Artificial intelligence can undoubtedly contribute significantly to human productivity. One area where this technology is most helpful today is in composing technical content, essays, and various other types of writing.
However, the need for human verification and validation remains indispensable. Cases where AI agents fabricate statistical data or legal precedents—and confuse medical information or other technical and scientific reports across various fields—are not uncommon.
Regarding research, for instance, I have personally witnessed ChatGPT present outdated information as if it were the most recent data on a given subject. Specifically, in a query made in April 2026, when I requested the latest iPhone model, the program pointed to the iPhone 15 (released by Apple in September 2023). In reality, the most modern smartphone available on the market—as of the publication date of this article (April 19, 2026)—is the iPhone 17 Pro Max, which was launched in September 2025 in the U.S. and is already available here in Brazil.
In that same vein of erroneous results, the same service identified the Samsung S23 as the most modern Samsung phone in Brazil as of April 2026. It is worth mentioning that in Samsung’s S-line, the most modern model currently available is the Galaxy S26 Ultra, launched in February 2026 in the U.S. and also already on sale here.
And this was just one example of a simple market search driven by personal curiosity.
When it comes to producing technical content for professional applications, human validation is, with even greater reason, a fundamental necessity and a conditio sine qua non. Medical, legal, and scientific texts from various fields, when drafted by AI, must be checked before being presented for their respective purposes; users must confirm the information through other secure and established sources.
Nevertheless, the advantages of using Artificial Intelligence for text composition are undeniable: more accurate grammar, more elaborate writing, and the use of more sophisticated vocabulary—including technical terminology when requested—not to mention the sheer speed of production.
Blog Tip
A suggestion for the effective use of AI tools in text production would be for the user to draft the original text (or a rough draft) themselves and then submit it to the AI for improvement and/or grammatical correction. Once the AI has finished its work, the user performs a “manual check” of the text, reading and verifying that the AI actually followed the instructions.
I have personally made reasonable use of AI tools for text formatting: I wrote the content—in other words, I told the story in my own words, while trying to use good vocabulary—and subsequently submitted it to the AI. I asked the tool to improve the text, correcting any punctuation or comma errors, avoiding word repetitions and pleonasms, and, in some cases, using technical legal language.
And I must mention the primary request: I always made it clear to the tool that it could NOT alter the content of the text. It could and should, at most, rewrite the text using a more polished vocabulary, but never touch the content of the message being conveyed. After the AI finished its work, I performed a final reading and review. It worked very well.
How about you? Have you ever used AI to compose texts or increase your productivity?