liveBook · Manning

Deletion-aware pipeline

As explained in AI Governance

A deletion-aware pipeline is a data-processing sequence designed to propagate deletion requests through the data it processes and stores.

A deletion-aware pipeline carries erasure requests into the data lifecycle of an in-house model. It supports executing requests and tracking whether deleted data’s influence is removed, while acknowledging limits on reliable model-level unlearning. section 5.5.4

Account for data and safeguards across the model lifecycle:

  • Maintain training-data provenance records and document the scope and limits of unlearning. section 5.5.4
  • Include derivative artifacts such as checkpoints, embeddings, evaluation datasets, and backups in governance. section 5.5.4
  • Keep fallback mechanisms such as output filtering active while technical deletion remains incomplete. section 5.5.4

How AI Governance covers it

How deep it goes
Identifies pipeline construction and provenance records, but gives no implementation steps and acknowledges limits of model-level unlearning.
Examples
Names checkpoints, embeddings, evaluation datasets, and backups as artifacts that may retain personal-data traces.
What you can do after
Build deletion-aware pipelines, document provenance and unlearning limits, and maintain fallback filtering while technical deletion remains incomplete.
How it connects
Connects data provenance, machine unlearning, checkpoints, embeddings, evaluation datasets, backups, and output filtering.

Related concepts

This explainer was generated by Manning's knowledge graph from the text of AI Governance. The section links point to the parts of the book it draws on.