Elon Musk: AI agent traffic will inevitably consume the internet — forecasts and reality
The world stands on the brink of a fundamental transformation of the internet, and judging by recent statements, this transformation will happen much faster than many expect. Elon Musk has openly supported the ambitious forecast by Cloudflare's CFO Thomas Seifert, according to which artificial intelligence activity online will vastly exceed human activity. This is not just another hypothesis—it is a strategic vision that is already beginning to find empirical confirmation.
Speaking on the second-quarter earnings call, Seifert presented truly shocking figures: if current growth rates hold, non-human traffic could surpass user traffic by 1,000 times within just five years. He aptly noted that people risk becoming a "statistical rounding error" in the global web. Musk, commenting on this assessment, emphasized that such a scenario is not a question of probability, but a question of time.
The infrastructural shift is already underway
It is important to understand that we are not just observing theoretical speculation. Cloudflare's initial forecasts suggested that machine traffic would exceed human traffic only by 2027. However, reality has proven far more dynamic: this tipping point arrived as early as May 2026. The pace at which AI agents are being integrated into everyday digital processes surpasses even the boldest analyst expectations.
Data from independent research fully correlates with this picture. The HUMAN Security 2026 report records automated traffic growing eight times faster than user traffic. Notably, between January and December 2025, the monthly volume of traffic from AI agents grew by 187%, with 95% of all this activity concentrated in three key sectors: retail and e-commerce, streaming and media, as well as travel and hospitality.
Imperva's statistics in the Bad Bot Report 2026 show that in 2025, 53% of all web traffic already consisted of automated requests. Fastly, in turn, estimates bots account for 49% of all requests in January 2026, with 99% of them deemed undesirable by services.
Economic implications for digital markets
The scale of this phenomenon extends far beyond technical metrics. According to ARK forecasts, by 2030 AI agents could drive nearly 25% of digital spending, compared to roughly 2% in 2025. The company estimates that by the end of the decade, agents will process up to $8 trillion in online commerce. This means that familiar models of e-commerce, advertising, and user experience will be completely reimagined.
For the crypto industry and blockchain payments, this trend opens a window of opportunity: machine-to-machine micropayments, decentralized identity, and smart contracts are precisely designed to serve this kind of automated interaction. The only question is which players will manage to adapt their infrastructure to the new reality, where the primary consumer will be not a human, but an algorithm.
My view: the market significantly underestimates the speed of AI agent adoption in commercial infrastructure. Investors who are now looking at projects focused on machine-to-machine transactions and automated financial flows could gain a significant advantage over the next 2-3 years. This is not just a trend—it is the new foundation of the digital economy.