liveBook · Manning

Compliance

As explained in Architecting Autonomous Agentic Systems

Compliance is continuous alignment between information use and applicable enterprise, legal, regulatory, policy, and evidentiary obligations.

Compliance is continuous alignment between information use and enterprise obligations, not a label on stored data or a review after work is done. A legal agent may use a privileged email for a lawyer’s review but be barred from sending it to an external provider, retaining it in memory, or retrieving it across matters. section 5.6

Carry obligations through the information lifecycle:

  • Enforce them during retrieval, transformation, context assembly, generation, tool use, disclosure, retention, reuse, and action. section 5.6
  • For legal work, respect jurisdiction, privilege, evidentiary and citation standards, authority limits, approval requirements, and episode reconstruction. section 5.6
Table 5.6 Verifying and managing agentic data quality
Quality dimension
What you should verify and manage
Accuracy
Reconcile source data with authoritative records, test transformation fidelity, and run versioned evaluation datasets against retrieval, grounding, factual claims, tool use, and expected outcomes.
Completeness
Test whether required fields, sources, evidence, exceptions, and policy inputs are present for the decision surface.
Timeliness
Monitor freshness, effective dates, supersession, pipeline delay, retrieval-index lag, and context expiry.
Consistency
Detect conflicting values, duplicates, incompatible formats, broken relationships, and disagreement across sources.
Traceability
Connect sources, transformations, retrieval, model and tool activity, policy decisions, and the final decision episode.
Privacy and confidentiality
Discover, classify, mask, redact, and restrict personal, privileged, confidential, or regulated information.
Compliance
Evaluate policy conditions, processing restrictions, evidence requirements, approval routes, and control outcomes.

Use the verification dimensions to make information conditions testable, including traceability and privacy controls. section 5.6

How Architecting Autonomous Agentic Systems covers it

How deep it goes
Traces obligations across the information lifecycle and names runtime controls, but does not detail their implementation.
Examples
Uses a legal-agent dispute involving privilege, jurisdiction, external communication, and evidence reconstruction.
What you can do after
Operate agent workflows by enforcing obligations across retrieval, generation, tool use, disclosure, retention, and action.
How it connects
Connects compliance to Policy as Data, pre-agent guardrails, Context Graphs, and Trusted Context.

Also explained in

Compare how 16 books explain compliance →

Related concepts

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