An article by OpenAI CEO Sam Altman has stirred up the tech world. They’re sparking a debate about the authenticity of social media. In a post, Altman admitted that he struggles to tell the difference between human and bot-generated content, saying that platforms like X and Reddit “feel very fake.”

This admission came after he noticed a strange pattern on a Reddit forum dedicated to a competitor’s product, where a flood of users were posting about how they were switching to OpenAI’s new Codex tool. Despite knowing his company’s product was growing, he couldn’t shake the feeling that the posts were bots. This subsequently led him to question the true nature of online discourse.

Sam Altman says AI bots are making social media feel fake

Sam Altman laid out a few reasons why he believes social media is feeling so synthetic. He pointed out that real people have started to mimic the quirky language of large language models, or LLMs. Altman also noted that platforms pressure creators to juice engagement, and rival companies use “astroturfing.” If you’re not aware, “astroturfing” is a practice where paid accounts create the illusion of genuine grassroots support.

What makes this situation particularly ironic is that the very technology making social media feel fake was created by his company, OpenAI. After all, LLMs work mimicking human communication. Adding to the irony, OpenAI’s models were trained on data from sites like Reddit. Altman himself was a board member and remains a major shareholder of the platform. The technology learned from our communication, and now we’re adopting its style, creating a strange feedback loop.

A calculated move?

Many critics have been quick to point out the double standard here. They argue that Altman’s public concern over AI-generated content seems disingenuous, especially since he is a key figure in the very industry that created the problem.

Some have suggested that his public lament is a calculated marketing move to prepare for OpenAI’s rumored entry into the social media space. The company has reportedly been exploring building its own platform to compete with X and Facebook. If this is true, painting the current social media landscape as a “fake” and bot-ridden wasteland would be the perfect way to market an “authentic” alternative.

Whether his motives are genuine or strategic, Altman’s observation highlights a real and growing issue. Data security company Imperva reported that over half of all internet traffic in 2024 was non-human. X’s own bot, Grok, has estimated that hundreds of millions of bots are active on its platform alone.

Ultimately, if the CEO of the world’s most influential AI company can’t tell whether his biggest supporters are real, then perhaps no one can.