Whose login was it?
A personal email, a shared login, or a managed seat? Can the company see the history, remove access, and know what happens when the person leaves?
Shadow AI / staff and client data
Your staff are already using AI. Can the company see the path?
If people are using personal or unmanaged ChatGPT accounts for work, a blanket ban does not create a usable alternative. We map the real workflows, classify the data, assess whether a managed team plan fits, and set up an approved route when a private assistant is the better answer.
Use-case mapData-path reviewApproved routeStaff handover
Pattern demonstration / not a staff audit
The question nobody wants to answer
Who put the client file into ChatGPT?
Usually someone was trying to finish a real task. Now the partner, IT lead, or privacy lead has to work out which account was used, what crossed the boundary, and what staff will use tomorrow.
What exactly was pasted?
A PDF, working paper, matter note, or prompt may have been copied to get a faster answer. If nobody recorded the upload, the firm may not be able to reconstruct the path.
What will staff use tomorrow?
Telling people to stop does not answer the client question due at 4pm. Without an approved tool that is usable in the same moment, the next upload happens somewhere less visible.
Three decisions after the upload
Do not solve a workflow problem with a policy PDF.
The right answer may be a managed team route, a private workflow, or a split between ordinary work and restricted client material. It starts with the actual use case and current provider terms.
Use a team plan when it fits
A managed team or enterprise workspace may be enough when account ownership, access, retention, contractual terms, and administration meet the firm's requirements.
Deploy the sensitive workflow
A private assistant can keep the application, retrieval layer, and document store in client-controlled infrastructure when the data path requires more control.
Separate low-risk and restricted work
Public research and ordinary drafting may use one route while client files, matter documents, and restricted records use another.
From shadow use to an approved path
Start with what staff actually do.
Map the real use cases
Ask who is using AI, for which tasks, with which data, on which accounts, and what output returns to the business.
Classify the information
Separate public material, internal information, client files, personal data, matter documents, and records that need a tighter route.
Assess the available route
Review the current managed-plan terms and controls, or compare them with a client-owned private assistant and its actual data path.
Give staff a usable alternative
Deploy and test one approved workflow, publish the boundary, train the team, and hand over the accounts and update process.
The question is not whether staff will use AI. It is whether the company has given them a usable path for the work that matters.
Why trust Pristine3D?
We build and operate production software.
Pristine3D Ltd builds and operates live digital products, and we run private AI workflows internally as part of our own operations. We scope around your real workflow: the documents you own, the questions your team asks, and the access boundary you approve. Based in Lagos, Nigeria, we work remotely with clients worldwide.
We already run live products
Based in Lagos, Nigeria, Pristine3D currently builds and operates smartcards.ng, venu.ng, photoshoot.ng, and ugc.ng in production.
We start with the actual workflow
One input, one output, one test set, and one person who owns the result. We scope a real workflow instead of a transformation programme.
The boundary stays visible
Cloud, model, storage, and messaging accounts stay in your name. The chosen data path, access rules, test record, documentation, and training are part of the agreed scope.
Pricing / fixed scope
Know the starting numbers before you ask.
The final quote follows the workflow. Infrastructure and model bills stay on your accounts.
One Workflow
Starting from$3,500 / ₦2.5m
2 to 3 weeksOne workflow for a small team, with training and handover.
See the one-workflow packageFirm-wide
Starting from$7,500 / ₦8m
4 to 6 weeksMultiple sources, roles, integrations, and admin handover.
See the deploymentAnnual support
Starting from$3,000 / ₦1.5m
per yearStandard care for one delivered workflow. Optional. Larger deployments and active monitoring are separately scoped.
See supportArchitecture review from $500. Standard annual support is $3,000 / ₦1.5m per year for one delivered workflow. New workflows, integrations, active monitoring, and infrastructure are separately scoped; infrastructure, model, storage, and messaging bills stay on client accounts. Full pricing and what changes the quote
Straight answers
Common questions.
Does this mean ChatGPT Team or Enterprise is automatically unsafe?
No. There is no blanket answer. The relevant questions are account ownership, provider terms, retention, access, audit needs, data classes, and the company's own policies. We help assess the route rather than making an unsupported guarantee.
Do we have to replace ChatGPT?
No. If a managed team route fits the requirements, it may remain part of the approved setup. A private assistant is useful when a specific workflow or data path needs more control.
Can you tell us what staff already uploaded?
We can help map the process and design controls, but a reliable historical review depends on the accounts, provider logs, retention, and access records the company actually controls.
What should we do first?
Choose one real use case and one data class. Do not start by uploading more sensitive files. First agree the route, source, users, permissions, and review boundary.
Own your knowledge base
The model is not the product. The knowledge base is.
Documents, the retrieval index, access rules, and the workflows built around them are the asset, and they compound. We deploy so the knowledge base stays yours: on your accounts, in the environment you choose, under access rules your team defines. The model behind the answers is a connector, so the knowledge base moves with you, not with a vendor.
Your corpus, your index
The document store, metadata, and retrieval setup live on accounts you own. No vendor holds the corpus.
Models are swappable parts
Change the model provider, move regions, or go local without rebuilding the knowledge base or the workflow.
The knowledge base is the alpha
Every improvement to the corpus improves the answers, and the improvement stays with you, not with a vendor.
Keep exploring
Related setups.
Start with the actual use
What are staff already using AI to do?
Tell us the task, account type, information involved, and the route the company wants staff to use instead.