Of course, the words in the headline are an exaggeration. Sometimes I still have to go to the scene of an event, talk to people, and then turn everything I have heard and seen into coherent sentences. But in many other cases, I simply give a task to artificial intelligence, after which all I have to do is polish the text it produces in seconds and adapt it to the media outlet where it will be published.
For example, the other day I “wrote” a long feature article for Farmer’s Day. To do this, I gave Gemini a task in which I outlined the parameters of the future text: the language of the article, its length in thousands of characters, and its content. In terms of content, I asked it to explain when this day first appeared, how the lives of Ukrainian farmers have changed in the context of the war, what difficulties they face today, and what contribution they make to ensuring the country’s food security. To support the text with facts, I added one more instruction: to provide basic statistical data. Finally, I asked it to congratulate Ukrainian farmers on their professional holiday. The neural network handled the task brilliantly. I only adjusted the headings and changed one grammatical case in one place.
Speaking of Congratulations
Some time ago, I needed to write a congratulatory message for Journalist’s Day of Ukraine on behalf of the director of a vocational school. My request to Gemini was brief. I told it on whose behalf the greeting should be written and explained whom the school trains: pastry chefs, gas welders, and electricians. The result was so impressive that I realized I would have racked my brain over that greeting for a good hour and most likely would not have come up with such elegant comparisons as the artificial intelligence did.
Let’s Be Honest
Sometimes I have to prepare materials based on what has already been published in other media outlets. Simply because I did not have the opportunity to attend the event, while the news feed still needs to be filled. A neural network handles this task perfectly: it is enough to upload several texts copied from the Internet and tell it what you want to get as the output. Alternatively, you can send it links to those materials — and they may be in different languages. Yes, it produces pure rewriting, and I, for that matter, am engaging in petty theft. But let’s be honest. These days, anyone and everyone does this, and personally, I have absolutely no objection if someone uses my original text to save time and effort.
Analysis from Discussion
However, there are situations in which a neural network helps me write a much more complex analytical text. And the main advantage of working with it lies, above all, in the enormous saving of resources. Instead of going to Google with topic-related queries and digging through a pile of websites, I simply ask the same Gemini, and it readily answers me. Moreover, it often offers me a discussion, and a kind of polemic develops between us. The result is that the article becomes sharper because it reflects different opinions — at the very least, mine and that of the public or expert community on whose behalf the neural network argues with me.
This is exactly how I wrote a long article about whether Russian kamikaze drones should be shot down over urban areas, given that the risk of a downed UAV falling onto residential neighborhoods and killing civilians is very high. The neural network explained to me that if the drone is not shot down over the city, it will reach its target, and not only people but also the infrastructure facility the drone was aimed at will suffer. I reflected this point of view in the article, but expressed my disagreement with it and explained why. The article received a lot of positive feedback from city residents who constantly face the risk of being killed by fragments of a downed Shahed or by the explosion when it crashes.
To Summarize
In preparing written materials — news, extended notes, analytical pieces — artificial intelligence is an excellent tool that saves you time and finds the information you need faster than any search engine. Does this mean that you stop being an author capable of writing a coherent text on your own? Not at all. You simply stop wasting time on routine work and focus on more serious tasks. In the end, it is your own vision of the problem that helps you properly formulate the task for AI and then bring into publishable form what it has produced according to your instructions.