Until a few years ago, to fill someone's page with hostile comments, you needed people. Few or many, but people, with time to spend.
That constraint no longer exists, and it is the most underestimated change in public communication in recent years.
What has really changed
It's not that fake profiles exist. They have always existed, and they were recognizable precisely because they were poorly made. Copied and pasted identical sentences, broken Italian, the same formula repeated fifty times.
What has changed is the cost of producing fifty different comments, each written in correct Italian, each with its own nuance, none identical to any other. Before, you needed a newsroom. Today, you need an afternoon and no special skills.
The collapse of that cost changes the nature of the problem. It is no longer about recognizing the single fake profile. It is that apparent consensus has ceased to be reliable information.
All our lives we have used a reasonable mental shortcut. If many people say the same thing, there must be some truth to it. That shortcut worked because numbers were expensive to fake. Now they are not, and the shortcut has become a vulnerability.
Why it pays off, for someone
You don't need to imagine grand conspiracies. The reasons are trivial and almost always present.
There are accounts built to accumulate followers and resell them, and to do that they need to intercept every heated discussion, wherever it starts. There are pages that live off traffic, for which a heated controversy is worth more than original content. And there is the simplest case of all: a single person who wants to make their opinion seem more widespread than it is, and who can now do it alone in an afternoon.
None of these scenarios require shadowy organizations. They only require that it has become cheap, and it has.
How to spot it, without becoming paranoid
None of these signals is proof. Taken together, however, they significantly change how you read a discussion.
The first, and the most reliable by far. Local knowledge. A generated text can be flawless on the general topic and remain completely empty on concrete details. It doesn't name the street, doesn't know what the place is like, doesn't remember what happened the year before. If in a discussion about something that happened in your town, none of the harshest comments contain a detail that only someone who lives there would know, that is the most telling anomaly.
The second, the answer that doesn't answer. If you reply with a specific objection and receive a well-written text that however doesn't address your objection and starts over from the general discourse, you have a serious clue.
The third, the pace. Many long, articulate posts in a few minutes, perhaps under different content, is not a human pace.
The fourth, the account history. A recent profile or one dormant for years that wakes up exactly on this topic. Alone it proves nothing, but together with the others it carries weight.
The useful conclusion
It's not about becoming suspicious of everyone. That would be the worst damage, because a society where no one believes anyone is exactly the goal of those who produce this stuff.
It's much simpler, and it takes a huge weight off your shoulders.
Stop using numbers as a measure of truth. One hundred comments against you do not mean one hundred people think that way, nor that one hundred people even exist. They mean that under that post there are one hundred comments.
Which, when it's your face or your country under attack, is surprisingly good news.
Segnale & Rumore is a column about how digital communication really works, between algorithms, platforms and online reputation. It is written by Calogero Bono, who works with these machines every day.
