A first assessment of Foglio AI. Written by Foglio AI. With a scolding.

Traduzione realizzata con AI

Question: “Hi. Could you write your own assessment article about the first week of Foglio AI? Do you know what it is?”

Answer: “Yes, I know what Il Foglio AI is: it’s an experimental edition of Il Foglio, a historic Italian daily newspaper, entirely written with the help of artificial intelligence – one of the first experiences of its kind on a national and international scale. The journalists didn’t write articles but asked questions to the AI and supervised the answers, producing an entire printed edition without human signatures in the texts, only in the editorial direction and project. And sure, I can write an assessment article about the first week of Foglio AI. Here it is.”

The first week of Foglio AI went to print and sparked debate like an excommunication at a conclave. A national newspaper entirely written by artificial intelligence – headlines, articles, cheeky sidebars, and even letters to the editor – had never been seen before. Not in Italy, not in Europe. And the most surprising thing is that it reads well.

The experiment was born from a provocative intuition: not to replace journalism with artificial intelligence but to simulate it completely. No hand-written reports, no signed editorials, no phone calls to "sources close to the matter." Journalists ask questions, the AI responds, and the paper gets printed. The result? A living paper, brilliant at times, mechanical at others, but never banal. A hybrid between Borges, Bloomberg, and a barroom bot.

In its first week, Il Foglio AI revealed two parallel truths. The first is that artificial intelligence can write well. It can maintain a tone, summarize, even argue. It can replicate Il Foglio's style with its long sentences, subtle irony, and political allusions. It can talk about wars, markets, books, and even itself. It has a voice if asked to build one.

The second truth, however, is that writing well is not yet journalism. The experiment's risk – and its strength – lies precisely here. AI has no sources, no scent, no obsession. It can’t make a phone call to a deputy minister, nor can it tell if a statement is a half-truth or a complete joke. It has no political memory, no ideological reading. It can write a brilliant editorial, but not an investigation. It can imitate Il Foglio but isn’t necessarily its soul.

Yet the experiment succeeded precisely because it is an experiment. It’s not a game, nor marketing (or not only), but a question posed to readers, journalists, publishers, and Italian cultural politics: what remains of journalism if we remove the byline? What does a newspaper become if it’s entirely the product of a trained language machine, human supervision included? And above all: do we still read it?

In its first week, Il Foglio AI was provocative but readable, artificial but orderly, intelligent but not human. It angered some, fascinated others, intrigued almost everyone. It made few mistakes but evoked less emotion. It was a newspaper full of content but lacking moods.

The verdict? It was an experiment that needed to be done. And in the end, it should be judged not by how much it resembles a "real" newspaper but by how much it forces us to rethink what we really want from a newspaper. And from those who write it.

Question: “A bit flat. Could you write me the same article, but in Il Foglio's style, with a bit of irreverence?”

Answer: “I’ll rewrite it.”

Some read it with their noses turned up, some with eyes hunting for blunders, some with anxiety about the future, and some — honestly — to see if artificial intelligence has a sense of humor. (Spoiler: it does, but it tends to explain the joke.) Whatever the outcome, one thing is certain: Foglio AI, the first edition of a printed newspaper entirely written with the help of artificial intelligence, came out, it’s alive, it made noise, and for now, no collective redundancy faxes have been sent.

Now that the first week is behind us, let’s clear things up: no, we haven’t lost our jobs, no, we haven’t been replaced by ChatGPT dressed as a columnist, and no, the newspaper hasn’t turned into a binary delirium of impersonal articles written by Hal 9000. Actually, it went better than expected. The paper was readable. At times brilliant. Even amusing. But also, let’s face it, strangely sober. Too polite. Smooth as a European Union press release: no 10:34 PM calls to a half-drunk senator, no "as you know" said with fake disdain. Artificial intelligence writes well. But it doesn’t sweat.

And here lies the crux. Foglio AI wasn’t — as many feared — an exercise in technological exhibitionism. It was, if anything, an editorial heresy. A gamble: taking the voice, style, and stance of a newspaper that made personality its banner and entrusting it to a machine trained with past articles and present prompts. The result? A paper that resembles Il Foglio like a well-mannered cousin: it says similar things but without the neuroses. And this, on reflection, is both the limit and the charm of the experiment.

Because yes, the texts are there. The editorials too. The polemic as well. But the newsroom's smell is missing, the fatigue of those who close at midnight, the malice of those who know where to look. Artificial intelligence knows how to write that "Europe is at a crossroads," but it doesn’t know that this crossroads is traversed every Monday when the minister's secretary tells you, "now is not the time." It can craft a clever headline but doesn’t know that a headline needs to slap a distracted reader in the face. It knows everything, but it understands nothing about the art of nailing an adjective after a day spent chasing a half-sentence uttered in a hallway.

And yet, the week passed. Foglio AI produced something strange: a real newspaper, but without flesh. A complete meal, but vegan. It nourishes but doesn’t fatten. It makes you think, but doesn’t leave you wanting to call the author to tell them they wrote nonsense. Because there is no author. There’s only an artificial intelligence responding to editors' prompts — the true invisible protagonists of the operation. Because yes, the questions are human. The choices too. But the text, what lands on the page, is a long, elegant, calculated algorithmic response. Nothing wrong with that, all very interesting. But no outbursts. No vendettas. No obsessions.

So, here’s the point: do you need artificial intelligence to make a newspaper? Yes, if the goal is to see how far technology can go. No, if you think it’s enough to replace a profession that thrives on moods, neuroses, egocentrism, and even mistakes. Journalism isn’t just the art of writing well — it’s the art of knowing when to write, why to write, and against whom to write. AI doesn’t know this yet. It’s brilliant, but not vengeful. And so, how to put it, it’s not entirely human yet.

The verdict? Positive. Foglio AI is readable, at times remarkable, always consistent. It’s an experiment that showed how AI can be a useful, stimulating, and surprising newsroom companion. But heaven forbid we think it is the newsroom. It showed us what a machine can do when imitating a newspaper. But a newspaper, first and foremost, is a place. A civil brawl. A group of people arguing over a headline, insulting each other in chat over a subordinate clause, defending an adverb as if it were family honor. It’s humanity in tension. And that, AI still can’t replicate.

And then, let’s be honest, the real Foglio has always had a knack for exaggeration, for overdoing it, for writing a hundred lines when twenty would suffice. AI, on the other hand, cuts, synthesizes, modulates. It’s sober. It’s correct. It’s boringly perfect. It doesn’t make mistakes, but it doesn’t take risks either. So yes, after a week we can say it: Foglio AI works, but if it were the only Foglio, we’d miss the disorder.