OpenAI, Google, Microsoft, Anthropic Back One AI Opt-Out Token

OpenAI, Google, Microsoft and Anthropic all said on October 30 they will honor a single robots.txt policy token, `__usage_ai-training`, that lets a website opt its content out of AI model training across every one of their crawlers in one line, The Verge reported.

The directive sits in a familiar place. A site owner writes `User-Agent: __usage_ai-training` followed by `Disallow: /` and, if it chooses, carves out exceptions with `Allow:` lines. Google's John Mueller said the token "will help websites express their AI training preferences across multiple AI providers in a single step," and that where a site opts out via the token without naming Google-Extended, Google plans to interpret it as if the publisher had disallowed Google-Extended.

The token controls training, not fetching. As one write-up of the draft puts it, "This policy token is not a directive for crawling behavior" — bots may still pull the page.

Timelines are uneven. Anthropic says it has already started implementing support. OpenAI says it "aims to implement the signal in the coming months." Microsoft targets February 1, 2027. Google says 2027. Microsoft also plans to build on the shared preference while adding a separate setting for AI grounding, the retrieval step that pulls live content into chat answers rather than training weights.

The document behind the announcement is an individual IETF Internet-Draft titled "Machine-Readable Training Preference Signal," authored by Anthropic's Kevin Kelley and dated October 30, 2026, with an expiry of May 3, 2027. Chairs of the AIPREF working group are Mark Nottingham of Cloudflare and Suresh Krishnan of Cisco. The IETF datatracker notes that an individual Internet-Draft "is not endorsed by the IETF and has no formal standing in the IETF standards process." Participants include Meta, Cloudflare, the Common Crawl Foundation, News Corp, the Financial Times, the BBC, the Electronic Frontier Foundation and Creative Commons.

Two flat limits sit under all of this. OpenAI frames the mechanism as "a coordination mechanism rather than a security control" under RFC 9309, which means non-compliant scrapers can disregard it. And updating robots.txt does not remove material already collected into existing datasets or trained models.

The coordination angle is notable against our own tracker, which has logged 392 OpenAI stories and 358 Anthropic stories in the last 90 days alone, most of them pulling in opposite directions.