Definition
Really Simple Licensing (RSL) is an open, XML-based standard for expressing machine-readable licensing terms for digital assets. It was introduced in September 2025 and reached version 1.0 in December 2025, stewarded by the nonprofit RSL Collective, whose founders include RSS co-creator Eckart Walther. RSL deliberately builds on two familiar foundations: the RSS syndication format and the Robots Exclusion Protocol behind robots.txt.
Where robots.txt can only say allow or disallow, RSL describes terms. A publisher can declare that content is free for search indexing but licensed for AI training, or attach a payment model—free, attribution, subscription, pay-per-crawl, pay-per-inference, or a good-faith contribution. Pay-per-inference is the notable addition, billing when a model uses content to produce an answer rather than only when it fetches the page.
The standard ships with supporting protocols. The Crawler Authorization Protocol (CAP) uses an HTTP Authorization: License header so a server can verify permission and return 401 or 402 when it is missing. The Open License Protocol (OLP), an OAuth 2.0 extension, lets crawlers discover and acquire license tokens automatically. An encrypted media standard covers nonpublic assets such as books, video, and datasets. RSL documents are discovered through robots.txt, HTTP headers, RSS feeds, or HTML link elements.
RSL is a declaration layer, not an enforcement layer—much like robots.txt, it depends on good-faith compliance plus separate technical gates such as pay per crawl or Web Bot Auth. Early adopters included Reddit, Yahoo, Quora, and Medium, followed by publishers such as BuzzFeed, USA Today, and Vox Media. For GEO teams it is the most precise tool available for separating "index and cite me" from "train on me," which matters because those two decisions have opposite effects on AI search visibility.
Examples of Really Simple Licensing (RSL)
- A publisher adds an RSL license file at the site root and references it from robots.txt, declaring free access for search indexing but pay-per-inference terms for AI answer generation.
- A crawler without a valid license token requests a protected article and receives HTTP 401 Unauthorized rather than the content.
- A media group uses RSL's pay-per-inference option so it is compensated whenever its reporting contributes to a generated answer, not just when a bot fetches the page.
- A GEO team audits its RSL terms after noticing that a blanket AI restriction also suppressed the brand from AI Overviews and ChatGPT citations.
