Draft standards for content-use preferences
The IETF AIPREF working group's two drafts are still Internet-Drafts, not RFCs: draft-ietf-aipref-vocab-08 (13 September 2026) defines three usage categories, train-ai (training a generative model), ai-use (using the asset as input to a generative model when the user did not supply it) and search (selecting assets and directing users to them, with links and excerpts), each set to y or n, where any n wins over a y; draft-ietf-aipref-attach-05 (18 August 2026) attaches them through a Content-Usage HTTP header and a Content-Usage rule in robots.txt (for example Content-Usage: train-ai=n). Its labels differ from Cloudflare's Content-Signal (ai-train, ai-input, search).
Sources
- 1
- 2IETF draft · 2026A Vocabulary For Expressing AI Usage Preferences (draft-ietf-aipref-vocab-08)datatracker.ietf.org
- 3IETF draft · 2026Associating AI Usage Preferences with Content in HTTP (draft-ietf-aipref-attach-05)datatracker.ietf.org
AGrade A: official documentation or a standard. A grade describes the provenance of the sources above; it is not a probability that a change will work. The effect basis says how the size of the effect is known: measured by a study, taken from an industry report, only documented without a size, or not quantified.
Put this evidence to work
Related entries
More in Policies, or search the whole base.
All Policies entries