OpenAI chief executive Sam Altman has voiced concern that discussions on social media no longer feel authentic. In posts on X, he reflected on Reddit threads about Codex, a programming assistant from OpenAI, and said the style of the conversations struck him as oddly manufactured.
Altman remarked that the threads gave him a strange impression, noting that the language in many comments resembled the phrasing of large language models. He suggested two possible reasons: real people may be adopting AI-like habits in their writing, and automated accounts may be amplifying the effect.
A Shift in Online Style
According to Altman, this shift reflects broader cultural patterns in how communities interact online. Social platforms push content that drives engagement and monetization, which can make real users sound as if they are following the same template. At the same time, the presence of bots adds to the sense of uniformity.
Concerns Beyond OpenAI
Altman’s observations add to a wider debate about the growing presence of synthetic content. Analysts and technology leaders have warned that low-quality AI output, often designed to capture clicks, is filling feeds across major platforms. Substack’s leadership, for example, has cautioned that automated posting could saturate timelines with material aimed only at engagement.
Struggling to Separate Human From Machine
The overlap between human and AI expression has created a loop in which both sides influence each other. As AI systems learn from online writing, and people in turn echo AI phrasing, the boundary between the two continues to narrow. Altman’s comments highlight how even genuine discussions now risk being perceived as artificial.
Read next: Meta and Anduril Secure Role in U.S. Army Combat Goggles Project
