Single policy token controls AI training across providers by 2027. Industry-wide standardization of robots.txt AI preferences now underway
Google to Adopt IETF AI Training Control Standard in 2027
Original: Google Adopts New AI Training Control
Importance: 2027年から全ウェブ運営者のAI学習制御フローが統一化される破壊的仕様変更。SEO・コンテンツ戦略へ直結
Summary
Google announced it will adopt an IETF-developed AI training control standard by 2027, similar to Bing's approach. Publishers can now express AI training preferences across multiple AI providers using a single 'policy token' in robots.txt, rather than managing separate instructions for each AI developer. This implements AIPREF working group standards and responds to publisher requests and EU AI Act compliance. Company-specific controls like Google-Extended remain available alongside the new shared signal.
Key Points
- Single robots.txt policy token controls AI training across multiple providers simultaneously
- Planned 2027 deployment in coordination with Bing and IETF standards
- Complies with EU AI Act; existing per-provider controls remain functional
- Search Console AI training opt-out for Gemini grounding continues unchanged
View developer summary
Google will deploy IETF AIPREF standard AI training control by 2027 via robots.txt policy token mechanism. Publishers set single token instead of individual provider Disallow directives; Google interprets token opt-out as equivalent to Google-Extended disallow when not explicitly specified. Other AI providers can adopt matching semantics. Existing per-provider tokens (Google-Extended, etc.) remain functional alongside shared signal. Fulfills EU AI Act General Purpose AI Code of Practice requirement recognizing robots.txt as primary machine-readable preference vector. Multistakeholder IETF refinement ongoing.
Source: https://www.seroundtable.com/google-ai-training-control-42430.html
Outlet: Search Engine Roundtable
This article is an AI-generated summary (Claude Haiku 4.5) of publicly available information from Google, X, TikTok, Meta, LINE, YouTube and other marketing platforms. The original source URL is always provided in accordance with fair-use citation requirements. Summaries are AI-generated and may contain mistranslations or misinterpretations. Always verify details with the original source.