We assess how effectively an organisation is identifying and managing ethical AI risk across the lifecycle of its AI use. The Health Check looks at areas including bias, transparency and privacy, and provides a clear picture of strengths, gaps and priorities for action.

We test governance arrangements against real or emerging AI use cases to see how they would work in practice. The Stress-Test looks at potential harms, safeguards, human oversight, escalation, accountability and who has the authority to pause or stop a use case when a boundary is crossed.

We help organisations assess which frontier AI models are most fit for a particular organisation, use case and risk profile. The assessment looks beyond raw capability to consider reliability, safety, privacy, transparency, governance and operational fit, using independent evidence rather than provider claims alone.

Where relevant, the assessment can also include a Te Ao Māori fitness lens to test cultural competence, tikanga, data sovereignty and governance fit. description