Wiki · citable explainer
Training cutoffs, retrieval indexes, and API-queryable identity records operate on different timelines.
There is no single index timer. Training lag can be long; retrieval can be days–weeks; structured public records can be observed sooner by systems that fetch live sources.
Is AI indexing the same as Google indexing?
No. Training data, RAG retrieval, and tool/API access behave differently.
What is fastest?
Publishing structured public identity evidence that retrieval systems and tools can consume without waiting for a model retrain.
Can I force ChatGPT to update?
No. You can only improve observable evidence quality.