- AI bot traffic has surpassed human traffic on the Internet, according to Cloudflare founder Matthew Prince.
- Bot traffic could grow dramatically faster than human traffic, putting pressure on the Internet’s advertising-funded business model.
- Unlike human users, AI bots generally do not click ads or generate referral revenue for publishers.
- A possible alternative is pay-per-crawl, using micropayments to charge AI systems for accessing publisher content.
- The shift towards an AI-driven Internet could reshape advertising, search, publishing and the value of human-focused media such as TV, audio, film and live events. *
Advertising pays the bill for the Internet, funding most websites and generating all or a significant amount of the profits at Alphabet (Google), Meta, TikTok, Alibaba, Amazon, Microsoft and Apple.
Almost nobody has a better view of the Internet than Matthew Prince, the founder and co-CEO of Cloudflare, the infrastructure company that handles more than 10% of the world’s Internet traffic.
Prince recently gave several powerful interviews on the impact of fast-growing AI bot traffic on the Internet’s business model. In a nutshell, he believes that AI bots will break the Internet’s 28-year-old business model — advertising — and soon.
Bot traffic grows
Historically, bot crawlers represented less than 30% of Internet traffic, a number that had been stable for years, but that’s all changed. Just four months ago, bots on the Internet surpassed human traffic for the first time and are expected to double human traffic by year end.
Further, Prince contends that the math of the Internet now suggests bot traffic growth will dramatically outpace humans, driving the bot to human traffic ratio to 1000x within five years.
This is a big problem for the Internet. Referral traffic metered on human clicks is how Internet platforms and publishers are paid today. That works great to pay for the cost of supporting Internet traffic when two of every three pages are generated by humans.
Bots don’t click
But bots don’t click. However, supporting fast-growing bot traffic requires that publishers and platforms continue to scale up their infrastructure, which costs a lot of money in data centres, as we all know.
Unfortunately, the Internet’s traffic growth will increasingly be ‘freeloading’ bots, not humans paying with clicks. Something will have to give. Either platform and publisher profits drop, or everyone has to do more with less on the infrastructure side. You can’t have both.
Prince suggests that an answer may be ‘pay-per-crawl’ business models for publishers leveraging crypto-based micropayments, effectively moving the Internet from an ad-supported business to a subscription-based one. That would be dramatic.
But the shift from a human-driven Internet to a bot-driven one doesn’t just pressure those who power publishing on the Internet; it puts pressure on the entire advertising industry. Just as the Internet became dependent on advertising these past decades, the world of advertising became dependent on the Internet.
What happens to the ad industry if several hundred billions of dollars of performance ad spend each year is siphoned off by subscription micro-payments?
This question begets some deeper fundamental questions about the future of advertising, including:
Is it realistic that brands and retailers will have the ability to influence the consumer answer-engine experience?
From launch and for many years after, Google had no ads or few ads, and only scaled advertising up when it could on its own terms, not on the terms brands or retailers really wanted.
Might answer engines become brands and retailers themselves rather than trying to service them?
Amazon is doing this. So is Shein — and it’s not a new concept. Montgomery Wards and Sears did it a long time ago.
Will media platforms that retain true human interactions become scarcer, thus more valuable for advertisers?
In a world funded by those desiring to influence human consumption, it only makes sense that those who can continue to maintain direct human interactions will have leverage. Who knows, maybe high-engagement media channels with limited agentic mediation — live events, television, film, audio and reading — will drive more advertising value going forward, not less.
What do you think? Who funds the Internet in its post-advertising future?
* Summary created by AI
This story was first published by MediaPost.com and is republished with the permission of the author.
Dave Morgan, a lawyer by training, is the CEO and founder of Simulmedia. He previously founded and ran both TACODA, Inc, an online advertising company that pioneered behavioural online marketing and was acquired by AOL in 2007 for $275 million, and Real Media, Inc, one of the world’s first ad serving and online ad network companies and a predecessor to 24/7 Real Media (TFSM), which was later sold to WPP for $649 million. Follow him on Twitter @davemorgannyc
