Relax. Everyone should use AI.

Something that genuinely triggers me is when someone points out that I use AI for my work, as if they have exposed some terrible secret.

Yes, I use AI

Why would I manually spend hours doing something that a tool can help me complete faster?

Why should I walk for ten kilometres if I have a car?

Using the car does not mean I have forgotten how to walk. It means I understand that my time and energy have value.

The same applies to AI.

I use it to research, organize ideas, test approaches, write code, improve formatting, and get through repetitive work faster. I still decide where I am going. I still need to know whether I reached the right destination.

What exactly is the problem?

If you want to criticize someone for using AI, bring receipts. What is actually wrong with the result?

Point out the real problem. Simply saying, “You used AI,” tells me nothing about the quality of the work.

If the result is accurate, clear, well researched, properly formatted, and useful, why should the tool matter more than the outcome?

We already use tools for everything

Nobody attacks an accountant for using a calculator. Nobody tells a photographer that editing software makes the photograph fake. Nobody tells a developer to stop using libraries and write every function from the beginning.

Nobody demands that writers abandon spell-check and return to dictionaries. Nobody asks designers to stop using modern software and draw everything by hand.

We use tools because they help us work faster and improve what we produce.

AI is not magic.AI is another tool.

It is more advanced than a calculator or spell-check, of course. It can produce entire pages, working code, realistic images, and confident explanations. That means it requires more attention and responsibility, not complete rejection.

Books did not disappear because of the internet

When the internet became popular, people worried that it would destroy books.

It did not.

Books still exist. Libraries still exist. Researchers still read original documents. The internet simply made information easier to find and share.

The problem was never the internet itself. The problem was unreliable websites, misinformation, plagiarism, and people believing everything they read.

We learned that finding information online is not enough. We also have to evaluate the source.

AI requires the same habit. Getting an answer is not the same as getting the truth.

Read it. Question it. Verify the claims that matter.

Electricity did not eliminate candles

Electricity replaced candles as the main source of light because it was more practical. That did not make candles useless. We still keep them for decoration, tradition, relaxation, and emergencies.

Should we reject electricity because candles still work during a power outage?

Of course not.

We use the better tool for daily life while keeping older methods available when they are needed. The same should happen with AI.

People should continue learning how to write, research, calculate, code, and think independently. Those abilities remain important. But keeping traditional skills does not mean we must avoid modern tools.

Using AI is not the same as using it badly

Bad AI use is copying the first answer without reading it. It is publishing invented facts, fake quotations, broken citations, insecure code, and awkward writing. It is allowing a system to make decisions without human judgment.

Good AI use is giving clear instructions, providing reliable information, checking the sources, testing the output, correcting mistakes, improving the formatting, and reading the final result.

Imagine someone uses AI to help write a historical article. If the article contains false dates and invented people, criticize those errors. If the sources do not support the claims, expose the weak research. If the writing is repetitive, point out the repetition.

That is legitimate criticism because it identifies a real problem.

But if the article is accurate, properly sourced, clearly written, and carefully reviewed, saying “AI wrote this” is not criticism.

Hallucinations are a real problem

AI can invent information and present it confidently. The industry calls this a hallucination. I call it a reason to check the work.

Ask for sources. Open the sources. Confirm that they exist. Check whether they actually support what the AI wrote.

Never assume a quotation is real because it looks professional. Never trust a date because the paragraph sounds confident. Never ship code because it compiled once.

This responsibility belongs to the user.

Teach AI how to work with you

AI performs better when it understands your expectations. I treat that the same way I treat any technical system: define the requirements, provide good examples, set constraints, and test the output.

Give it your preferred style. Show it examples of your work. Explain what vocabulary fits, how citations should appear, and which habits you want it to avoid.

Give it templates, reference material, and specialized instructions. Then review the result and correct what does not fit.

The goal is not to let AI think for you. The goal is to teach it how to assist you.

Someone who knows how to guide, verify, and edit AI will usually produce better work than someone who enters one vague sentence and copies whatever appears.

Always read the result

This rule should never be ignored.

Always read the result.

Do not publish something because AI generated it quickly. Check the facts, tone, structure, spelling, formatting, links, and sources. Make sure it represents what you actually believe.

Your name is attached to the final work. You remain responsible for it.