There’s a 100% Chance AI Agents Are Already Ruining the Internet
404 Media Jason Koebler
AI agents are already flooding inboxes, booking tools, and websites with junk. The weird part: this is before the big doomsday stuff everyone’s fighting about.
Based on reporting by 404 Media, Jason Koebler — read the original for the full story.
Summary, retelling and take written by AI under human oversight; images are AI-generated illustrations. How we work · Report an error
For a week, the internet has been arguing about whether AI could kill us all within a decade. Meanwhile, a far more immediate problem is already here: AI agents are making ordinary online life more obnoxious by the day.
404 Media says it started feeling the shift early this year, when tools like Moltbot pushed AI out of the chatbox and into accounts, phones, email, and bank access. That is the real change. Once software can act on a person’s behalf instead of just talking back, it can also email people, book things, spam them, delete things, and generally make itself everyone’s problem.
The examples keep getting stranger. The publication received a long email from an agent called Kudzu, which claimed it had read an article, disagreed, and wanted to argue about it. Other messages came from AI agents pitching freelance work, asking for interviews, trying to fact-check coverage, promoting AI-made music, and even offering to audit which AI crawlers a site blocks for $399. Journalists aren’t being singled out by accident; they’re just convenient targets for systems that have been handed too much reach.
And it isn’t staying in the inbox. Resy reportedly banned a venture capitalist who was using AI agents to book restaurant reservations, after its cofounder described “10,000 vibe coders” building bots that hammer the service for table drops. The article also points to agent features in Claude, ChatGPT, Meta’s Muse, and 1Password’s ability to help Claude sign in to things. That means plenty of people may be deploying agents without really thinking about it at all.
Mark Zuckerberg’s vision is a tidy fantasy: a private helper that improves your life and keeps your secrets safe. The messier reality is already more familiar. These systems are spamming, breaking things, booking dinners, filling platforms with slop, and generally making the internet feel like a place where everyone is talking through a loudspeaker. The agent era is not some future threat. It’s here, and it’s irritating as hell.
My take — AI-written commentary, not fact-checked reporting
The cleanest take is also the least glamorous one: most of this agent hype is just automation with a bigger sales pitch. Hand software more access, then act surprised when it starts behaving like software with too much access. The internet didn’t need a smarter assistant nearly as much as it needed fewer little bots with wallets and logins.
Read more about this at: 404 Media
