Generic AI answers fail because they can't know what you specifically misunderstand, but detailed prompts—naming what you've already tried, what confused you, what gap you're trying to fill—produce answers actually calibrated to your learning needs. Specificity in asking directly increases usefulness of answers.
Prompt engineering sounds fancy, but it's really just the art of asking AI questions in ways that get you useful answers. Think of it like asking your professor for help—a vague question gets a vague answer, but a specific one gets a detailed explanation.
When you ask an AI "Can you explain economics?" you'll get a generic overview that might not match what your professor emphasized in lecture. But if you ask "Can you explain how supply and demand affects housing prices, using examples from the lecture notes I'm attaching?" you get something actually useful for your exam.
AI language models (the technology behind ChatGPT, Claude, etc.) process context clues. The more specific you are about what you need, the more the AI can tailor its response to match your situation. It's like the difference between asking a librarian "Do you have books?" versus "Do you have books on 19th century American history that focus on the Industrial Revolution?"
Good prompts include four key elements: the task ("Explain this concept"), the context ("from my biology lecture"), the format you want ("as a study guide with bullet points"), and any constraints ("in 5 minutes of reading time").
The biggest one? Assuming the AI knows what your professor cares about. Your AI tutor hasn't been to your lectures. It doesn't know which topics your professor spent three days on versus five minutes. When you're asking for help understanding material, include that context.
Another mistake is treating AI like a search engine. You don't need formal language—actually, conversational prompts often work better. "Hey, I'm confused about photosynthesis and my exam is Friday" will get you better results than "Provide a comprehensive overview of photosynthesis."
Start with what you need: "I need help studying for..." Follow with context: "My professor emphasized..." Then specify format: "Can you give me this as..." Finally, add any special instructions: "Use examples from..." or "Explain like I'm..."
The goal isn't to write a perfect prompt on the first try. It's to iterate. If the first answer isn't quite right, ask a follow-up question that refines what you're looking for. "That's helpful, but can you focus more on X and less on Y?" This back-and-forth conversation is where AI becomes truly useful.
Try this: Take one confusing concept from your current coursework. Write two prompts for an AI tool—one vague (like you might naturally ask), and one specific with context and format details. Compare the answers and notice the difference in usefulness. Start using the specific format for all study questions.
Peri can explain this concept, give practical examples, help you decide whether it applies to your situation, or recommend a journey if appropriate.
Explore related journeys or tell Peri what you're working through.