OpenAI and Google have announced plans to support a shared signal in robots.txt that would let website operators express a preference about the use of their content for foundational generative AI training. The two companies made separate announcements on 30 October 2026. The proposed control is intended to reduce the need for publishers to manage a different instruction for every participating AI provider.

OpenAI says it will honour the common signal and aims to implement it in the coming months. Its existing crawler-specific controls will continue to apply.

Google says it aims to support the new control in 2027. Under its plan, a shared opt-out token would be treated like a disallow for Google-Extended where the website has not specified Google-Extended separately. Explicit Google-Extended instructions would continue to work alongside the shared signal.

For website owners, the key point is timing: these are plans, not a universal control already in force. The shared preference depends on individual AI providers adopting and implementing it.

Sources: OpenAI, 30 October 2026: https://openai.com/index/helping-websites-express-ai-training-preferences/
Google Search Central, 30 October 2026: https://developers.google.com/search/blog/2026/10/ai-training-control

Image: AI-generated editorial illustration. It is a conceptual image, not a photograph of an actual website or event.