How we use AI
AI is central to our products. Here's what it does across MailOps, CXO Assist, and ArtistPro:
Classifying and routing information
Drafting content
Extracting data from documents
Answering questions from your knowledge
Automating repetitive tasks
The models we use
We use a mix of frontier and open-weight models, selected per task for quality, cost, and latency. We do not lock into a single provider.
How we handle your data with AI
Your data is yours
Inputs and outputs belong to you. We don't claim ownership of anything you send through the system.
No training without consent
We do not use your data to train or fine-tune models unless you explicitly opt in. Our upstream providers are bound by the same commitment.
Isolation
Your data is isolated from other customers. Prompts and responses are not shared across accounts.
Grounded answers
Where possible, AI features are grounded in your data (RAG) rather than relying solely on the model's training data.
Human in the loop
AI assists — it doesn't decide. Critical actions (sending emails, approving content, financial operations) always require human confirmation before execution.
Accuracy & limitations
AI models are powerful but imperfect. They can generate incorrect, incomplete, or outdated information. We design our products to minimise this risk (grounding, guardrails, confidence scores), but no system is perfect.
You should review AI-generated outputs before relying on them for significant decisions — especially in financial or regulated contexts.
Transparency
Where you interact directly with an AI feature, we make that clear.
Where we publish AI-generated content, we label it appropriately.
We follow the transparency principles of the EU AI Act and align our AI governance with ISO/IEC 42001 practices.
Responsible AI
We build with these principles in mind:
Secure
Data stays protected at every stage.
Accountable
We own outcomes. AI augments, humans decide.
Fair
We test for bias and work to mitigate it.
Correctable
If AI gets it wrong, users can flag and fix.
Contact
Questions about our AI practices?
Email contact@zyqo.ai
Or our data contact at [DATA CONTACT EMAIL]