Knowledge cutoff

A knowledge cutoff is the fixed point in time after which a model's training data ends, so the model has no built-in knowledge of events, pages, or facts that first appeared after it¹. It is a property of how the model was trained, not of the product you are using: two systems built on the same base model share its cutoff, even if one can search the web and the other cannot.

The distinction that matters for AI search is between parametric knowledge and retrieved knowledge. Parametric knowledge is what the model absorbed during training, frozen at the cutoff. Retrieved knowledge is content the system fetches at answer time from outside the model. The cutoff bounds the first and says nothing about the second. A model whose training ended last year can still answer about this morning's news, but only by retrieving it, not by recalling it.

Status in 2026

Major models today ship with a documented cutoff, and in practice it sits months behind the model's release and further behind the moment you actually use it. That lag is the reason the consumer engines added retrieval. Anthropic's own documentation describes its web search tool as giving the model "direct access to real-time web content, allowing it to answer questions with up-to-date information beyond its knowledge cutoff," and states in the same place that "the response includes citations for sources drawn from search results"¹. Read those two sentences together and the GEO opportunity falls out of them: the cutoff creates the need to retrieve, retrieval is how the engine gets current information, and citation is how it credits what it retrieved. The cutoff sits upstream of the citation surface.

The honest corollary is that retrieval is not automatic. For stable knowledge that sits well within its cutoff, an engine can answer from parametric memory alone, with no search and no sources; Anthropic's same web search documentation describes this case directly, noting that Claude answers without searching when a request draws on stable knowledge¹. When that happens there is nothing to cite, which is one reason some queries return no citations rather than citing anyone. So the cutoff does not only enable citation; it also marks the line below which an engine may not bother to retrieve at all.

How to apply

The cutoff is a model property you do not control, but it tells you where citation opportunity concentrates. The work is to aim content at the queries that force an engine to retrieve:

• Target fast-moving and post-cutoff topics. A question the model cannot answer from training (this year's data, a just-shipped feature, an evolving standard) pushes the engine to retrieve, which is exactly where your content can be cited. Long-settled topics are the ones an engine is most likely to answer from memory without citing anyone.
• Keep recency signals honest and current. When an engine retrieves for a time-sensitive query it tends to favor content it can tell is recent, so maintain accurate datePublished / dateModified rather than cosmetic date-bumping (see freshness signals).
• Optimize the retrieval layer, not the model weights. You cannot influence what the next training run absorbs or when the next cutoff lands, and even a brand recalled from parametric memory carries no link. The citable surface is the retrieved answer, so the leverage is in being retrievable and quotable.

What to skip: do not worry about which cutoff date a given model carries, and do not chase parametric inclusion. Both are outside your control and neither produces a citation on its own. Spend the effort on the retrieval layer that does.

How it relates to other concepts

• The cutoff is why retrieval-augmented generation exists: RAG (Lewis et al. 2020) pairs a retriever with the model so an answer can draw on current external content rather than only frozen parametric knowledge². The cutoff is the problem; RAG is the architecture that answers it.
• It defines the job of the retrieval pipeline: fetch, rank, assemble, and attribute all exist to supply the model with what its training left out.
• It is the upstream reason for generative engine optimization: because engines must retrieve to answer beyond the cutoff, there is a citation surface to optimize for at all.
• It is documented engine-side in Claude citations: Anthropic ties web search, the knowledge cutoff, and citations together in one tool description, the clearest vendor statement of the cutoff-to-citation chain.
• It raises the stakes for freshness signals: post-cutoff content competes on recency, and most assistants tend to prefer fresher content for time-sensitive queries.
• It is a standing reason for hallucination grounding: answering a post-cutoff question from stale parametric memory is a common way to produce confident but wrong output, and grounding the answer in retrieved sources is the mitigation.