In order to stop AI’s knowledge-base from stagnating or corrupting, frontier AI models need to keep ingesting vast quantities of current data from the internet. In this respect, the major AI providers, and their avid competitors, are currently meeting two obstacles: a) the publishers, social networks, platforms and other sources of fresh data now realize what this access is worth, and are looking to monetize it; and/or b) domains are now getting so hammered by relentless AI scraper-bots looking…