Who are running these bots? I presume developers at all of the frontier labs know (or at least would know to look for) Wikipedia has bulk APIs for automated access. Unnecessary scraping increases their workload/costs too, so why in 2026 is this still a problem?