The Simplest Way to Work Through a Problem with AI
How to use ChatGPT, or another AI tool, to untangle an ordinary problem even when you do not know how to explain it yet.
You have a problem, but it does not arrive neatly.
It might be an email you do not know how to answer, a document you do not fully understand, a screenshot of something confusing, a message you are unsure how to reply to, a decision you keep avoiding, or a situation with too many moving parts.
So you open ChatGPT, or another AI tool, and immediately hit the next problem:
How do I even explain this?
This is where many beginners get stuck. They think they need to understand the problem clearly before they can ask AI for help.
But often, the conversation is what helps you understand the problem.
You do not need the perfect prompt. You can start with the messy version. You can paste the email, upload the document, show the screenshot, or simply explain the situation badly and ask AI to help you make sense of it.
The simplest way to work through an ordinary problem with AI is not to sound clever. It is to keep the conversation going until the problem becomes clearer.
The mindset: talk to AI like a helpful person, not a perfect machine
Before we get into the steps, it helps to change how you think about the chat box.
Many beginners treat AI like a search engine or a command box. They try to type the perfect instruction, then wait for the perfect answer.
That makes AI feel harder than it needs to be.
A better way is to treat it like a conversation with a capable helper who knows nothing about your situation yet.
You would not walk up to a helpful person and say only, “Fix this.” You would explain what happened, what feels hard, what you are trying to do, and where you feel stuck. If they misunderstood, you would correct them. If they asked a question, you would answer it. If their first suggestion was not quite right, you would keep talking.
That is the mindset for using AI to work through an ordinary problem.
You are not trying to perform intelligence. You are not trying to write a perfect prompt. You are simply explaining the situation in plain language and letting the conversation help you find the next useful step.
AI is not responsible for your life, and it should not replace your judgement. But it can be very useful as a thinking partner when you talk to it naturally, keep steering, and check what matters.
The simple process
You can think of the process like this:
Show the mess.
Make the request clearer.
Let AI ask questions.
Sort the parts.
Look at options.
Check weak spots.
Choose one next step.
That may look like several steps, but it is really just the natural shape of a useful conversation.
You show what is going on. AI helps you make sense of it. You answer a few questions. The problem becomes less foggy. Then you decide what to do next.
This guide is for ordinary problems, not serious professional decisions. If the issue involves legal, medical, financial, tax, immigration, employment, safety, crisis, or anything where being wrong could cause real harm, use AI to prepare, organise your thoughts, and ask better questions. Do not treat it as the final authority.
For everyday problems, though, this simple conversation can take you a long way.
1. Show the mess
The first step is not to explain the problem perfectly.
The first step is to bring AI what you have.
That might be a rough explanation, a pasted email, a screenshot, a document, a form, a few notes, or a message thread. Where your AI tool allows uploads, you can use that instead of typing everything out.
This is useful because many people do not know what matters yet. They only know something is confusing, awkward, annoying, unclear, or hard to deal with.
Before you paste or upload anything, pause for privacy. Remove private details where you can. Replace names, addresses, account numbers, client details, medical details, financial information, passwords, and anything sensitive with simple placeholders.
You do not need to write a beautiful explanation. You can start with something like this:
I have a problem and I do not know how to explain it clearly yet. Here is the messy version, or the thing I am looking at: [type, paste, or upload it here]. First, tell me what you think is going on. Do not solve it yet.
For a document, email, or screenshot, you can also say:
I do not know what matters here. Please help me understand what I am looking at before giving advice.
That line is useful because it takes the pressure off you. You do not need to know the correct question before you start. You can ask AI to help you see what the question might be.
2. Ask AI to turn the mess into a clearer request
Once AI has seen the messy version, do not rush straight to the answer.
Ask it to turn the mess into a clearer request first.
This is not prompt engineering. It is just tidying the question before trying to answer it.
This step is especially helpful for beginners because it shows you what you are really asking. Sometimes the problem you thought you had is not the real problem. Sometimes you are asking for a reply, but what you actually need first is to understand the situation. Sometimes you are asking for a decision, but what you really need is to sort your options.
Use this:
Turn what I just gave you into a clearer request I can review. Include the situation, what seems hard, what I might want, and what information is still missing. Do not answer it yet.
Then read what AI gives back.
This part matters. Do not skip it.
If AI has misunderstood the situation, correct it. If it has made a wrong assumption, say so. If it has missed something important, add it.
You are still in charge of the conversation. AI is helping you shape the request, not taking ownership of the problem.
You can also ask for a stronger version:
Before you answer, rewrite my messy explanation into a clearer prompt. Show it to me first, then ask me if anything is wrong or missing.
This is a simple way to get a better prompt without needing to know how to write one yourself.
3. Let AI ask questions first
A lot of AI answers are generic because the tool answers too quickly.
For ordinary problems, it often helps to slow the conversation down. Instead of asking AI to solve everything immediately, ask it to ask questions first.
This is easier for beginners because you do not have to explain everything in one go. You only need to answer the next question.
Try:
Ask me the three most important questions before giving advice.
Or, if you feel overwhelmed:
Ask me one question at a time, in plain language. Start with the easiest question.
This works because AI does not know your life unless you tell it. The questions help pull out the details that matter.
For example, if you paste an awkward email, AI may ask who the person is, what your relationship is like, whether you want to sound warm or firm, and what outcome you want. Those are useful questions. They turn a vague problem into a clearer one.
You can also tell AI how to behave if you do not want too many questions:
Only ask the questions that would genuinely change your answer.
That keeps the conversation useful without making it feel like homework.
4. Sort the problem into parts
Many ordinary problems feel hard because they are not one problem.
They are a mix of facts, feelings, timing, people, uncertainty, responsibility, and possible consequences. When all of that sits in one pile, the problem feels bigger than it is.
Ask AI to separate the pile.
Use this:
Separate this problem into parts: what is practical, what is emotional, what is uncertain, what I can control, what I cannot control, and what decision actually needs to be made.
For an even simpler version, ask:
Break this into smaller pieces so it feels less overwhelming.
A useful extra question is:
What is the real problem underneath all of this?
This step can bring a lot of relief. The problem may still be real, but it is no longer one large cloud.
For example, you may think the problem is, “I need to reply to this message.” But after sorting it, you may realise there are several parts underneath: you are annoyed, you do not want to damage the relationship, you need to set a boundary, and you are unsure what tone to use.
That is already progress.
AI has not solved the problem for you. It has helped you see the shape of it.
5. Ask for options, not orders
This is important.
Do not ask AI to decide for you. Ask AI to show you your options.
There is a big difference between:
What should I do?
and:
What are my realistic options?
The second version keeps you in charge.
Try:
Give me three realistic options. For each one, explain the benefit, the risk, what I would need to check, and when that option would make sense.
You can also ask:
Which option is the simplest low-risk first move?
Or:
Which option would you avoid, and why?
This helps AI become a thinking partner instead of a decision-maker.
For ordinary problems, you usually do not need ten options. Too many choices can make the problem feel heavier. Two or three realistic options are usually enough.
You can also ask AI to label the options by style:
Give me one gentle option, one direct option, and one middle option.
That is useful for emails, conversations, family situations, small work problems, and awkward replies.
You are not asking AI to control your life. You are asking it to help you see the paths in front of you.
6. Check what might be missing
A calm AI answer can still be too simple, too confident, or based on assumptions that are not true.
Before you rely on the answer, pause and ask what might be missing.
Use this:
Before I rely on this, what could be wrong, missing, or too simplistic about your answer?
Another useful version is:
What are you assuming about my situation that might not be true?
And before taking action:
What should I check before I act on this?
This step does not need to become heavy. You are not trying to doubt everything forever. You are simply checking the part that matters before you use the answer.
For example, if AI helps you draft a message, you might ask:
How might this message be received badly, even if I mean it well?
If AI helps you understand a document, you might ask:
What parts should I check against the original before I rely on your summary?
If AI helps you compare options, you might ask:
What important detail would change your recommendation?
These questions keep your judgement awake.
AI can help you think more clearly, but you are still the person who knows the real situation, the people involved, and the consequences.
7. Choose one next step
Do not end the conversation with a giant plan.
For most ordinary problems, the best ending is one small next step.
Ask:
Based on everything we have discussed, what is the smallest useful next step I can take today? Also tell me what I should not decide yet.
Or even more simply:
Give me one next step, not a full plan.
This matters because AI can easily create too much work. A beginner does not need a 20-step strategy for every problem. They need the next clear thing.
You can also ask:
What can wait until later?
That question is often just as useful as asking what to do.
Sometimes the next step is to send a short reply. Sometimes it is to check one fact. Sometimes it is to ask one person a question. Sometimes it is to sleep on the decision because the conversation showed you that you are tired, rushed, or missing information.
A good AI conversation does not always end with a dramatic solution.
Often, it ends with the next sensible move.
That is enough.
A simple example
Imagine you receive an email that feels awkward.
You do not know whether the person is annoyed, whether you need to apologise, whether you are overthinking it, or whether you should reply warmly or firmly.
You could start by pasting the email and saying:
I received this email and I am not sure how to read it. Please help me understand the tone and what the person might be asking for. Do not write a reply yet.
Then you could ask:
Turn this into a clearer request I can review before you answer.
AI might turn your messy concern into something like:
The user wants to understand the tone and meaning of an email, work out whether a reply is needed, and decide how to respond without sounding defensive or too formal.
You can then correct it if needed.
Next, ask:
Ask me three questions that would help you understand the situation better.
After answering, ask:
Separate the problem into parts: what I need to respond to, what I might be assuming, what tone I should use, and what I should avoid saying.
Then ask for options:
Give me two possible reply options: one warmer and one more direct. Tell me the risk of each.
Before sending, check:
What might I be missing before I send this?
Then finish with:
Give me the simplest next step.
That is how an ordinary problem becomes easier. Not because AI magically knows everything, but because the conversation helps you slow down, sort the parts, and choose what to do next.
What not to use this for blindly
AI can help you organise your thoughts, understand a document, prepare a message, compare options, notice assumptions, and work out what to check next.
But it should not be treated as the final authority for serious matters.
Be careful with legal, medical, financial, tax, immigration, employment, safety, crisis, or professional decisions. In those cases, AI can help you prepare, but a qualified person or reliable source still matters.
A simple rule is:
Use AI to think more clearly. Do not use it to hand over responsibility.
This boundary is not there to make AI scary. It is there to keep AI in the right place.
AI can be a useful helper. It should not become the person responsible for your life.
Try this now
Paste this into ChatGPT, or another AI tool:
I have an ordinary problem I want to work through, but I do not know how to explain it clearly yet. Here is the messy version, or the thing I am looking at: [type, paste, or upload it here].
Please do not jump straight to advice. First, tell me what you think the problem is in plain English. Then turn it into a clearer request I can review. After that, ask me a few simple questions, one at a time. Once the problem is clearer, help me separate it into parts, show me two or three realistic options, tell me what might be missing, and help me choose one small next step.
You can use that prompt with an email, a screenshot, a document, a rough note, a message, or just a messy explanation of what is going on.
Before pasting or uploading anything, remove private details where possible.
The calm takeaway
You do not need to solve the whole problem before you ask AI for help. You do not need to write a perfect prompt, and you do not even need to know exactly what the problem is yet.
Start with the messy version. Upload or paste the thing in front of you if that is easier. Ask AI to help you understand it, turn it into a clearer request, ask questions, sort the parts, show options, check what might be missing, and choose one small next step.
That is the simple way to work through a problem with AI: not by sounding clever, not by becoming technical, but by talking naturally and staying in the conversation.
If this helped, save the guide and try the prompt next time a problem feels too messy to explain.
And if you want the fuller beginner-friendly path, you can find the Kindle series here: