If you have browsed comments on YouTube, Facebook or X lately, you have probably met accounts that look a little odd: nearly identical comments repeated, offers of an 'amazing investment', or profiles without real photos replying within seconds to every post. Many are bots, accounts run automatically by software rather than a person.
A bot, broadly, is software that performs automated actions online. Not every bot is bad: some index information for search engines, some handle customer service, some post weather updates. But on social networks the term usually refers to fake accounts impersonating people. They are used for scams, product promotion, artificially inflating followers and likes, and sometimes to sway public opinion on political issues.
AI changed the game. Bots used to be easy to spot: broken language, identical messages, empty profiles. Today language models can write varied, natural comments relevant to any post in any language, and image tools generate profile photos of people who do not exist. So one operator can run thousands of accounts that look like real people.
How to spot them? A few signs still help. An account opened recently with very intense activity. A profile photo that looks 'too perfect' or has no others like it. A history focused on one topic, such as only investments or only politics. Comments that arrive very quickly after posting. And a username with a random string of numbers. No single sign proves anything, but a combination raises suspicion.
Why does it matter to you? Because bots can make an opinion, product or scam look more popular than it is. When you see hundreds of comments praising an 'investment', it is easy to think it is real. The advice: do not judge by the number of likes and comments, check the source, and above all do not click links in strangers' comments. And if you spot an account that looks fake, report it to the platform; it helps clean things up for others too.
