A client's unlaunched campaign is confidential by contract.
Don't paste it into someone else's model.
Private LLM hosting on a dedicated machine in EU-Central, for marketing and creative agencies drafting and ideating on client campaigns, briefs, and strategy that shouldn't sit on shared AI infrastructure.
Not legal advice. This page describes GPUwerk's infrastructure and contractual commitments only. Whether a given use of client campaign material is permitted under your client contracts, confidentiality clauses, and applicable regulation is a determination for your agency's own legal counsel, made against each client's specific terms.
The account you'd hate to lose, one shared answer
A leaked campaign concept or a client recognising their own confidential brief inside a public AI tool's outputs is the kind of thing that ends an account relationship. That raises the same infrastructure question before anything else: where does the model actually run.
"Strategists paste client briefs into consumer AI tools for a faster first draft"
Drafting positioning language or a creative brief summary in a public AI tool is common, and rarely reviewed for what confidential client detail went in alongside it, or what a vendor's retention and training policy does with it afterward. On a GPUwerk instance, the machine is assigned to your agency alone, in EU-Central. GPUwerk operates the machine but, under the data processing agreement, does not access your content except at your request for support or where a legal obligation requires it.
"An unreleased campaign concept is worth more to a competitor than to us, for a few weeks"
Between concept and launch, a campaign's value is almost entirely in its novelty. A dedicated single-tenant machine means no other tenant's workload, and no vendor model training, is anywhere near that material before it ships.
"Client contracts increasingly name AI tools directly in confidentiality clauses"
More clients now ask agencies which AI tools touch their material and on what terms. A dedicated instance gives you a concrete answer: single-tenant hardware, EU-Central, an Article 28 DPA, and no sub-processors on the workload, rather than a vendor's general-purpose terms of service. See the private ChatGPT setup →
What's actually in the contract
For your agency's legal, IT, or client-services lead to review directly, or to pass along to a client asking how you handle AI tools. This is a description of infrastructure, not a certification record.
| Question | Answer |
|---|---|
| Where does it run? | EU-Central, on a dedicated single-tenant machine assigned to your agency |
| Who operates it? | PRINT IT! SE, a Societas Europaea registered in Prague, Czech Republic. No US parent entity. |
| Who can reach the instance? | Through the instance itself, only holders of your SSH keys; password login is disabled fleet-wide. GPUwerk keeps infrastructure administrator access to the underlying machine, as on any hosted service, and under the DPA does not use it on your content except at your request for support or where a legal obligation requires it. |
| Does GPUwerk read client campaign material? | No. We host the hardware and, under the DPA, do not access, read, copy, index or analyse workload content. |
| Sub-processors for the workload? | None, listed at /legal/sub-processors |
| DPA (GDPR Art. 28)? | Published at /legal/dpa, no charge |
| Data on termination? | Container and workspace volume deleted from the node, then the node is sanitised before reassignment. Filesystem deletion, not a cryptographic erase; export what you need before terminating, since it isn't reversible. |
Questions we get from marketing and creative agencies
Can we run a client's confidential brief through an LLM here?
On a dedicated GPUwerk instance, the machine is assigned to your agency alone, in EU-Central, and GPUwerk does not access, read, copy, index or analyse workload content except at your request for support or where a legal obligation requires it. Whether a given brief is cleared under your client contract to be processed on any external infrastructure, AI-specific or not, is a separate question for your agency's account and legal leads.
Does GPUwerk train on our client campaign material?
No. GPUwerk operates infrastructure, not models. Whatever you run on your instance is your software and your data, and under our data processing agreement GPUwerk does not access that content except at your request for support or where legally required.
Do client contracts that ban third-party AI tools cover this?
We can describe the infrastructure: dedicated single-tenant machine, EU-Central, an Article 28 DPA, no sub-processors on the workload, and no vendor model training on your content. Whether that satisfies a specific client contract's clause on AI tools or subcontractors is a determination for your legal team and your client, not something GPUwerk can certify.
Do you sign a DPA?
Yes, a standard GDPR Article 28 DPA is published at /legal/dpa at no charge, and there are no sub-processors for instance workloads.
Related pages
Put the campaign work on hardware your own team controls.
Talk to the people who run the racks, or start with a pilot and a practical rollout plan.
Talk to us Deploy an instance