Trust

Responsible AI & Security

Our approach to responsible and secure technology development. This page describes how we work; it does not represent a formal certification.

Responsible AI

We consider appropriate human oversight, transparency, security and data-protection requirements when designing and developing AI systems.

AI systems are designed so that people remain able to review, question and override outputs where decisions matter.

We aim to be clear about what an AI system does, the data it relies on and its known limitations.

AI components are subject to the same secure-development practices as the rest of the software.

Personal data is used only where necessary and handled in line with UK GDPR considerations.

Models are evaluated against agreed criteria before use and reviewed as data and requirements change.

We consider whether AI is the right tool at all, and prefer simpler approaches where they meet the need.

Security is considered from design onwards, including code review, dependency management and testing.

Access to systems and data follows least-privilege principles, with strong authentication.

We collect and retain only the data needed for a defined purpose.

Data is encrypted in transit and, where appropriate, at rest.

Cloud services are configured with security in mind, using reputable providers and managed infrastructure.

Backup and recovery arrangements are planned to suit the needs of each system.

Security incidents are identified, contained, investigated and communicated promptly.

Data-protection requirements are considered throughout design, development and operation.

Scheme	Status	Notes
Cyber Essentials	Planned	This page will be updated when certification is obtained.

For security or data-protection questions, or to report a concern, contact info@anvicgroup.com.