Your data is your own.
Where does it go, and who can act on it? The questions IT and procurement ask us first - answered plainly.
Where is our data processed?
In the UK or EU by default. Where a workflow requires a model hosted elsewhere, we say so in writing before deployment and you decide whether to proceed.
Do you train models on our data?
No. Client data is used to serve that client only. We contract with providers on terms that exclude training on submitted data, and we can show you those terms.
Who can the system act on behalf of?
Nothing that touches a customer, a payment or a record of account is sent without a named person approving it, unless you explicitly ask for an unattended step and accept that in writing.
What happens if the model gets it wrong?
Every workflow logs its inputs, outputs and decisions so an error can be traced and reversed. Pilots run alongside the existing process precisely so failures surface before anyone depends on the output.
What if we end the engagement?
You keep the workflow, the documentation and the accounts it runs on. We do not hold your integrations hostage, and there is no proprietary layer you have to keep paying us for.
Which other companies touch our data?
The third-party services we use to run your workflow - for example the telephony provider and the model provider. We keep a current, named list of them for every client, with where each one processes data, and we work to UK GDPR throughout. Our data-processing terms are shared on request.
The ICO publishes guidance on applying UK GDPR principles to AI systems; we work to it. ICO AI and data-protection guidance →
Ask us the hard questions first.
Bring your IT lead or your DPO to the demo call - we'd rather answer this before you commit than after.