Skip to content
Trust

Security & data handling

How we protect your data, keep you in control of your accounts, and keep people in charge of the decisions that matter.

Our commitments

Your data, your accounts, your decisions

You own your accounts and data

Whenever practical, AI provider accounts, software licenses, cloud environments, and strategic data stay in your name. You can revoke our access at any time.

Not used to train AI models

We select and configure AI services so your content is excluded from provider model training, and we document the settings and terms we rely on.

Permission-aware by design

AI only sees what the person using it is allowed to see. Access follows your existing roles, groups, and approval rules.

People approve important actions

High-impact processes start in read-only, draft, or human-approval mode. Consequential actions always have an approval and escalation path.

Documented and auditable

Every engagement documents data flows, connected systems, model and provider choices, prompts, and the controls in place.

Least access, for the least time

We request only the access a task needs, prefer service accounts you control, and remove our access when the work ends.

In practice

How we work with your systems

Before we start
  • We map the data involved, who may see it, and any client, contractual, or regulatory restrictions.
  • We agree which systems AI may read from and which actions, if any, it may take.
  • We confirm data residency and hosting requirements, including private or client-hosted options.
While we build
  • We use fictional or authorized sample data for demonstrations and testing.
  • Credentials live in your systems or a password manager, never in documents or chat.
  • Testing covers accuracy, permissions, and failure cases before anything reaches users.
After go-live
  • Monitoring, access reviews, and cost reviews are part of Managed AI Operations.
  • Incidents are routed to named owners with an agreed response path.
  • Wider rollout happens only after acceptance evidence supports it.
Straight answers

What we do not claim

VEOTEQ does not currently hold third-party security certifications, and we do not promise AI accuracy or autonomy beyond what has been tested and approved in your environment. We work within your security policies, complete your vendor security questionnaires, and sign data processing and confidentiality agreements when required.

Questions from your security or IT team? Email ssallam@veoteq.com. See also our privacy policy.

Start with an AI Value Sprint

A focused discovery engagement that identifies your highest-value AI opportunity, the controls it needs, and a practical pilot.