What happened
A six-prompt chain that turns a chatbot into a one-on-one tutor is doing the rounds in the Telegram channel @GPThelp_ru, and while the raw numbers are small — 1,172 views, 17 reposts, 8 reactions — the format is worth a closer look. This isn't another “10 ChatGPT hacks” listicle. It's a repeatable method, and structured AI prompts for learning are turning out to be one of the few genuinely dependable ways to use these tools.
The rules are refreshingly simple. You drop your topic into the first prompt and the other five adapt themselves to whatever you typed. You send them one at a time, in a single chat, and you wait for each answer before sending the next one. No plugins, no paid tier required, no code.
The six prompts, in plain English
Here's the whole chain, written the way a native speaker would phrase it — copy it as-is:
1. Set the frame. “I want to get up to speed on [topic]. Before you explain anything, ask me three short clarifying questions: why I want this topic, what I already know about it, and what result 30 days from now would feel useful. Wait for my answers and don't explain anything yet.”
2. Get the map. “Given my answers: you're an experienced teacher in this field. Give me a structured overview. Focus on the 20% of concepts that explain 80% of the subject. One short paragraph per concept, plain language.”
3. Clear out the junk. “List the 5 most common misconceptions beginners have about this topic. For each one, say what people get wrong and what's actually true.”
4. Slow down on the hard parts. “Take the 3 hardest concepts from your overview and explain each one again as if I know nothing about the field. Use one everyday analogy per concept.”
5. Make it real. “Now the practical part. Build a step-by-step plan for applying this over the next 30 days, with one concrete action per week.”
6. Check your work. “Test me. Ask 10 questions covering everything above, one at a time. After each answer, tell me whether I was right, and correct me if I wasn't.”
That's the whole thing. Six messages and you've got a syllabus, a misconception filter, plain-English explanations, a 30-day action plan and a quiz.
Why the order does the heavy lifting
Most people prompt an AI like a search engine: one question, one answer, done. This chain does something different — it builds context before it builds knowledge.
Step 1 forces you to say out loud what you actually want. That single move changes the quality of everything that follows, because the model stops guessing who you are. Steps 2 and 3 give you the shape of the field and the traps, in that order. Step 4 is where real learning happens: a good analogy does more work than three paragraphs of jargon. Step 6 leans on active recall — answering questions from memory instead of re-reading — which is the most reliable study technique we know of.
What it means for you
At home: the skill you keep postponing
Maybe you want to finally understand your home Wi-Fi setup, cook proper Thai food, or figure out what your pension statement actually says. Prompt 2 gives you the 20% of concepts that carry the other 80%. Prompt 4 hands you the analogy that makes it click. You can be done in an evening.
At work: walking into a meeting you're not ready for
You've got a call about a topic you only half understand. Run steps 1 through 3 the day before. You'll walk in knowing the vocabulary, the standard arguments and the three things newcomers usually get wrong — which is most of what sounding competent actually requires.
In business: market research without hiring a consultant
Founders use this chain to map a new niche: the core concepts, the rookie mistakes, then a 30-day validation plan. It won't replace real customer conversations, but it stops you from burning three weeks learning the basics the slow way. If you'd rather not pay for a subscription, a directory of free AI tools is a decent place to start.
Studying: exam prep that actually sticks
Prompt 6 is a private tutor that never gets bored of quizzing you. Ten questions per round, immediate feedback, corrections on the spot. Run it three times a week for a month and you'll cover more ground than re-reading your notes ever would.
Creativity: picking up a craft or a medium
Music theory, video editing, watercolour, screenwriting — whatever it is. Ask for the 20% overview, then the analogies, then a 30-day practice plan with one deliverable per week. You finish with a portfolio, not just a vague sense of having learned something.
Income: testing a side-hustle skill in 30 days
Pick something people pay for — copywriting, bookkeeping, no-code automation. Run the full chain. By the end of the 30-day plan you'll know whether the skill fits you, and you'll have four weeks of practice behind you either way.
How to try it right now
You can run the whole thing tonight, for free.
1. Open ChatGPT. The free tier is enough. Go to chatgpt.com and sign in — or use any other AI chatbot you already have.
2. Start one chat and stay in it. Every prompt goes into the same thread. Open a new chat and the AI loses the context from steps 1–5, and the quality falls off a cliff.
3. Paste prompt 1 and swap in your topic. Replace [topic] with something specific. “Organic chemistry” works far better than “science.”
4. Answer the three questions honestly. Your goal and your current level are the two things the AI genuinely can't guess.
5. Send prompts 2 through 6, one at a time. Wait for each answer. If an explanation is too dense, just type “simpler, please, with a shorter example.”
6. Do the quiz. Don't skip it. That's the step that turns reading into remembering.
If you want to go deeper, these are worth your time: Sogeti Labs on using AI to learn faster, a hands-on walkthrough on YouTube, Meta's own guide on how to study with AI, and MIT Sloan EdTech's notes on effective prompts.
Upsides and what changes
The upside is leverage. Instead of paying for a course you'll abandon in week two, you get a syllabus shaped around your own goal, in about twenty minutes. Because the chain checks your understanding at the end, it also builds in something most self-study lacks: proof that you actually learned it.
What changes is your default move when you hit an unfamiliar subject. Right now it's probably “I'll Google it later.” After a few runs, it becomes “I'll run the six prompts.” That shift matters more than any individual topic, because it means the cost of learning something new drops to almost nothing. Independent sources back the underlying approach — structured, multi-step prompting with AI as a tutor is a well-supported technique in prompt engineering for education, even though this exact six-prompt set isn't a formally recognised methodology.
Limitations
Be honest about what this can't do. The AI can be confidently wrong, especially on niche, fast-moving or highly technical subjects, so treat every factual claim as a draft you still need to verify against a trustworthy source — a textbook, a primary document, a practitioner. Six prompts also won't turn you into an expert in a month; they'll take you from zero to conversational, which is usually the hard part anyway. And a chatbot can't give you feedback on a physical skill, a live negotiation or a real client project. Use it to get oriented, then go do the thing in the real world, because that's where the learning actually sticks.
Conclusion
Six prompts, one chat window, thirty minutes of your evening. That's the entire barrier to entry here, and the payoff is a study method you can point at literally any topic for the rest of your life.
The one action for today: open ChatGPT, paste prompt 1, and replace [topic] with the thing you've been meaning to learn for the last six months. Answer the three questions, send prompt 2, and see what comes back. If the output is useful, finish the chain. If it isn't, you've lost ten minutes — and you now know exactly how to phrase the next attempt.
Comments 0