Masto-Mailo-Inator (michal.sapka.pl)

Anubis is a server-side anti-scraping mechanism being deployed to protect sites from large-scale AI-driven scraping. It issues a client-side proof-of-work challenge inspired by Hashcash: individual visitors incur a tiny computational cost, but mass scrapers face exponentially higher expense. The system is presented as a JavaScript challenge page and is explicitly a stopgap while operators develop more precise fingerprinting techniques to detect headless browsers (for example, by analyzing font rendering quirks). Sites warn that modern JavaScript is required and that privacy plugins like JShelter can break access. This approach matters to the AI/ML community because it shifts scraping costs back onto clients, raising the barrier for training-data harvesters and automated crawlers while preserving access for most human users. Technically, it’s a low-complexity deterrent rather than a foolproof block: it reduces server load and discourages mass scraping but can produce false positives and interfere with legitimate bots, researchers, or accessibility tools. The roadmap toward fingerprinting headless browsers signals an escalating arms race—scrapers may adapt with distributed architectures or better browser mimicry—so engineers should plan for resilient, compliant data collection and respect site-owner protections.