A machine-readable reservation is a rights-holder signal, usually embedded in metadata or another structured format, that tells automated systems not to use a work for text and data mining or similar processing. Unlike a human-readable notice, it is designed so crawlers, dataset builders, and compliance tools can detect and honor it automatically.
In cyber security and AI operations, this matters because data ingestion is part of the attack surface. If a model training pipeline ignores opt-out signals, it can create legal exposure, policy violations, and weak provenance controls. Defenders use reservation checks, source filtering, and audit logs to prove that protected material was excluded. Attackers may try to hide reserved content inside mixed datasets or stripped metadata, so robust pipelines validate reservations before storage, training, or indexing.