AI agents are slowly changing the way we interact with artificial intelligence. Earlier, chatbots mostly acted like a window to get answers, but now AI agents are becoming active partners that can actually perform tasks for us. Companies like Anthropic, OpenAI, Google, and Microsoft are rolling out AI agents that can browse the internet, collect information, and complete work on behalf of users. As these agents become a regular part of everyday computing, Cloudflare CEO Matthew Prince suggests that by 2027 the internet may soon have more bots than humans online.
Speaking at the South by Southwest conference in Austin, Prince suggested that given the pace at which companies are pushing smarter and more capable AI agents, by next year, the internet might host more automated bot traffic compared to human activity.
Cloudflare CEO highlighted how powerful AI agents are becoming, saying they can already perform tasks at a scale that would normally take many humans. As these agents get smarter, he suggested that both companies and individuals will rely on them more to improve speed and productivity. “If a human were doing a task — let’s say you were shopping for a digital camera — you might go to five websites. Your agent will often go to a thousand times the number of sites that an actual human would visit,” he said. That means a single AI request can generate thousands of page visits in seconds, creating what Prince called “real traffic and real load which everyone is having to deal with.”
Prince explained that before the generative AI boom, bot traffic made up about 20 per cent of the internet, mostly from search engine crawlers like Google, while the rest largely came from spam or malicious activity. However, the situation is now changing quickly as companies train their AI systems using huge amounts of data to answer questions, build models, and perform tasks. “With the rise of generative AI and its insatiable need for data, we suspect that in 2027 the amount of bot traffic online will exceed the amount of human traffic,” he said.
Meanwhile, Cloudflare, which helps websites stay secure and online, says it is already preparing
for a future where bot traffic becomes a major part of the internet. The company whose services power a large portion of the web, is building tools to help businesses handle the rise in automated traffic, protect their servers, and control how AI bots access their websites.
As for the prediction about bots taking over internet activity, the Cloudflare CEO is not alone. Just a few days ago, Brian Armstrong, CEO of Coinbase, made a similar prediction, saying that AI agents could soon carry out more transactions online than humans. According to him, as AI agents start doing more work on behalf of users, they will also need the ability to make payments on their own. He suggested that many of these transactions could happen using cryptocurrency, as bots cannot easily open traditional bank accounts but can operate through crypto wallets.
Whether it is browsing the internet, completing tasks, or even making payments, one thing that many tech leaders in the AI space are making clear is that bots and AI agents are going to play a much bigger role online. As AI systems become more capable, they are expected to change how people use the internet, how businesses operate, and even how digital economies work. In short, the message coming from across the industry is the same: AI agents are not just tools anymore, they are changing the online world and will indeed rewrite human presence on the internet.
