Meta commercial terms Oct. 30: Meta can crawl your site to train AI

If your business runs ads or a page on Facebook or Instagram, Meta's new commercial terms give it written permission to crawl your website and train AI on what it finds. The updated Meta Commercial Terms take effect October 30, 2026. You accept them by using Meta's business tools. There is no checkbox to decline. Nieman Lab reported on October 15 that legal experts read the clause as a broad training license. Meta says nothing changes. The contract text is what you sign.

What actually happened

We read the preview of the terms on Meta's own legal page. The licensing section says:

• Automated access. You authorize Meta, and anyone acting for Meta, to use crawlers, bots and similar tools on domains, pages and URLs "associated with" your products or services. The terms do not define that phrase.
• What the bots can do. Access, index, cache, analyze, collect, retrieve and crawl that content, and pull updates periodically.
• Test credentials. Meta may create test credentials to get that access. Cornell law professor James Grimmelmann told Nieman Lab this could let Meta sign up as a user to get past registration walls.
• Training. Meta may use the content to develop, improve, research and train its products and features.
• No robots.txt. The terms do not mention robots.txt.

Meta spokesperson Andy Stone called the reports false and said there is no change to how Meta uses web crawlers. He said the update helps Meta pull accurate product information from third-party sites, and that site owners can still use robots.txt. Whether a contract clause overrides a robots.txt block is untested in court.

Why Meta's commercial terms matter for your business

For a store, "pull accurate product information" is not the scary part. That is how catalog ads work. The scary part is scope. The clause covers caching, training and test logins on any domain tied to your business, and you agree to it by running ads or a business page.

What we would do before October 30:

• Decide your crawler policy, per bot. Meta's crawler docs list meta-externalagent for AI training and meta-externalads for ads. Block the first if you do not want training. Keep the second if you run catalog ads. Meta says robots.txt can be cached for up to 24 hours.
• Audit what sits behind a login. Wholesale price lists, B2B portals, customer order pages. If a test account could see it, assume Meta's bots could too. Put real authorization on it, not only a signup form.
• Log the bots. Check your server or CDN logs for Meta user agents after October 30. Then you know if your robots.txt is respected.
• Read your other platform terms. Meta will not be the last vendor to put crawl rights in a business agreement.

Sources: Meta Commercial Terms (preview), Nieman Lab, Meta web crawler docs.