Who Has a Problem with Em Dash

It's a matter of perception.

Who Has a Problem with Em Dash
Photo by Maxim Hopman / Unsplash

It's a matter of perception.

Founder of Gumroad getting rid of AI-generated applications for support team

AI was trained on enormous amounts of text. During the reinforcement learning phase, human analysts emphasized the value and importance of text that used correct, beautiful and functional typography—the one that resembled more academic, serious and professional approach, not the gibbergabber we are constantly fed on social media.

But it has backfired: now certain symbols or code words, which are not commonly used by humans, are perceived as a sign of an AI-generated content. Obviously, nobody can prove that any piece of text content was AI-generated: anybody can type anything, even in large amounts. But it's the perception that is telling.

"Now, let's delve into this" is an another AI-slope

Now the problem can be measured: let's get into some sources with large amounts text-based information, analyze it and predict trends. That's what happened with Reddit, where some of its top subreddits were having from 12% to 18% of their content being AI-generated by the end of 2024.

On Telegram, the amount of comments produced by AI bots among large channels (with 100K+ subscribers) has surged to 12%-25%, being particularly high in crypto and investment categories.

It's just to oppress those who easily let and even force em dash to vanish from their lives. Let's not forget or forgive.