Compliance Operations
AI Compliance Knowledge System
Organizes regulations and internal controls into a searchable knowledge base.
Compliance Operations
AI Audit Trail
Every AI input, output and decision in scope is recorded as it happens, preserved for later review.
When an auditor, regulator or customer asks what your AI did and why, you need a record, not a recollection. This audit trail captures the inputs, outputs and decisions from the AI systems in scope as they happen, in a structured format anyone can read. Records are kept apart from the systems they describe, searchable by application and date, and stored in your accounts.
- Every AI input, output and decision recorded as it happens
- Records searchable by AI application and date
- Logs kept apart from the systems they describe
$5,900
20–30 working days
Optional maintenance $295/mo
What it is
AI Audit Trail records the inputs, outputs and decisions from your AI systems as they happen, preserving them in a structured log you can pull for a later audit, incident review or regulator request. Records are stored apart from the AI tools themselves, so changing vendor never erases your history.
Why businesses buy it
Without a record, every question about an AI decision turns into a week of guessing and an uncomfortable answer. That matters in a customer dispute, an incident review or a regulator request, and it matters most when you cannot predict which decision will be questioned. Logs inside each AI tool are often incomplete or deleted on a schedule. This audit trail keeps a consistent, separate record you control. The written scope sets which systems and log types are covered before you buy.
Problems this solves
Nobody can show what an AI system said or decided last month
The audit trail records inputs, outputs and decisions from up to three AI systems as they happen. You can pull the exact record later for an audit, incident review or regulator request.
Switching AI vendor would wipe the history you rely on
Records are stored apart from the AI tools themselves. If you change provider, the structured log stays with you rather than leaving with the old vendor.
Incident reviews stall while someone digs through raw logs
Entries are kept in a structured format you can search and export. The admin guide and recorded handoff show your team how to pull a record when it is needed.
A good fit if
You need to be able to show what a specific AI system did and decided on a specific date, after the fact.
An auditor, regulator or internal investigation could reasonably ask for a record you do not currently keep.
You want a permanent log kept separately from the AI system itself, so it survives a change of vendor or model.
You already have monitoring or governance in place and need the record-keeping piece to sit alongside it.
Not the right fit if
You need active alerts when something looks wrong, not a passive record. See AI Agent Risk Monitor, or AI Compliance Monitoring.
You need a summarized status view, not the underlying log itself. See AI Governance Dashboard.
You need the full governance build including policy and risk, not just the audit record. See AI Governance System.
You need to know which AI tools exist before deciding what to log. See AI Model Inventory.
How it works
- We agree what gets logged with you. Which AI systems, what counts as a decision worth recording, and how long records need to be kept.
- Logging is connected to the systems in scope. Typically the AI applications themselves plus Google Workspace, Microsoft 365 or Jira, wherever the relevant actions actually happen.
- Each entry is structured, not a raw dump. Timestamp, system, input, output and the decision made, recorded in a consistent format an auditor can actually read.
- The log is kept separate from the systems it watches. A later change of AI vendor or model does not take the historical record with it.
- Retrieval is tested against real scenarios. Pulling a specific record for a specific date, the kind of request an audit or investigation actually makes.
- A test report confirms it works as agreed. Covering the scenarios you named, delivered alongside the admin guide at handoff.
What you provide within 72 hours
Your portal lists these the moment you order. Delivery starts the working day after everything is in and your access is checked.
- Your business information (company basics, hours and brand voice), shared during intake, or confirmation we can reuse what you gave us on an earlier order.
- The AI tools and systems whose actions and decisions should be logged.
- Any internal policy that defines what counts as a decision worth recording.
- A governance owner who can confirm retention requirements and log categories with us.
- Written authorization for any testing of the logging connections.
- Admin access and API keys for every connected system, shared by adding our team as a user or creating a key you control, never by email.
- One decision-maker who signs off the final log structure before handoff.
The 72-hour rule. Send everything on your checklist, including access, within 72 hours of ordering so your build starts on time.
Timeline
Start
business information confirmed and the systems to be logged agreed.
Weeks 1–2
Logging connections built and structured entries defined with your governance owner.
Week 3
Retention rules and log categories configured across the systems in scope.
Weeks 3–4
Test report produced by pulling real records for agreed scenarios.
Week 4
Your review and two rounds of changes to log structure or categories.
Go-live
Live, recorded handoff session, and the 30 days of support begins.
| Option | Delivery time | Price |
|---|---|---|
| Standard | 20–30 working days | $5,900 |
| Express | 15–25 working days | $6,800 |
| Rush | 10–20 working days | $7,650 |
What you receive at handover
- AI Audit Trail configured and deployed, recording inputs, outputs and decisions from up to three AI systems
- A structured log stored separately from your AI tools, with up to five configured controls
- Workflow, prompt and guardrail configuration and connections to the agreed systems
- A test report with a control test log for the agreed scenarios
- An admin guide and a recorded 60-minute handoff session
- 30 days of post-launch support
Expected benefits and how to measure them
- Can make it much quicker to answer an auditor or regulator about a past AI decision.
- Helps incident reviews start from facts rather than memory.
- Keeps your AI history safe if you change vendors.
- Time to retrieve a past AI decision
- Pick a sample request and time how long it takes to find the record. Compare with the effort before launch.
- Logging coverage
- Check each month that every in-scope system is writing records, using the console's record counts.
- Gaps found during reviews
- Count cases where a needed record was missing during audits or incidents, before and after launch.
Before you start: Try to retrieve one AI decision from last month and note how long it takes. List where AI inputs and outputs are kept today.
These are things to track, not promised results. Outcomes depend on your data, your team and your market.
Running costs
The build price is one-time. Running the log afterwards costs roughly $20 to $300 a month in your own accounts, mostly storage and hosting for the records themselves.
AI usage during build and testing also bills to your own accounts, so every provider relationship stays yours, not ours.
Ask about running costsCustom requirements
Need to log more than three connected systems, or more than five log categories? Add a note at checkout for a written Change Quote, while your original order keeps its normal timeline. Larger or bespoke projects use the Custom AI request form.
Request a custom versionKeep it running
We recommend monthly maintenance: monitoring, fixes and a monthly allowance of change hours, which covers adding a new system to the log or adjusting retention as requirements change. It starts after the 30 days of support.
$295/mo (5% of the build price a month, includes 2 change hours)
Optional. Starts after the post-launch support period. Cancel any month.
How maintenance worksHow this compares
| Do it yourself with a SaaS tool | Hire a custom agency | Upmark Labs | |
|---|---|---|---|
| Price | A general logging platform charges per event or per gigabyte stored | An agency quotes for building custom logging | Ours is one fixed price, shown on this page before you buy |
| Scope | A logging platform captures whatever you send it, structured or not | An agency's scope depends on how the engagement is billed | Ours is structured around AI inputs, outputs and decisions specifically |
| Delivery | A logging platform still needs real integration work before it is useful | An agency engagement can run several weeks | Ours has a fixed delivery time shown before you buy, with faster options |
| Retrieval | A logging platform leaves you to build your own queries | An agency deliverable depends on what was agreed | Ours is tested against the specific scenarios you named |
Questions about this product
How much does AI Audit Trail cost?
The audit trail is built for one fixed price covering the Scope Card: agreeing what gets logged, connecting up to three systems, structuring entries and setting retention and retrieval, with an admin guide and recorded handoff. You can check the figure at each delivery speed in the price panel on this page before ordering. Storage afterwards is paid in your own accounts.
What does AI Audit Trail cost to run each month?
Running the log afterwards costs roughly $20 to $300 a month in your own accounts, mostly storage and hosting for the records themselves. The exact figure depends on how much activity your logged systems generate and how long you choose to retain records. You keep the storage account in your own name throughout.
How is AI Audit Trail different from AI Compliance Monitoring?
AI Audit Trail is passive: it records what happened so you can retrieve it later, without evaluating whether anything was wrong. AI Compliance Monitoring actively checks AI workflows against your internal compliance rules and flags issues as they occur. Many buyers run both together, since a compliance flag is far more useful when the full record behind it is already preserved.
What happens if an AI Audit Trail record shows something concerning?
The audit trail preserves the record; it does not interpret it or decide whether something was wrong. If a review of the log raises a concern, that assessment sits with your governance owner or whoever runs the audit, using the preserved record as evidence, because judging what a decision meant is a job for a qualified person, not the log itself.
Can we log more systems or change retention later?
Yes. Adjusting retention, adding a new log category, or refining what counts as a loggable decision usually fits inside the monthly change hours that come with maintenance. Logging a fourth or fifth system beyond the three in scope goes through a Change Quote agreed before that extra work starts, so pricing stays clear from the outset every time.
Can we cancel the AI Audit Trail maintenance later?
Yes. Maintenance has no lock-in, so you can cancel any month once the 30 days of included support has ended. Unused hours do not roll over. Because a log runs quietly once it is set up, you can decide after the first few months whether you still need it.
Who owns the AI Audit Trail log once it is live?
You do. Records are stored in your own accounts, kept separate from the AI systems they log, so nothing is lost if you later change vendor or model. We build the structure and the retrieval process, then hand it over with an admin guide and a recorded session, and the full history stays accessible and exportable by your team.
Keep a record that survives a change of vendor.
Order AI Audit Trail and a structured, retrievable record of your AI systems' actions is live on a fixed delivery date.
Works well with
Compliance Operations
AI Policy Assistant
Answers staff questions about AI and data policies with citations.
