The purpose of this AI Ethics Policy is to establish the ethical principles, governance requirements, and accountability mechanisms governing the design, development, deployment, operation, monitoring, and continuous improvement of Artificial Intelligence (AI) systems within the AverCare ecosystem.
This Policy ensures that AI technologies are developed and utilized in a manner that:
Protects patient safety
Promotes human wellbeing
Respects individual rights and dignity
Preserves human autonomy
Ensures fairness and non-discrimination
Protects privacy and confidentiality
Promotes transparency and trust
Supports clinical accountability
Complies with applicable legal and regulatory requirements
Aligns with AverCare's mission to improve health outcomes through responsible innovation
2. SCOPE
This Policy applies to:
Artificial Intelligence systems
Machine Learning systems
Large Language Models (LLMs)
Generative AI systems
Clinical Decision Support Systems
Predictive Analytics Systems
Health Coaching AI Systems
Symptom Assessment Tools
Risk Stratification Systems
Health Monitoring Algorithms
Wearable Device Algorithms
AI-powered Chat Interfaces
AI-generated Reports and Recommendations
Third-party AI technologies integrated into the AverCare Platform
This Policy applies to:
Employees
Contractors
Consultants
Medical Advisors
Developers
Data Scientists
Clinical Reviewers
Vendors
Third-Party Technology Partners
3. POLICY STATEMENT
AverCare is committed to developing and deploying AI systems that are ethical, safe, transparent, accountable, explainable, and human-centered.
AI shall be used to support human decision-making and shall not replace appropriately qualified healthcare professionals in making clinical decisions.
Patient safety, human oversight, and ethical responsibility shall always take precedence over automation, efficiency, or commercial objectives.
4. ETHICAL PRINCIPLES
4.1 Beneficence
AI systems shall be designed to promote positive health outcomes and improve the wellbeing of users.
AI systems shall:
Support preventive healthcare
Enhance patient engagement
Improve healthcare accessibility
Facilitate informed decision-making
Support healthcare professionals
The primary objective of AI shall be to create measurable benefit for patients and users.
4.2 Non-Maleficence
AI systems shall not intentionally cause harm.
AverCare shall implement controls to reduce risks associated with:
Incorrect recommendations
Hallucinations
Bias
Data inaccuracies
Automation failures
Security vulnerabilities
Misuse of AI outputs
AI systems shall undergo risk assessment before deployment.
4.3 Human Autonomy
Individuals retain control over decisions affecting their health and wellbeing.
AI systems shall:
Support human decision-making
Not manipulate users
Not coerce users
Not undermine informed consent
Respect individual choice
Users retain the right to seek human review and professional medical advice.
4.4 Human Oversight
AverCare adopts a Human-in-the-Loop (HITL) governance model.
Clinically significant AI outputs shall require human review before action is taken.
AI systems shall never function as autonomous medical decision-makers.
Human reviewers shall have authority to:
Approve outputs
Reject outputs
Modify outputs
Override outputs
Human decisions shall always take precedence over AI recommendations.
4.5 Fairness and Equity
AI systems shall be designed to promote equitable outcomes.
AverCare prohibits discrimination based on:
Race
Ethnicity
Nationality
Gender
Age
Disability
Religion
Socioeconomic status
Sexual orientation
Genetic characteristics
AI models shall be evaluated for bias and fairness throughout their lifecycle.
4.6 Transparency
Users shall be informed when AI is being utilized.
Transparency measures shall include:
AI disclosures
Explanation of AI involvement
Description of system limitations
User-facing disclaimers
Documentation of AI capabilities
Users shall be able to understand when content has been generated or influenced by AI.
4.7 Explainability
Where appropriate, AI outputs shall be explainable and interpretable.
AverCare shall strive to provide:
Rationale for recommendations
Supporting evidence
Confidence indicators
Decision pathways
Where full explainability is not technically feasible, appropriate risk mitigation measures shall be implemented.
4.8 Accountability
AverCare remains accountable for AI systems deployed within its ecosystem.
Responsibility for AI-generated outputs cannot be delegated to the AI system itself.
Accountability shall remain with:
AverCare Management
Clinical Reviewers
AI Governance Committee
Medical Advisory Board
Development Teams
Responsible Personnel
4.9 Privacy and Confidentiality
AI systems shall operate in accordance with privacy principles and applicable laws.
Personal information shall be protected through:
Data minimisation
Encryption
Access controls
Retention controls
Secure processing
Secure disposal
Special protections shall apply to health-related information.
4.10 Safety by Design
AI systems shall incorporate safety controls throughout the lifecycle.
Safety controls shall include:
Risk assessments
Human review processes
Escalation procedures
Override mechanisms
Monitoring controls
Validation testing
Patient safety shall be prioritized above performance metrics.
5. CLINICAL AI ETHICS
AI systems shall not independently:
Diagnose diseases
Prescribe medication
Modify treatments
Make emergency decisions
Replace clinical judgment
Clinical decisions shall remain the responsibility of appropriately qualified healthcare professionals. All clinically significant outputs shall be subject to Human-in-the-Loop review.
6. RESPONSIBLE DATA USE
AverCare shall use data responsibly.
Data practices shall include:
Lawful collection
Purpose limitation
Data minimisation
Accuracy controls
Retention management
Security safeguards
Data shall not be used in ways that violate user trust or applicable regulations.
7. AI MODEL DEVELOPMENT REQUIREMENTS
All AI systems shall undergo:
Ethical risk assessment
Clinical validation
Security assessment
Bias testing
Safety testing
Performance evaluation
Regulatory assessment
Prior to deployment.
8. BIAS MANAGEMENT
AverCare shall implement processes to:
Identify bias
Monitor bias
Measure bias
Reduce bias
Correct bias
Bias testing shall occur:
Before deployment
Following significant updates
Periodically during operation
9. AI MONITORING
AI systems shall be continuously monitored for:
Performance degradation
Model drift
Hallucinations
Unsafe outputs
Bias
Security concerns
User complaints
Monitoring results shall be reviewed by the AI Governance Committee.
10. USER RIGHTS
Users have the right to:
Know when AI is being used
Receive information about AI-assisted outputs
Request human review where applicable
Raise concerns regarding AI outputs
Exercise privacy rights
Submit complaints regarding AI systems
11. PROHIBITED USES OF AI
AI shall not be used for:
Autonomous diagnosis
Autonomous prescribing
Autonomous treatment decisions
Manipulative behaviour
Deceptive practices
Unlawful surveillance
Discriminatory profiling
Unauthorized data collection
Activities prohibited by law
12. REGULATORY COMPLIANCE
This Policy supports compliance with:
World Health Organization (WHO) Guidance on Ethics and Governance of Artificial Intelligence for Health
FDA Artificial Intelligence and Machine Learning Guidance
IMDRF Software as a Medical Device Framework
European Union AI Act
GDPR
UK GDPR
HIPAA
HITECH Act
UAE PDPL
Privacy Act 1988 (Australia)
Privacy Act 2020 (New Zealand)
Digital Personal Data Protection Act (India)
ISO 42001
ISO 27001
ISO 14971
ISO 13485
13. AI GOVERNANCE COMMITTEE
The AI Governance Committee shall:
Review ethical risks
Review AI incidents
Review model performance
Review bias assessments
Review safety concerns
Approve high-risk AI deployments
Monitor policy compliance
The Committee shall meet at least quarterly.
14. ROLES AND RESPONSIBILITIES
14.1 Chief Executive Officer (CEO)
Responsible for:
Executive accountability
Strategic oversight
Resource allocation
14.2 Chief Medical Officer (CMO)
Responsible for:
Clinical governance
Clinical validation
Patient safety oversight
14.3 AI Governance Committee
Responsible for:
Ethical oversight
Risk management
Compliance monitoring
14.4 Development Teams
Responsible for:
Ethical design
Secure development
Testing and validation
14.5 Clinical Reviewers
Responsible for:
Reviewing AI outputs
Exercising human judgment
Escalating concerns
15. POLICY VIOLATIONS
Failure to comply with this Policy may result in:
Corrective action
Removal of system access
Disciplinary action
Contract termination
Regulatory reporting where required
16. REVIEW AND MAINTENANCE
This Policy shall be reviewed:
Annually
Following significant AI system changes
Following material incidents
Following regulatory changes
17. VERSION CONTROL
09 June 2026 Date: ______________________________ 09 June 2026
This is the published version of AC-POL-AIE-135, version 1.0. The signed original is retained on record and is available to customers, auditors and regulators on request via ai-governance@avercare.global.
