As AI-generated writing spreads across the platform, users are scrutinizing em dashes, emojis and repetitive phrasing to call out inauthenticity.
https://www.bloomberg.com/news/articles/2026-01-30/chatgpt-written-linkedin-posts-have-users-analyzing-emojis-other-ai-signs
3 Comments
*Lora Kelley for Bloomberg News*
Over the summer, Kiara Stent set out on a hunt. She started scrolling through the LinkedIn posts of various people — including leaders in the marketing field and people she knows — reading some 200 posts across about 50 different profiles. What she found “made me lose a little bit of respect,” she said, especially for the people in more senior roles. About 75% of the posts, in her opinion, seemed to be generated by artificial intelligence.
Stent, a 25-year-old marketer in Cape Town, began this informal research project after noticing that the posts in her feed had started to sound the same — dramatic metaphors, grabby hooks, no voice. She posted in August about what she had identified as the telltale signs of AI use, and the post quickly became her most popular, generating tens of thousands of impressions. The reaction was split, with most commenters laughing and agreeing, and some getting defensive. But her points resonated.
“My LinkedIn was going crazy,” she said.
Stent is part of a small but growing army of concerned citizens vigorously calling out what they identify as AI-generated writing on LinkedIn. Some consider the em dash a dead giveaway that something was written by ChatGPT; others side-eye any use of the Oxford comma. Big words? Repetition of ideas? “Not just X, but Y”? Winding syntax? Emoji? Red flags, all.
[Read the full dispatch here.](https://www.bloomberg.com/news/articles/2026-01-30/chatgpt-written-linkedin-posts-have-users-analyzing-emojis-other-ai-signs?accessToken=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJzb3VyY2UiOiJTdWJzY3JpYmVyR2lmdGVkQXJ0aWNsZSIsImlhdCI6MTc2OTc2NjYxOSwiZXhwIjoxNzcwMzcxNDE5LCJhcnRpY2xlSWQiOiJUOU82Q1VLR0lGUTIwMCIsImJjb25uZWN0SWQiOiJEMzU0MUJFQjhBQUY0QkUwQkFBOUQzNkI3QjlCRjI4OCJ9.5Eo9CcVWT_sSk5_wux_i7KVdOT-_ffUVYgV3H8cxNrQ)
Fuck AI posts, death to clankers, eat the technocrats.
where the performance of being “real” is itself a valued currency.
& if AI can’t be genuinely human, then demonstrated humanity becomes a premium skill.
Damn. That’s fascinating. insightful read! Thanks for sharing OP