By Matt G. Southern
Publication Date: 2025-12-19 17:40:00
Microsoft has shared new guidance on duplicate content that’s aimed at AI-powered search.
The post on the Bing Webmaster Blog discusses which URL serves as the “source page” for AI answers when several similar URLs exist.
Microsoft describes how “near-duplicate” pages can end up grouped together for AI systems, and how that grouping can influence which URL gets pulled into AI summaries.
How AI Systems Handle Duplicates
Fabrice Canel and Krishna Madhavan, Principal Product Managers at Microsoft AI, wrote:
“LLMs group near-duplicate URLs into a single cluster and then choose one page to represent the set. If the differences between pages are minimal, the model may select a version that is outdated or not the one you intended to highlight.”
If multiple pages are interchangeable, the representative page might be an older campaign URL, a parameter version, or a regional page you didn’t mean to promote.
Microsoft also notes that many LLM experiences are grounded in search…
