Cloudflare Reports Bots Generate Majority of Web Traffic

Cloudflare reports that automated systems now generate most web traffic. According to Cloudflare Radar, 57.4% of HTTP requests worldwide are bot-originated and 42.6% are human, figures echoed by NBC News, CNET, and Tom's Hardware. CEO Matthew Prince said he had not expected this crossover until late 2027 and that the shift happened over just a few months; he cautioned the data is still rough but the trend is unmistakable. Per Mashable, Prince's reaction was, "Welp, that happened faster than I predicted." On the web's future economics he wrote that "clearly it's going to be pay to crawl."

Reporting attributes the growth to autonomous AI "agents," software that programmatically browses and fetches pages on behalf of higher-level services. NBC and CNET note that a single agent can visit thousands of pages in the time a human visits a handful, so agent-driven activity is disproportionately large per request. CNBC, citing HUMAN Security's 2026 State of AI Traffic report, reported that agentic AI traffic grew roughly 7,851% year over year and that automated traffic expanded about eight times faster than human activity in 2025, with OpenAI-operated bots making up a large share of observed AI traffic.

A sustained majority of automated requests changes capacity planning, cost allocation, and the meaning of metrics long assumed to reflect human attention. Common mitigations include refining bot classification, instrumenting higher-fidelity engagement signals such as authenticated sessions and client-side events, and separating billable APIs from open crawl endpoints. Cloudflare launched a platform in 2025 that lets sites gate and charge AI crawlers; Prince says it has not yet gained broad traction and that supporting the required volume needs more protocols and infrastructure.

Track new bot-classification signals from CDN and WAF providers, server- and client-side metrics that distinguish agentic activity, and whether "pay to crawl" or similar access-and-monetization models gain adoption among large publishers and platforms.