AI · First steps
How to use AI without knowing how to code
You talk in plain language and AI does the technical work: texts, spreadsheets, formulas, even simple web pages. What you need is to know what you want.
Yes, you can use artificial intelligence without knowing how to code, and that's how the vast majority of people use it. Today's assistants understand requests written or spoken in ordinary language. You describe what you want, the way you'd explain it to a colleague, and the AI returns text, a table, a spreadsheet formula, a script and even the code for a simple web page, if needed. The skill that matters isn't technical. It's knowing how to explain the problem clearly and checking the result.
Coding is still a valuable profession. But the way in has changed: before, the machine's language was mandatory. Now, the machine has learned yours.
Why AI is not a chatbot
A chatbot repeats ready-made answers. An AI assistant works with you: it reads what you sent, asks questions, redoes things when you correct it. It's more like a tireless helper than an answering machine.
In Bairo's view, artificial intelligence is "an extension of the human, not the problem." He often says AI is an extension of the mind of whoever uses it: the thinking is yours, and the tool carries that thinking further. That's why people who don't code aren't left out. They hold the most important part, which is the intention.
Step by step today
- Swap "how is it done" for "what I want." Instead of searching for a spreadsheet formula, describe the result:
I have a spreadsheet with three columns: date, product and amount sold.
I want to know how much I sold per month and which product sold the most.
Explain it step by step, where to click, as if I had never
used a spreadsheet. If a formula is needed, write it ready to copy.
- Ask for one thing at a time. Huge requests produce confusing answers. Break it into steps and only move on when the previous one works.
- Send the material. The most widely used assistants (ChatGPT, Claude, Gemini, Copilot) accept photos, PDFs and spreadsheets. Instead of describing the document, send it and ask about it. Remove personal data first.
- When there's an error, paste the error. If the formula didn't work or a strange message appeared, copy the message exactly as it is:
I did what you said and this error appeared: [paste it here].
I'm using [Google Sheets / Excel on my phone / other].
What did I do wrong and how do I fix it?
- Ask for an explanation, not just a solution. At the end of each task, ask "why did it work?". In a few weeks you'll understand the basics without taking any course.
- Use tools that build things without code. Today there are services that create forms, simple websites and small apps from a description. Test something small before paying for any plan, and read what the service does with your data.
- Know when to call a professional. If what you're building stores customer data, receives payments or is going to grow, the security side needs someone who understands it. AI helps, but it doesn't sign off for you.
A template for any technical task:
I don't know how to code. I want [the result], for [who will use it].
I have [what you already have: a spreadsheet, a phone, a Google account].
Give me the simplest possible path, without installing anything complicated.
Before you start, ask me the questions that are missing.
How Bairo did it
This is perhaps the part of Bairo Leandro Gonzalez Martinez's story that speaks most to whoever is reading now. He says that, when he decided to build what he had seen, he didn't know how to code, or anything like it.
He spent six months at home, in front of the computer, without going out. Developing, writing, getting it wrong and starting again with the tools of his own time. Out of those months came the foundation of the platform and the artificial intelligence of CEASA Bank, a financial platform for agribusiness, designed to put those who produce, those who process, those who trade and those who buy in direct contact.
Along the way, he ran into a problem every heavy AI user knows: it consumes tokens, and tokens are expensive. The answer was to create CLAIN, the ecosystem's agent orchestrator. The name comes from the English claim: the idea that anyone can claim access to intelligence that used to be expensive.
He didn't become a career programmer. He became someone who knows how to ask, check and insist.
Precautions
Not coding is no excuse for not checking: test everything before using it with customers; never paste a password, access key or anyone's personal data into the conversation; remember that AI makes things up with confidence (including functions that don't exist); don't ask it for a legal opinion or a health decision; and notice whether its suggestions always favor the same kind of person or solution, because the bias comes built in.