Context
Cloudflare introduced Redirects for AI Training, a paid-plan AI Crawl Control feature that issues HTTP 301 redirects to same-origin canonical URLs for verified AI training crawlers when pages contain non-self-referencing canonical tags.
What changed
Made publisher-declared canonical destinations enforceable for a defined training-crawler category rather than relying only on advisory HTML signals. Cloudflare also added AI crawler response-code analysis in Radar.
Why it matters
Separates the governance of model-training crawlers from live search crawlers and browsing agents, giving publishers more precise control over which version of content is fetched for training.
Visibility OS interpretation
Visibility OS interpretation: training-data freshness is distinct from search indexing and AI citation visibility. A 301 response to training bots cannot be interpreted as an immediate improvement to model answers.
What remains uncertain
Will retraining on current content reduce outdated model answers over time? How do unverified crawlers behave? Cloudflare's first-party redirect metrics do not independently establish downstream model changes.