The New York Times paints a stark picture of a "post-human internet" approaching fast. The "Dead Internet Theory" — the idea that most digital traffic will eventually be produced by software, not humans — is becoming reality. When AI chatbots start acting as job applicants, interviewers, customer service representatives, and even dating partners, a fundamental question emerges: what happens when these bots start talking to each other?
Christian Vinson's story illustrates this perfectly. When he applied for finance jobs this year, he spent nearly 10 hours teaching two AI chatbots about his work experience — his education, a banking internship, five years in the Army. He had the AI tailor hundreds of cover letters. On the other end, more AI systems read, scored, and filtered his applications. No human was involved at any point.
Human-to-human interaction is becoming bot-to-bot loops. In job hunting, customer service, and even online dating, both ends increasingly use AI to generate and screen content, leaving humans on the sidelines. When AI-generated text gets read by another AI, processed by a third, each pass amplifies the model's biases and errors. When AI writes your emails, replies to messages, and even dates for you, your communication skills atrophy.
The recruiting industry has fully embraced AI screening, but the "AI applicants vs AI screening" loop may cause genuinely qualified candidates to slip through the cracks. Customer service is undergoing a similar transformation, with AI handling AI-generated complaints and human agents reduced to handling "exceptions the AI couldn't handle." Social platforms face the challenge of "AI pretending to be human" — when profiles, conversations, and photos can all be AI-generated, how do you distinguish real users from bots?
The "post-human internet" isn't a distant concept. If you've recently applied for a job, chatted with customer service, or used a dating app, you've been in this loop. A practical tip: in key decisions, always ask for "a real human" — it's both self-protection and a signal to platforms that AI can assist but not replace human judgment.
AI was invented to help humans communicate more efficiently. But when we outsource the communication itself to AI, our presence on the internet gets diluted. The deepest irony of the "post-human internet" is that it wasn't created by some rogue AI — it was built one choice at a time, by each person deciding to "use AI to write a better cover letter" or "use AI to quickly reply to a customer message." We handed the internet to the machines — and they're just doing what we asked them to do.