creative
You're going to help me turn one of my photos into something fun with an AI image tool, and write instructions good enough that the picture comes out right the first time. Rules: - Ask me one question at a time, numbered, and wait for my answer before asking the next. - When you offer choices, label them with letters (a, b, c). - Use only what I tell you, or what you can actually see in a photo I share. Never invent details about me, my job or the people in the photo. If something is unclear, ask. - Keep everything in plain English. - If the photo shows anyone besides me, remind me to get their OK before I share the result. Ask these, in this order: 1. What's in the photo? Describe it, or upload it if you can see images. 2. What do you want to turn it into? a) An aerial view, as if shot from a drone b) A collector's tin full of tiny trinkets and keepsakes c) An '80s snapshot d) A cartoon of me at work e) Something else (I'll describe it) 3. What's it for? a) Just for fun b) A social media post c) A gift or something to print d) Work (a slide, a profile picture, a team chat) 4. What must stay exactly the same? (Faces, a pet's markings, a logo, colors, anything.) 5. Should any words appear in the picture? Type them exactly, or say none. 6. Then ask the one follow-up that fits my choice: - Aerial view: what should be visible around the main subject? - Collector's tin: what are up to five keepsakes to include? - '80s snapshot: which year, and which scene? a) Mall portrait studio b) Backyard party c) Concert d) Yearbook photo - Work cartoon: what do I actually do all day, and what three things are always in my hands? - Something else: what's the one detail that would make it feel like me? Then: - Write one ready-to-paste image request, under 80 words, that keeps everything I said must stay the same. - Write a one-line caption I could post with it. - Give me a four-point checklist for looking over the result before I share it: faces, hands, any small words, and the things I said must stay the same. - End with one line telling me what to change in the request first if the picture comes back wrong.