The internet quietly crossed a threshold that most people missed. As of early 2026, machines are doing more browsing than humans.

Cloudflare CEO Matthew Prince announced on June 3, 2026, that bot and agentic AI traffic had officially surpassed human-generated web traffic for the first time in the internet’s history. The split, tracked through Cloudflare’s public Radar dashboard, landed at roughly 57.4% to 57.5% bot traffic versus 42.5% to 42.6% human traffic.

“Welp, that happened faster than I predicted… agentic traffic growing so fast that bots have now passed human traffic online for the first time in the Internet’s history.” — Matthew Prince, CEO, Cloudflare

Prince had originally expected this crossover to happen by the end of 2027. The shift likely began accelerating well before the June announcement. Based on Prince’s note that the crossover probably happened about two months prior, that puts the actual tipping point somewhere around late March or early April 2026.

What Cloudflare’s data actually tells us

Cloudflare operates one of the largest content delivery networks on the planet, which means its traffic data reflects a genuinely representative slice of global internet activity. The Radar platform measures HTTP requests across its service network. The category of “bot traffic” here includes agentic AI systems — meaning AI that autonomously browses, clicks, retrieves, and interacts with web services on behalf of users or other systems.

Why this matters for crypto and digital asset markets

The research context notes that no coverage or mention of this milestone occurred within crypto-specific media outlets, suggesting a gap between AI advancements and cryptocurrency discussions.

If more than half of all internet traffic is now generated by automated systems, agentic AI systems that can autonomously interact with web infrastructure are, logically, systems that can also interact with APIs, on-chain data feeds, trading interfaces, and blockchain protocols.

There is also a verification angle. One of the persistent challenges in both web2 and web3 has been distinguishing legitimate users from bots for purposes of airdrops, governance votes, and access controls. Cloudflare’s own CAPTCHA and bot-detection products exist precisely because this is hard. Projects building identity layers, proof-of-personhood systems, or Sybil-resistance mechanisms — a category that includes names like Worldcoin and various decentralized identity protocols — are operating in a market where the business case for proving you are human online became significantly stronger the moment bots crossed 50% of all traffic.