AI & Automation

# Volume 08: AI and Automation
> **Document ID:** V08-AIA-README
> **Version:** 1.0.0
> **Owner:** AI and Automation
> **Status:** Approved
> **Last reviewed:** 2026-07-26

## Purpose
This volume is the governing operating system for designing, building, approving, operating, and retiring deterministic automations, AI-assisted workflows, bounded agents, and multi-agent orchestration.

## Start Here
1. Read the [Department Charter](./01-Department-Charter.md).
2. Use the [Operating Model](./02-AI-and-Automation-Operating-Model.md) and [Execution Mode Selection](./06-Execution-Mode-Selection.md).
3. Classify autonomy with [Risk Tiers and Permission Envelopes](./10-Autonomy-Risk-Tiers-and-Permission-Envelopes.md).
4. Build from the [SOP library](../../sop/automation/README.md).
5. Require [QA and Acceptance](./26-AI-and-Automation-QA-and-Acceptance.md) before release.
6. Operate with [Observability](./27-Observability-Tracing-and-Audit-Evidence.md) and [Recovery](./31-Kill-Switch-Rollback-Reconciliation-and-Recovery.md).

## Five Execution Modes
| Mode | Use when | Default authority |
|---|---|---|
| Human work | Judgment, empathy, accountability, or novelty dominates | Human performs and decides |
| Deterministic automation | Rules and inputs are stable and testable | System executes explicit rules |
| AI assistance | AI improves speed or synthesis while a human decides | AI drafts, classifies, or recommends |
| Bounded agent | A decision loop needs tools inside a narrow envelope | Agent acts within explicit limits |
| Orchestration | Multiple specialties or independent checks improve outcomes | Coordinator controls typed handoffs |

## Non-Negotiable Boundaries
Humans retain approval for production release, external messaging or publication, spending and commercial commitments, credentials and permissions, sensitive-data use, destructive actions, rights-affecting decisions, autonomy expansion, and residual risk. No solution ambition overrides truth, consent, law, security, privacy, or recoverability.

## Navigation
1. [Department Charter](./01-Department-Charter.md)
2. [AI and Automation Operating Model](./02-AI-and-Automation-Operating-Model.md)
3. [Strategy Outcomes and Use Case Portfolio](./03-Strategy-Outcomes-and-Use-Case-Portfolio.md)
4. [Process Discovery and Automation Suitability](./04-Process-Discovery-and-Automation-Suitability.md)
5. [Solution Space and Ambitious Architecture](./05-Solution-Space-and-Ambitious-Architecture.md)
6. [Execution Mode Selection](./06-Execution-Mode-Selection.md)
7. [Deterministic Automation Engineering](./07-Deterministic-Automation-Engineering.md)
8. [Agent Architecture](./08-Agent-Architecture.md)
9. [Multi-Agent Orchestration and Handoffs](./09-Multi-Agent-Orchestration-and-Handoffs.md)
10. [Autonomy Risk Tiers and Permission Envelopes](./10-Autonomy-Risk-Tiers-and-Permission-Envelopes.md)
11. [Human in the Loop and Approval Design](./11-Human-in-the-Loop-and-Approval-Design.md)
12. [Prompt and Context Engineering](./12-Prompt-and-Context-Engineering.md)
13. [Structured Outputs and Validation](./13-Structured-Outputs-and-Validation.md)
14. [Memory and State Management](./14-Memory-and-State-Management.md)
15. [Knowledge Retrieval and RAG](./15-Knowledge-Retrieval-and-RAG.md)
16. [Model Evaluation Selection and Routing](./16-Model-Evaluation-Selection-and-Routing.md)
17. [Tool Function and MCP Governance](./17-Tool-Function-and-MCP-Governance.md)
18. [Integration Webhook and Event Contracts](./18-Integration-Webhook-and-Event-Contracts.md)
19. [Data Governance Privacy and Retention](./19-Data-Governance-Privacy-and-Retention.md)
20. [AI Security Prompt Injection and Exfiltration Defense](./20-AI-Security-Prompt-Injection-and-Exfiltration-Defense.md)
21. [Secrets Identity Tenant and Environment Isolation](./21-Secrets-Identity-Tenant-and-Environment-Isolation.md)
22. [Sandboxes Code Execution and Computer Use](./22-Sandboxes-Code-Execution-and-Computer-Use.md)
23. [Queues Idempotency Retries and Concurrency](./23-Queues-Idempotency-Retries-and-Concurrency.md)
24. [Evaluation Datasets Metrics and Thresholds](./24-Evaluation-Datasets-Metrics-and-Thresholds.md)
25. [Simulation Red Teaming and Adversarial Testing](./25-Simulation-Red-Teaming-and-Adversarial-Testing.md)
26. [AI and Automation QA and Acceptance](./26-AI-and-Automation-QA-and-Acceptance.md)
27. [Observability Tracing and Audit Evidence](./27-Observability-Tracing-and-Audit-Evidence.md)
28. [Cost Latency Capacity and Budget Controls](./28-Cost-Latency-Capacity-and-Budget-Controls.md)
29. [Reliability Degraded Modes and Circuit Breakers](./29-Reliability-Degraded-Modes-and-Circuit-Breakers.md)
30. [Deployment Release and Autonomy Expansion](./30-Deployment-Release-and-Autonomy-Expansion.md)
31. [Kill Switch Rollback Reconciliation and Recovery](./31-Kill-Switch-Rollback-Reconciliation-and-Recovery.md)
32. [Incidents Problem Management and Learning](./32-Incidents-Problem-Management-and-Learning.md)
33. [Maintenance Drift and Continuous Evaluation](./33-Maintenance-Drift-and-Continuous-Evaluation.md)
34. [Roles KPIs and Capacity](./34-Roles-KPIs-and-Capacity.md)
35. [Internal and Client Delivery](./35-Internal-and-Client-Delivery.md)
36. [AI and Automation Maturity Model](./36-AI-and-Automation-Maturity-Model.md)
37. [Glossary](./37-Glossary.md)
38. [Revision History](./38-Revision-History.md)

## Execution Assets
- [SOPs](../../sop/automation/README.md)
- [Templates](../../templates/automation/README.md)
- [Checklists](../../checklists/automation/README.md)
- [Prompts](../../prompts/automation/README.md)
- [Examples](../../examples/automation/README.md)
- [Diagrams](../../diagrams/automation/README.md)
- [Specialist capability tests](../../agents/tests/automation/README.md)

## Governance
Volume 0 and Volume 1 remain superior authority. This volume governs department execution and does not grant a tool, model, vendor, employee, or agent authority beyond an approved role and permission envelope.