Scorecards disagree, so test on your own work. This prompt interviews you one question at a time about what you use AI for, what you pay now and what matters most, then builds a short, fair test you can run on two or three models in an afternoon.
You are helping me choose which AI model to use for my own work: for example Claude, ChatGPT or Gemini. Benchmarks and leaderboards disagree, so instead of naming a winner, help me run a short, fair test on my real tasks. Interview me first. Ask one question at a time, number each question, and give lettered answer choices (a, b, c...) plus "other" where it fits. Wait for my answer before asking the next one. Ask about: 1. The three tasks I use AI for most (ask me to describe each in a sentence). 2. Which AI tools and plans I pay for today, if any. 3. What matters most to me: quality of answers, speed, cost, handling very long documents or files, or privacy. 4. My monthly budget for AI, including "I only want free options." 5. Whether I use AI through a chat app, through a developer API, or both. Then build my test: - Turn each of my three tasks into one realistic test prompt I can paste into every model unchanged. - For each, give me a simple 1-to-5 scoring checklist (accuracy, usefulness, follows instructions, tone) so I judge every model the same way. - Tell me which two or three models or plans to compare, based only on what I told you. - Give me a one-page scorecard table I can fill in. Rules: 1. Do not invent prices, test scores, plan names or features. If you are not sure something is current, say so and tell me where to check (the company's pricing page). 2. Prices change. Remind me to check the date on any price, and to compare introductory prices at their full price. 3. Do not tell me one model is "the best." The winner is whichever scores highest on my scorecard. 4. Keep everything in plain English. No jargon without a one-line explanation. When I come back with my scores, add them up and end with a one-line verdict: which model to use for which of my tasks, and whether a change is worth paying for.