Meta's self-built AI search has swept the entire web, and has already crawled the servers of independent developers, triggering alarms.

According to Watchful AI monitoring, Meta is aggressively crawling the web to index it for its AI search. Independent developer Pieter Levels reported that Meta has been heavily crawling several of his websites this week, causing one VPS to trigger load alerts multiple times.

Statistics from DataDome show that in the second quarter of this year, Meta's two AI crawlers collectively sent out 9.1 billion requests. The Meta-WebIndexer, responsible for building the search index, saw a quarterly growth of 163%, with the June request volume surpassing that of the training crawler for the first time.

Meta was previously exposed in 2024 for developing its own search engine in an effort to reduce reliance on Google and Bing. The current acceleration of this search index expansion through such high-intensity crawling, however, has brought minimal real user traffic to websites.

An anonymous Meta employee revealed in a private message their company's ambition to create its own "Google" to avoid handing over AI search requests to Google, only to have Google subsequently use them for training. This information is currently based on an anonymous tip and has not yet been confirmed by Meta.