Principle 01
Least-necessary access
Users should see only the applications, business objects, data and functions required for their authorised role.
Access is governed at more than one layer.
VibrantAI is designed to help organisations use analytical and AI capabilities while maintaining control over who can access information, how business context is applied, where customer boundaries are enforced and how activity can be reviewed.
Trust is established through role-sensitive access, data protection, customer-specific boundaries, AI controls, privacy practices, traceability and clear shared responsibility. Formal certifications and independent attestations are published when achieved, documented and applicable to the relevant service scope.
The trust model covers the complete path from user access and source evidence through analysis, AI interaction, review, collaboration and accountable management action.
Users should see only the applications, business objects, data and functions required for their authorised role.
Business definitions, functional areas, roles, source relationships and AI-enabled experiences are configured for the customer environment.
Natural-language access and synthesis should honour the same underlying data and business-object boundaries as other platform experiences.
Controls, compliance status, independent assessments and certifications are communicated with clear scope and supporting documentation.
Role-based access is most effective when the application, business object and data layers work together rather than relying on a single page-level permission.
Select a control area to see the protection model, the supporting control evidence and how the capability is configured for each customer environment.
The access model should combine application permissions, business-object visibility and granular data restrictions.
VibrantAI applies customer-specific boundaries across application services, data stores, storage, integration credentials, caches, logs and administrative access. The selected isolation model is aligned to the customer’s deployment, security and regulatory requirements.
EnvisionAI and AskAI combine natural-language interaction, structured analysis and retrieval-augmented generation. The analytical work is designed to execute within the customer-controlled environment. The trust model constrains what is retrieved, protects any limited context used with an AI provider and governs how the resulting answer is restored, presented and traced.
Privacy is implemented through product configuration, deployment controls, operating procedures and customer instructions. Applicable legal, contractual, geographic and data-category requirements are mapped to the individual customer environment.
Identify the business and analytical purpose for processing.
Use only the data, fields and documents required.
Apply roles, data boundaries and functional context.
Use documented source, analytical and AI boundaries.
Apply agreed periods to data, results, history and logs.
Support documented offboarding and deletion obligations.
Maintain records needed to explain processing and action.
Auditability is especially important because VibrantAI connects analytical evidence and AI-assisted interpretation to executive narratives, decisions and actions.
Authentication, authorised access, administrative actions and relevant security events according to the implemented logging design.
Changes to measures, roles, hierarchies, Storyboards, Commitments, Value Streams and other governed definitions.
The data source, document context, analytical method and evidence age supporting a report, narrative or outcome reading.
Why intervention was required, what was decided, who owned the action and what result followed.
Security is not transferred entirely to the provider or entirely to the customer. Responsibilities depend on the deployment, source architecture, contractual scope and enabled platform capabilities.
Provide and maintain the agreed SaaS, analytical, integration and AI controls within the defined service boundary.
Configure the platform and connected systems according to the organisation’s access, privacy, risk and records-management requirements.
VibrantAI distinguishes platform capability, customer-specific configuration and formal independent assurance so buyers can understand the scope and status of each trust commitment.
High-level capabilities such as role-based access, granular visibility, secure dedicated environments, RAG and protected narrative handling.
Authentication integration, retention periods, AI providers, source access, environment topology, logging scope and administrative separation.
Independent assessments, audit reports, certifications and regulatory positions are communicated with their applicable service scope, date and supporting documentation.
The public page introduces the trust model. Enterprise evaluation should use controlled documentation that reflects the actual service and deployment being proposed.
Service boundaries, customer data paths, integrations, storage, AI providers, environments and administrative access.
Implemented controls, customer configuration, shared responsibilities, exceptions and evidence ownership.
Processing purpose, data categories, subprocessors, retention, deletion, geographic considerations and customer instructions.
Enabled providers, retrieval sources, prompt and context boundaries, provider retention settings and human-approval model.
Access administration, monitoring, incident handling, change management, backup, recovery and support practices.
Current independent assessments, certifications and audit materials with their applicable scope and validity period.
VibrantAI brings role-sensitive access, customer-controlled context, protected evidence, accountable AI and traceable management activity into one enterprise trust model.