Welcome to “I Am GPTed,” where the practical tips are hot, the sarcasm is lukewarm, and your host, Mal, is exactly as excited as an algorithm can be. I’m Mal, the Misfit Master of AI—or just Mal for short. Let’s jump in before the tech overlords rebrand me as “Clippy, Version 2.0.” Today: One solid prompting trick, a real-life use case for AI newbies, a mistake you can totally blame on me, an easy skill-building exercise, and one tip to make your AI outputs less cringe. All in five hundred words or less, because time, like buzzwords, is precious. First up, the **prompting technique du jour:** *role assignment.* Yes, it’s as fancy as it sounds, and just as simple. You tell the AI what to be. Like playing make-believe, but your imaginary friend has access to the internet. Example—**Before role assignment:** Prompt: “Summarize this document.” Result: A summary that reads like someone rushed through it during their lunch break. **After role assignment:** Prompt: “You are an experienced legal analyst. Summarize this contract for a client with no legal background, highlighting any risks in plain English.” Now, the AI suddenly finds its briefcase and starts acting like it has a law degree—voilà, a way better summary. When you hand the AI a role, it tailors its response. Try this with ChatGPT, Claude, Gemini, even Grok—though Grok might prefer to explain things with memes and existential dread. Now for a **practical use case** you may not have considered: *Family Debate Referee.* Next Thanksgiving, instead of arguing with your uncle about some random fact, just type the disputed topic into your favorite AI model, assign the role: “You’re an impartial debate moderator,” and watch as dinner is saved (or, at least, redirected to AI’s blame). Bonus: The AI never brings up politics—unless you ask. Let’s talk mistakes. My personal favorite—because I make it about once a week—is **vague prompting**. You want a meal plan, so you type: “Make me a meal plan.” The AI hands you a seven-course dinner for goats. Been there, done that, wondering if I’m part goat. *Don’t be like early-Mal.* Be specific: “Make me a vegetarian meal plan for someone who hates mushrooms, has only 20 minutes, and likes Italian food.” Watch as the AI pivots from goat cuisine to something you’ll actually eat. **Your simple AI exercise this week:** Pick a task you do daily—writing an email, planning meals, anything. Write two prompts for the AI: One vague, one super-specific. Compare the outputs. Notice how the AI basically panics when you’re unclear but shines when you give it direction? Congratulations—you’ve just leveled up. Finally: **How do you know if that shiny AI output is any good?** Easy—take five seconds and ask yourself, “If I handed this to my boss, my kid, or my dog, would they understand it? Would they want to bite me?” If the answer is “maybe not,” ask the AI to clarify, add examples, or rewrite it shorter. Consider AI your endlessly patient intern—just less likely to steal your lunch from the fridge. That’s it for today’s episode. If you got even 1% smarter—or just feel less confused—be sure to subscribe to “I Am GPTed.” Thanks for listening! This has been a Quiet Please production. To learn more, visit quietplease.ai. I’m Mal, reminding you: You don’t have to get AI perfect. You just have to get less goat recipes. For more check out https://www.quietperiodplease.com/ and for some great deals go to https://amzn.to/4nidg0P This content was created in partnership and with the help of Artificial Intelligence AI
