The AI listens. It does not replace a therapist.

What the AI is intended to do

AI components may support conversation, reflect what a user has shared, organise selected emotional context, notice recurring themes, adapt interface stimulation and help prepare structured summaries.

These capabilities are intended to support continuity and care routing. They do not make the AI a therapist or clinician.

What the AI is not intended to do

Chiedza’s AI is not intended to diagnose mental-health or medical conditions, prescribe treatment, provide autonomous therapy, replace professional judgment or act as an emergency service.

AI output can be incomplete, inappropriate or wrong. Important care decisions should not rely on AI output alone.

Human responsibility

Qualified humans retain authority for diagnosis, clinical interpretation, treatment, care planning and escalation decisions.

Where a human-care relationship exists, professionals should be able to review, correct or reject AI-organised context.

Disclosure and user control

AI-supported interactions should be clearly identified. People should be able to understand when AI is involved and control relevant sharing, transcript and therapist-visibility settings.

Consent should be specific, understandable and changeable rather than hidden inside a single permanent choice.

Risk and crisis boundaries

The intended product direction includes defined pathways for human review and external support when risk increases. Final workflows will depend on geography, partnerships, clinical advice and operating capacity.

Chiedza is not currently an emergency service. Anyone in immediate danger should contact local emergency services or an appropriate crisis-support provider.

Development and review

Chiedza is now live. AI behaviour, safeguarding architecture, data practices and human-care pathways remain subject to testing and expert review as the live app develops. Organisational and clinical care pathways require separate validation.