Humans are already outnumbered on the web

I've had a question stuck in my head for a few days, and it's not a fun one for someone who just started a blog. Who are you writing for when half your visitors aren't there to read? Because that's where we are. On the web, bots now send a little more than half of all requests to web pages, ahead of the flesh-and-blood readers¹. The 50% line, which we'd watched creep closer for years like a horizon, was crossed in early June, on an ordinary day, without so much as a ripple.

More than half of all requests come from a machine

The number has the cold precision of bad news. Some 57.5% of requests² come from automated systems, against 42.5% from people. Matthew Prince, Cloudflare's co-founder and CEO, announced it in early June, based on the public dashboard where the company shares its traffic data¹. Cloudflare sits between a big chunk of the world's websites and their visitors, so it's in a pretty good spot to count who's knocking. Prince admits the data is "still rough, but the trend is unmistakable"³. Let's not quibble over the decimal. For the first time in the history of the internet, most traffic isn't human⁴.

Assistants that browse in our place

What tipped the scales is so-called agentic AI, semi-autonomous programs that browse the web on behalf of assistants like ChatGPT or Gemini. The math is brutal. When you shop for a camera, you might open five tabs before deciding. An agent sent on the same errand hits thousands of pages in seconds without breaking a sweat. Multiply that by the millions of people now handing their digital chores to a machine, and the traffic tips toward the bots all on its own⁵.

The wild part is the timing. In March, on stage at SXSW, the same Matthew Prince predicted the crossover for the end of 2027⁶. He beat his own forecast by eighteen months. In an industry that usually gets things wrong by calling them too early, you have to admit that's rare.

What the number leaves out

Before we declare the web empty of humans, one caveat. Cloudflare counts requests, not presence or time spent, and those are very different things. A reader who spends ten minutes on an article triggers a handful of requests. A bot sends thousands in the same time. By volume, the machines win easily. By actual attention, humans still have a comfortable lead, busy as we are scrolling feeds and bingeing one more episode. We haven't been replaced. We just lost the weigh-in.

A web that never planned for this

That's where it pinches, because the web was never designed for this. Its whole business model rests on an unspoken deal. A publisher puts something out for free, a human comes to read it, sees an ad or subscribes, and the content ends up paid for⁵. A bot reads without clicking and eats bandwidth without buying anything. Worse, it scrapes whole pages to train models that will answer in your place tomorrow, without ever sending anyone back to the source. The publisher pays for hosting, the bill grows with every machine visit, and nothing much comes back.

Cloudflare draws a blunt conclusion. According to its CEO, bots will have to pay per visit³. For the occasion, the company dug up an old forgotten code from the protocol that runs the web, 402 "Payment Required." It was set aside in the 1990s for paid access and sat unused for thirty years⁷. Now it's been promoted to tollbooth for the automated web. The question is what happens to a web where every page barricades itself behind a turnstile, and where the original promise of a commons open to everyone turns into an archipelago of fortresses charging admission.

The "dead internet" theory, that old fever dream of a web mostly full of machines talking to machines, has never sounded less far-fetched. And by a different road, the question meets the one raised by an AI reclassified as a state secret. Who exactly owns what we put online?

I don't have a moral to draw from all this, just a strange feeling. The sense of having crossed, without noticing, a line everyone had placed much further out. Pages still load, sites still run, nothing looks different, and maybe that's the most unsettling part. To top it off, the blog you're reading rolls out the red carpet for these bots and explicitly invites them to index it⁸. So I'll hand my question over to you. If the web fills up with visitors who aren't there to read, who are we writing for?

Notes

1. Cloudflare Radar, "Bot vs Human", real-time dashboard accessed June 2026. ↩ ↩²

2. A request is each ask a browser or bot sends to a server, for a page, an image, or a bit of script. Opening a single page often fires off dozens of them. ↩

3. The Decoder, "Cloudflare CEO says the web's future is 'pay to crawl' as bots overtake human traffic", June 2026. ↩ ↩²

4. NBC News, "Bot web traffic has overtaken human web traffic, data shows", June 2026. ↩

5. Forbes, Josipa Majic, "Bots Now Outnumber Humans Online, And The Internet Was Never Built For This", June 4, 2026. ↩ ↩²

6. TechCrunch, "Online bot traffic will exceed human traffic by 2027, Cloudflare CEO says", March 19, 2026. ↩

7. The HTTP protocol numbers its responses. 404, page not found, is the famous one. 402 spent thirty years in a drawer waiting for someone to give it a job, which makes it the most patient status code in computing history. ↩

8. True to its word, this site's robots.txt, the file that tells bots what they may visit, explicitly welcomes indexing bots, AI included. A /llms.txt even hands them a map. Might as well receive them politely. ↩