For Those Who Can't Keep Up with AI: 10 'AI Habits' That Make a Difference. How to Go from Someone Who 'Uses' ChatGPT to Someone Who 'Masters' It

"I tried using ChatGPT. But before I knew it, I hadn't opened it for days."

Have you ever had that experience?

The reason you can't keep up with AI usage isn't because you lack talent or don't know enough about prompts.

In many cases, it's because "using AI" has become a special task.

If you have to sit down at your computer and psych yourself up by saying, "Okay, I'm going to study AI today," your habit will break the moment work gets busy.

Moreover, the world of generative AI changes so fast that while you're away, the buzz about new features and services grows, and you might feel like you "can't catch up anymore."

But is it really necessary to memorize all the latest information?

The video I referenced for this article discussed something much more fundamental.

Talk to AI using voice input.

Have AI articulate your problems for you.

Reduce aimless information gathering.

Accumulate your own context.

Have AI build tools for you.

Read AI responses quickly, think, and run other tasks simultaneously.

And, tell those around you that you like AI.

What makes a difference in the AI era is not the "number of AI tools you know," but whether using AI has become a part of your daily life.

Personally, when I use AI to create articles or images, I feel that my understanding of AI deepens more when I use it for the work or ideas right in front of me, rather than when I'm thinking, "I have to keep up with all the new features."

In this article, I have organized the 10 AI habits introduced in the video into 6 themes that are easy to start practicing today.

Even AI beginners will be fine.

Start with just 5 minutes a day, and change AI from a 'subject to study' into a 'partner to talk to every day'.

AI habits begin with 'talking to AI every day'

The first step to mastering AI is not memorizing advanced prompts.

It is to increase the number of times you talk to AI the moment a question or concern pops into your head.

The first thing introduced in the video was 'voice input'.

This may seem modest, but I think it is a very compatible method for building an AI habit.

When writing sentences on a keyboard, you unconsciously brace yourself, thinking, 'I have to write the question properly' or 'I have to explain it clearly'.

However, with voice, you can say:

'I'm thinking of doing this today, where should I start?'

'Where do you think this idea is weak?'

'I'm feeling a bit fuzzy right now, can you help me organize the reasons?'

You can start with this kind of casual feeling.

It doesn't need to be a polished piece of writing.

In fact, in conversations with AI, there is meaning in going back and forth many times rather than trying to complete the answer from the start.

Another important thing is the way of using it to 'have AI verbalize the problem itself'.

For example,

'I haven't been able to make progress on my work lately. But I don't even know the reason myself. Instead of 10, give me 5 possible causes in order of importance.'

You can ask it like that.

While looking at the candidates AI provides,

'This one is wrong'

That's pretty close.

I want to dig a little deeper into this cause.

You respond like this.

By doing so, rather than just getting answers from the AI, you end up in a state where you are organizing your own thoughts together.

This is what we call 'sounding board' interaction.

Instead of using AI like a search engine for one-off questions, you use it as a conversation partner.

If you want to start today, these three steps are enough.

1. When a question pops into your head during the day, talk to the AI about it on the spot.
2. Ask at least one follow-up question to the AI's response.
3. If the theme changes significantly—like from work to family or hobbies—start a new conversation.

You don't need to spend an hour on it from the start.

Five minutes at a time is fine.

What's important is not 'securing time to use AI,' but creating a reflex to 'try talking to the AI whenever you think of something.'

When studying English, it's easier to maintain your sense for the language by touching it a little every day rather than just spending a few hours on the weekend.

AI is the same; it becomes easier to continue if you think about increasing the frequency of your daily contact with it.

Latest Don't follow every AI. Gather information by working backward from your goals.

To keep using AI, you need not only to 'increase information' but also to 'reduce information.'

If you try to research every new AI that comes out, you'll get exhausted before you even use it.

Open X and there's a new model.

Open YouTube and you'll see 'This AI is amazing'.

A few days later, a different service becomes the talk of the town.

In this situation, thinking 'I'll use it once I understand everything' is quite difficult.

A key point mentioned in videos as well was the mindset of 'stopping information gathering without a purpose'.

For example, if you want to create a website, you should look up 'how to create a website with AI' at that exact moment.

If you want to create a document, look up the AI needed for document creation.

If you want to create an image, look up image generation.

The purpose comes first, and the tool comes second.

Do not reverse this order.

I think you can let go of the idea that 'someone who can use AI is someone who knows all the latest AI'.

What is important is being able to use AI when you want to achieve something.

And, an AI habit that takes this a step further is 'collecting context'.

It is easy to understand context if you think of it as background information that helps AI understand your work or purpose.

For example, if you are writing an article,

・ Who is the reader?

・ What kind of articles have you written in the past?

・ What kind of language do you use?

・ What do you value?

・ What kind of structure do you prefer?

・ What didn't work well in the past?

As this information increases, the need to explain everything to the AI from scratch each time decreases.

Even with the same request to 'write an article,' the direction of the output will naturally change between an AI that has almost no information about you and an AI that shares your past articles, reader profile, purpose, and prohibitions.

As you use AI more, the perspective of 'what should I have the AI know' becomes more important than finding a magic sentence.

Therefore, there is value in gradually organizing ideas that come up during work, meeting notes, frequently used text, and decision-making criteria.

However, when inputting personal information, confidential information, or third-party conversations into an AI service, do not forget to check the service's terms of use and your organization's rules.

Instead of 'saving everything,' think about 'what information will be useful if I give it to AI in the future.'

Even just doing this will significantly change how you use AI.

Don't just have AI provide answers; have it build tools specifically for you

The next stage of AI utilization is moving from 'asking questions' to 'creating systems.'

If you are doing the same task every time, think about whether you can leave that task itself to AI.

For example,

Summarizing text in the same format every week.

Extracting necessary items from a large amount of information.

Checking the same data every morning.

Creating similar email drafts every time.

For tasks like these, just asking AI 'How can I do this faster?' will yield improvement suggestions.

Furthermore, nowadays, by using AI that supports code generation, the hurdle for creating your own small programs or tools has become lower than before.

What is important here is not to try to build a grand system from the start.

'I don't know what I should build with AI.'

I think there are many people who feel this way.

At times like that, ask AI exactly that.

'Identify the repetitive tasks in my work in a question format. From those, choose three that could be streamlined with AI or simple tools.'

This is all you need to do.

For example, in my case, while continuing to produce articles,

Generating title ideas.

Organize the structure.

Extract key points from long texts.

Check for inconsistencies in writing style.

Convert into short versions for social media.

These are some of the processes involved.

You don't need to fully automate everything at once.

If even one step becomes faster, you can use that time for other work.

What you need to remember here is that 'creating AI tools' does not mean 'becoming a programmer'.

The goal is not to write code, but to reduce your tedious tasks.

There is a question I want you to try today.

'Among the tasks I did today, which ones are highly likely to be repeated tomorrow?'

Try thinking about this for just 3 minutes at the end of the day.

And once you find one,

'Suggest 3 ways to semi-automate this task, ordered from easiest for a beginner to most advanced.'

Try asking that.

If you continue this small habit, AI will change from a 'tool for creating text' to a 'tool for improving the work itself'.

Asking AI not just to generate answers every time, but to help you think of 'systems that eliminate the need to generate answers in the future' will take your AI usage to a deeper level.

What is needed in the AI era is not speed reading, but the 'ability to rethink at high speed'.

When you start using AI a lot, new problems arise.

There are so many AI responses that you run out of time just reading them.

A few hundred characters is no problem.

However, when you ask AI to handle research, planning, documentation, or analysis, it can return thousands of characters of text in an instant.

Furthermore, if you ask multiple AIs questions at the same time, the amount you have to read increases all at once.

That is why 'speed reading training' was introduced in the video.

However, this is not about 'reading at high speed while perfectly understanding every single sentence'.

For themes you are already somewhat familiar with,

What is it written about?

Where are the important keywords?

What parts are unknown to me?

What should I ask next?

Read with the intention of quickly picking up these four things.

For example, if it is a theme you are knowledgeable about, there is no need to read the AI's response from beginning to end, character by character.

Focus on headings, conclusions, numbers, proper nouns, and parts that feel off.

On the other hand, for fields you know almost nothing about, check them carefully rather than reading quickly.

This switching is important.

And another interesting habit is the 'AI puzzle'.

To simplify the concept discussed in the video,

Concrete → Abstract → Another Concrete

It is practice in converting to this.

For example, suppose a certain popular product is selling well.

So,

"Organize the reasons why this product is selling, not as product-specific elements, but as a reusable structure."

I ask the AI.

Suppose,

・ The difference is clear at a glance

・ It makes you want to tell others about it

・ The experience changes the moment you use it

a structure like this is produced.

Next,

"If you were to apply these three elements to a note article, what would that look like?"

I ask.

Then, you can repurpose the success factors of product marketing for article creation.

This is a powerful way to increase your ideas.

Even between seemingly completely different genres like education, soccer, work, social media, and product development, if you extract the "structure," there is a possibility of applying it.

You don't just adopt the AI's answer as is.

Think about the "why?" together with the AI.

Instead of dumping your thinking onto the AI, use the AI to increase the number of times you think for yourself. Used this way, the AI is not an entity that steals your thinking, but a partner that expands it.

A common pitfall is thinking that the first answer is the finished product.

AI answers can contain errors or oversights.

Therefore,

"Is that really true?"

"What are the opposing views?"

"What are the weaknesses in this explanation?"

"What does it look like from a different perspective?"

Respond with these.

If you think of conversations with AI as a 'thought rally' rather than 'checking for the right answer,' the range of how you use it will expand.

One by oneFrom the era of using AI to the era of advancing multiple tasks simultaneously

Once you get used to using AI, think about reducing the time you spend waiting for AI responses.

The video also introduced advanced concepts called 'parallel execution' and 'ambient agents'.

Parallel execution, simply put, is running multiple AI tasks at the same time.

For example, when creating an article,

Have AI A organize the key points of competitor articles.

Have AI B list the readers' pain points.

Have AI C generate title ideas.

Meanwhile, you think about the article structure.

In this way, instead of waiting for one task to finish before moving to the next, you run tasks that can proceed independently at the same time.

However, beginners do not need to run 10 tasks at once right away.

Starting with two is enough.

For example,

1. Have an AI summarize a long text
2. While that is processing, ask another AI for title ideas

Start with this level.

As you get used to it, your speed in deciding 'I will do this' versus 'I will hand this over to AI' will increase.

Beyond that lies a system where AI and automation handle tasks once the necessary conditions are met.

Gathering specific information every morning.

Categorizing incoming information.

Summarizing only what is necessary.

Completing routine preparatory work.

Once such workflows are in place, humans can shift their role from 'the person who presses the start button every time' to 'the person who reviews the results and makes decisions'.

However, it does not mean that everything should be fully automated.

There are tasks that require human verification, such as money, contracts, public documents, and important customer correspondence.

First, it is safer to start with tasks that:

・ Do not cause major problems if they fail ・ Are repeated frequently ・ Have relatively clear judgment criteria ・ Can be verified by a human at the end

It is safer to try with tasks like these.

When handing work over to AI, what is needed is not 'do everything,' but rather:

What exactly to delegate.

What you want checked.

At what point it should be returned to a human.

Designing up to this point is key.

Productivity in the AI era is not determined solely by typing speed.

Once you can run 'AI time' separately from your own time, the amount of work you can get done in the same hour will change.

When you tell people you like AI, your AI skills will quickly turn into 'practice'

The final AI habit is surprisingly analog.

'Tell people around you that you like AI'.

In the video, this habit was described as being extremely important.

I think this is an interesting perspective as well.

When you study AI alone, you end up solving only the problems you are interested in.

However, if you tell people around you,

'I've been using AI quite a lot lately'

they might say,

'Can't this be done with AI?'

'Can't this part of my work be made easier?'

'Can you build a service like this?'

and you may find yourself gathering challenges you wouldn't have thought of yourself.

And when you try to solve other people's problems with AI, your learning suddenly turns into practice.

For example, suppose a friend consults you, saying,

'It's a hassle to create documents in the same format every month.'

Then you use AI to look for improvements.

You actually try them out.

You fail.

You change your prompt.

You try again.

Try it again.

Once you have this much experience, you will be left with a much deeper understanding than just the knowledge that "AI can create documents."

Furthermore, explaining it to others is a learning process in itself.

Even if you thought you understood it yourself, when you try to explain it to someone else,

"I can't explain this part properly."

You will find parts like that.

Research those parts with AI again.

Understand them.

Explain them again.

This cycle will strengthen your AI utilization.

That is why I think those considering an AI side hustle don't need to overthink "What should I sell with AI?" from the start.

First, use AI for your own work or hobbies.

Next, listen to the problems of people around you.

If it seems like AI can solve them, give it a try.

Put that experience into words.

Following this order will cultivate a sense of "what AI is effective for," rather than just theoretical knowledge.

So, do you need to start all 10 of these today?

Of course not.

I recommend a method of gradually building habits over 30 days.

For the first 7 days, just talk to AI for 5 minutes every day.

For the next 7 days, dig deep into one work or life problem with AI.

For the next 7 days, try to streamline one repetitive task using AI.

For the final 9 days, try teaching one AI usage method to someone else.

This alone is enough.

The shortest path to mastering AI is not to keep learning about AI, but to increase the reasons for using AI in your own life.

Thinking 'I have to study AI' makes it feel like a burden.

But,

Consult about tonight's dinner.

Consult about a travel plan.

Get text corrected.

Ask for ideas.

Organize your worries.

Make one task easier.

You can start from things like that.

Perhaps it is only when AI is no longer a special existence that we can truly say we have developed an AI habit.

Conclusion | AI habits create the difference in the future

The faster generative AI evolves, the easier it is to feel anxious, thinking 'I have to memorize all the latest information.'

However, the essence that emerges from these 10 AI habits is the opposite.

Before chasing tools, talking to AI, thinking, delegating, creating systems, and teaching others—building this foundation will be more useful in the long run.

1. Key points

Voice input → verbalizing worries → gathering only necessary information → preserving context → small-scale automation → moving toward parallel processing with your thoughts. Incorporate AI into your daily life in this order.

1. Today's small step

Today, just once, please speak or write to an AI like ChatGPT about 'what you are currently struggling with the most.' And don't stop at the first answer; try asking one follow-up question.

1. Author's perspective

I believe that as AI continues to evolve, the value of 'what you have achieved with AI' will become greater than 'which AI you know'.

What will make a big difference in six months might not be the knowledge you learned today, but the small AI habits you started today.

You don't need to wait until you can use it perfectly to start.

You can learn while you use it.

Transform AI from a 'convenient tool you use occasionally' into a 'partner that is naturally by your side when you think'.

That change begins with just one question today.

【Reference/External information links】

*These are English links, but on iPhone, you can display them in Japanese by selecting 'AA' -> 'Translate' in Safari.

Reference: How to use ChatGPT and prompts (Search term: ChatGPT prompting)

・ Prompting (OpenAI Academy / Official)

・ Prompt engineering best practices for ChatGPT (OpenAI / Official)

Reference: Context to provide to AI (Search term: AI context engineering)

・ Effective context engineering for AI agents (Anthropic / Official)

Reference: AI agents and automation (Search term: AI agents workflow)

・ A practical guide to building agents (OpenAI / Official)

[Hashtags for this article]

#AIUtilization

#AISideHustle

#ChatGPT

#ChatGPTSideHustle

#LawOfAttraction

#note

#OrganizingLife

#AIHabits

#GenerativeAI

#AISage

#AISpecialist

🏆Popular Article Ranking**BEST 10(For first-time readers)**

(*For those who aren't sure which article to read first. We have compiled popular articles chosen by many readers in a ranking format.)

⭐️Next recommendation: Self-introduction

This is the reason why I started writing on note.

"I should have been living properly, so why is it painful?"

I wrote this for those who have felt that way.

📝A final word

Thank you for reading this far.