All articles

Publishers Draw a Line in the Sand to Make AI Pay for Web Scraping

Marketing Insider News Team
published
September 14, 2025
Credit: Outlever

Key Points

  • A coalition including Reddit and Yahoo launches the Really Simple Licensing (RSL) standard to ensure AI companies pay for using web content.

  • The RSL standard allows publishers to embed machine-readable licensing terms, replacing the traditional robots.txt file.

  • The RSL Collective, modeled after music rights organizations, aims to negotiate with AI firms collectively for content usage rights.

  • The initiative's success hinges on widespread adoption by publishers and compliance by AI companies.

A coalition of major publishers and tech companies, including Reddit and Yahoo, has launched a new technical standard and a nonprofit collective designed to ensure they get paid when AI companies use their content for training models. The move is a direct challenge to the long-standing practice of AI firms scraping web data for free.

  • From robots to royalties: The new Really Simple Licensing (RSL) standard allows any publisher to embed machine-readable licensing terms onto their website, effectively replacing the simple "allow/disallow" robots.txt file with a price tag. The framework supports a range of payment models, from subscriptions to "pay-per-inference" fees, where publishers are compensated each time their content is used in a chatbot's response.

  • The ASCAP playbook: To give the standard some teeth, the group also launched the RSL Collective, a nonprofit modeled on music rights organizations like ASCAP. By bundling the rights of millions of creators, the collective aims to negotiate with AI giants from a position of strength, creating a unified front that is harder to ignore than individual publishers acting alone.

The initiative marks a significant escalation in the battle over the value of online content. Its success will depend on broad adoption by publishers and, more critically, on whether AI companies, who have been striking one-off licensing deals, will choose to comply with a collective, industry-wide framework.

The move to formalize licensing comes as the conflict over data scraping intensifies. One study found that 83% of newspapers and magazines were already attempting to block AI bots. Meanwhile, some publishers are forging their own path, with companies like Vox Media signing individual content deals directly with OpenAI, highlighting the different strategies emerging to address the issue.