-to-clarity problem is real and it's the #1 thing holding people back right now. Good to see it tackled head on. Coming from 25+ years of enterprise digital transformation consulting, the pattern here is familiar. Every major tech shift has a phase where the tools outpace the understanding. The people who break through are the ones who stop collecting tools and start building fluency around their own workflows. Day 2 sounds like it's going to hit the practical application side. That's where it clicks for most people. Happy to support the community with their AI questions, been building in AI for almost 8 years. Let's gooooooo, see you all there @Tony Robbins and Dean Graziosi