Dead Internet Theory Revives as 25% of U.S. Adults Use Chatbots Daily
Updated
Updated · The Atlantic · Sep 4
Dead Internet Theory Revives as 25% of U.S. Adults Use Chatbots Daily
3 articles · Updated · The Atlantic · Sep 4
Summary
Nearly a quarter of U.S. adults now use chatbots every day, and OpenAI’s new ChatGPT integration for iMessage has pushed fears of AI-mediated communication from internet fringe theory into mainstream debate.
OpenAI’s tool still requires users to review suggested messages and works on desktop rather than phone, but it formalizes behavior already spreading through work chats, emails and personal messaging.
Reddit posts and user anecdotes show the sharpest anxiety around intimate communication, with people asking whether dating-app messages, apologies or family texts were written by AI and treating that possibility as a breach of trust.
Researchers studying AI-detection forums found users rely on cues such as implausible visuals and ornate, repetitive phrasing, but warned those mental models may quickly fail as large language models improve.
The broader question is whether today’s stigma around AI-written personal messages will persist or fade as automated communication becomes more normalized online.