Compliance Operations
AI Compliance Knowledge System
Organizes regulations and internal controls into a searchable knowledge base.
Compliance Operations
AI Data Governance Assistant
The data your AI systems use is classified, access rules are applied, and where it came from stays traceable as it moves.
AI tools are only as safe as the data you feed them. This assistant classifies the data your AI systems use as public, internal, confidential or restricted, applies the access and handling rules you set, and flags sensitive data before it is sent to a model. It also records where each dataset came from as it moves. Your governance owner keeps control.
- Data feeding your AI classified against your own categories
- Sensitive data flagged before it reaches a model
- A traceable record of where each dataset came from
$5,900
20–30 working days
Optional maintenance $295/mo
What it is
AI Data Governance Assistant classifies the data feeding your AI systems, applies the access and handling rules you set, and keeps a record of where each dataset came from as it moves between systems. Sensitive data is flagged before it reaches a model, so problems are caught early.
Why businesses buy it
Most AI data incidents start with a reasonable request: someone connects a folder, a spreadsheet or a mailbox to a new tool without checking what is inside. Once restricted data reaches a model, getting it back is hard. Classifying data and applying rules up front is cheaper than any cleanup. This assistant makes those rules consistent and visible, whichever model you choose. The written scope sets the systems and categories covered, so you can approve the work with confidence.
Problems this solves
Sensitive data can reach an AI model before anyone notices
The assistant classifies data feeding your AI systems and flags sensitive items before they reach a model. Problems are caught at the start of the flow, not after the fact.
Nobody can say where a dataset used by AI came from
It keeps a record of each dataset's origin as it moves between up to three systems. When someone asks where data came from, the answer is already written down.
Data handling rules exist but are applied inconsistently
The access and handling rules you set become up to five configured controls. They apply the same way every time, instead of depending on who handles the data.
A good fit if
AI systems in your business draw on data from more than one source and you need to know which rules apply to each.
Some of the data feeding your AI tools is more sensitive than the rest, and access needs to reflect that.
You need to trace a dataset back to where it came from, especially if a customer or regulator asks.
You already have an AI governance owner but data handling has been left to whichever team touches it first.
Not the right fit if
You need to govern which models get approved, not the data feeding them. See AI Model Governance.
You need the written usage policy itself, not data-specific classification and access rules. See AI Usage Policy System.
You need the full governance build spanning policy, risk and audit, not just data. See AI Governance System.
You need a record of AI decisions and outputs, not the data that fed them. See AI Audit Trail.
How it works
- We map the data your AI systems actually use. Where it comes from, which systems hold it, such as Google Workspace or Microsoft 365, and how sensitive it is.
- Data is classified against categories you set. Public, internal, confidential or restricted, so handling rules can be applied consistently rather than judged case by case.
- Access rules are configured per category. Who can see or use restricted data, and which AI workflows are allowed to touch it at all.
- Lineage is tracked as data moves. Each dataset carries a record of where it originated and what it has passed through, kept alongside the classification.
- Rules are tested against real workflows. A handful of actual AI processes are checked to confirm access and handling behave as agreed, not just as documented.
- A findings report closes out the build. Confirming classification, access rules and lineage tracking all work against your agreed scenarios.
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.
- A list of the data sources feeding your AI systems today.
- Any existing data classification or handling policy, even an informal one.
- A governance owner who can confirm which categories and access rules make sense for your data.
- Written authorization for any testing of the classification and access rules.
- 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 categories and access rules.
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 your data sources named.
Weeks 1–2
Data mapped and classification categories agreed with your governance owner.
Week 3
Access rules and lineage tracking configured across systems in scope.
Weeks 3–4
Rules tested against real workflows and findings reported back to you.
Week 4
Your review and two rounds of changes to categories or rules.
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 Data Governance Assistant configured and deployed across up to three systems
- Data classification and handling rules set as up to five controls
- A lineage record showing where each in-scope dataset came from
- 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 lower the risk of personal or confidential data ending up in an AI tool.
- Helps you answer data origin questions from clients or auditors quickly.
- Makes data rules apply consistently across systems.
- Sensitive items flagged before reaching a model
- Count flags each month in the console and check how many were real issues.
- Datasets with a recorded origin
- Compare in-scope datasets with a lineage record against those without, at launch and each quarter.
- Data handling incidents
- Count incidents involving AI systems in the quarter before launch and the quarter after.
Before you start: List the datasets that feed your AI systems and who can access each. Note any data incidents from the past year.
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 classification and access rules afterwards costs roughly $20 to $300 a month in your own accounts, mostly hosting for wherever the rules and lineage records are held.
AI usage during build and testing also bills to your own accounts, so every provider relationship stays yours.
Ask about running costsCustom requirements
Need to classify data across more than three connected systems, or set more than five 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
Monthly maintenance covers monitoring, fixes, updates, a health report and change hours for things such as a new data category or an adjusted access rule as your data sources change. Maintenance 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 data catalog tool bundles classification into a broader subscription | A data governance consultancy quotes for a classification project | Ours is one fixed price, shown on this page before you buy |
| Scope | A catalog tool classifies data generally, not specifically for AI use | An agency's scope depends on how the engagement is billed | Ours is written to the data feeding your AI systems specifically |
| Delivery | A catalog tool still needs real tagging work before it is useful | A classification project can run several weeks | Ours has a fixed delivery time shown before you buy, with faster options |
| Lineage | A catalog tool tracks lineage only if you configure it fully yourself | An agency deliverable depends on what was agreed | Ours tracks lineage as part of the build, tested against real workflows |
Questions about this product
How much does AI Data Governance Assistant cost?
The build is one fixed price for everything listed on the Scope Card: mapping your data, setting up to five categories or rules, configuring access and lineage tracking across up to three systems, and a recorded handoff. The price panel on this page shows the figure at each delivery speed. Running costs afterwards are paid directly in your own accounts.
What does AI Data Governance Assistant cost to run each month?
Running the classification and access rules afterwards costs roughly $20 to $300 a month in your own accounts, mostly hosting for wherever the rules and lineage records are held. There is no separate license fee on top of that figure. The exact cost depends on how many systems and data sources are in scope.
How is AI Data Governance Assistant different from AI Model Governance?
AI Data Governance Assistant governs the data your AI systems use: its classification, access rules and lineage. AI Model Governance instead governs which models get approved for which use, which is a separate decision about the tool rather than the data feeding it. Organizations with sensitive data often start here, since data rules apply regardless of which model is chosen.
What happens if AI Data Governance Assistant flags a data access issue?
The flag names the dataset, its classification and the rule it conflicts with, then routes to your governance owner for review. The assistant applies the access rules you set; it does not decide on its own whether to grant or revoke access, because a change like that needs a qualified person to weigh the request against your actual policy.
Can we add data categories or access rules later?
Yes. Adding a category, adjusting who can access a classification, or updating lineage tracking for a new source usually fits inside the monthly change hours that come with maintenance. Classifying data across a fourth connected system beyond the three in scope, or a sixth category, goes through a Change Quote agreed before that work starts on your build.
Can we cancel AI Data Governance Assistant 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. Once your data categories have settled into a stable, predictable pattern, you may find you need it less.
Who owns the data classification and access rules once they are live?
You do. The classification categories, access rules and lineage records all live inside your own connected systems, not a platform we control. We build the structure and hand it over with an admin guide and a recorded session, and every rule and category stays fully visible and editable by your governance owner or IT team.
Know what data your AI systems are actually using.
Order AI Data Governance Assistant and your data classification, access rules and lineage tracking are live on a fixed delivery date.
Works well with
Compliance Operations
AI Policy Assistant
Answers staff questions about AI and data policies with citations.
