Aeon AI Risk Management

Decide. Operate. Prove.

Three practical books for putting AI to work without handing over judgment. Read one for the decision in front of you, or connect all three from investment to operation to evidence.

The AI Investment File

How Leaders Find, Buy, and Implement AI That Pays. A practical route from finding valuable work to making an investment decision. Build the AI Investment File around results, economics, capability, evidence, information, people, and tracking. Inside this book: A business result worth pursuing. Capability evidence and full operating cost. Information, permissions, ownership, and stop-or-scale criteria. Working record: AI Investment File.

The Supervised Digital Workforce

From Shadow AI to Shared Advantage. Turn scattered AI use into a supervised way of working. Map the task, define human authority, manage exceptions, and preserve the skills people need when automation is wrong or unavailable. Inside this book: Task boundaries and human decision rights. Verification, exceptions, and recovery paths. Assisted, unassisted, and recovery capability. Working record: Digital Workforce Control File.

Trust Is a Product Feature

Release Gates for the Governable Corporation. Make trust inspectable through bounded claims, release gates, runtime signals, and recovery paths. Build an AI Trust Dossier that separates evidence quality, control status, and the decision itself. Inside this book: Bounded claims and release conditions. Evidence age, decision rights, and reserved matters. Monitoring, expiry, and recovery expectations. Working record: AI Trust Dossier.

One decision. A connected record.

Start with the AI Investment File. Carry the approved scope into the Digital Workforce Control File. Bring operating evidence into the AI Trust Dossier. Each book stands alone.

Start with a working sheet.

Free starter worksheets, without a purchase, email gate, or workshop. These concise working aids adapt the books' methods; they are not the full books. Keep sensitive information in approved systems. Completing a sheet does not establish that a control works or that a decision is approved.

AI Investment Receipt

Frame the business outcome, full cost, evidence gaps, and next investment decision. Prompts: Decision, sponsor, owner, date, and review date; Result: target business outcome and current baseline; Economics: build, run, integration, verification, and recovery costs; Capability and control: task boundary, human authority, and limits; Evidence: source, date, test conditions, limitations, and gaps; Information: data needed, permitted use, access, and retention; People, process, permissions: owners, workflow changes, and approvals; Tracking: outcome measures, review cadence, and accountable reviewer; Stop or scale: thresholds, next decision, and unresolved conditions.

Workflow Control Sheet

Define the task, human authority, verification steps, exceptions, and skills to retain. Prompts: Workflow, owner, affected roles, date, and review date; Task boundary: inputs, permitted actions, outputs, and excluded work; Human authority: who reviews, approves, overrides, and stops; Information: approved tools, data access, and handling rules; Verification: what is checked, by whom, against which evidence; Exceptions: triggers, escalation owner, and response expectation; Recovery: manual fallback, restoration steps, and exercise evidence; Skills plan: assisted, verification, unassisted, and recovery exercises; Operating record: incidents, outcomes, open gaps, and next review.

Release-Gate and Board Decision Brief

Connect a bounded claim to evidence, accountable decision-makers, conditions, and follow-up. Prompts: Decision requested, system scope, sponsor, date, and expiry; Bounded claim: what is being asserted and what is excluded; Authority: management delegation, reserved matters, and decision owner; Release criteria: required checks and acceptable thresholds; Evidence: sources, dates, limitations, contradictions, and open gaps; Control state: operating, partial, untested, or absent, with evidence; Decision state: pending, go, conditional go, or no-go, with rationale; Conditions: owner, deadline, expiry, and escalation for each; Runtime monitoring, stop triggers, recovery, and next review.