AI Policy

1. Introduction

MySafeTherapy uses artificial intelligence (AI) to support reflection, wellbeing, and self-help alongside access to qualified human therapists. This policy explains how AI is used, its limitations, and your rights.

We are committed to ethical, transparent, and responsible AI use in line with:

• UK GDPR and ICO guidance on AI and data protection
• The Data Protection Act 2018
• Professional mental health standards (including BACP principles)
• Emerging international AI frameworks, including the EU AI Act

2. How We Use AI

AI is used to enhance the user experience through optional features, including:

2.1 Weekly Insights

AI analyses mood entries and journal content to identify general patterns and trends. These insights are intended to support personal reflection and optional discussion with a therapist.

2.2 Navi – AI Chat Assistant

Navi provides supportive conversation, wellbeing information, and coping suggestions.

Navi:

• Is not a therapist
• Does not provide diagnosis, treatment, or clinical advice
• Does not replace human judgement or professional care

2.3 Journaling Prompts

AI may suggest personalised journaling prompts based on recent themes to encourage reflection and emotional awareness.

2.4 Progress Recommendations

AI may suggest worksheets, programmes, or resources based on your activity and stated goals. These suggestions are optional and non-directive.

3. Important Limitations

You should always rely on qualified professionals for clinical care. If you are in crisis, seek immediate help from emergency services or crisis helplines.

4. Data Processing

4.1 Data Used by AI

With your consent, AI may process:

• Mood entries and ratings
• Journal content
• Worksheet responses
• Interactions with the Navi assistant

AI processing is limited to what is necessary to provide the relevant feature.

4.2 How Data Is Processed

• Data is encrypted in transit and at rest
• Processing occurs on secure cloud infrastructure
• We use third-party AI providers (such as Google and OpenAI) under Data Processing Agreements
• Personal identifiers are minimised or pseudonymised where feasible

We do not use identifiable therapy content to train general-purpose AI models without explicit consent.

4.3 Data Retention

AI-generated insights form part of your account data. You can:

• Delete AI-related data via your account settings
• Withdraw consent at any time
• Request full data erasure, subject to legal and safeguarding obligations

5. Human Oversight & Control

AI features operate with meaningful human oversight:

• AI outputs are not automatically shared with therapists
• Therapists do not rely on AI for clinical judgement
• Users control what data is shared and with whom
• AI features can be disabled at any time
• AI outputs are reviewed internally for safety and appropriateness

No significant decisions about your care are made solely by AI.

6. Automated Decision-Making & Profiling

MySafeTherapy does not use AI to make automated decisions that produce legal or similarly significant effects under Article 22 of UK GDPR.

AI tools:

• Support reflection and self-awareness
• Do not determine access to therapy
• Do not replace human review or decision-making

You may request human review of any AI-related concern.

7. Accuracy, Bias & Fairness

We take steps to reduce risk, but acknowledge that:

• AI systems may reflect biases in training data
• Outputs may not fully reflect cultural, personal, or contextual nuance
• Errors or inappropriate content may occasionally occur

We mitigate these risks through:

• Regular review and testing
• Content filtering and safeguards
• User feedback mechanisms

If you encounter concerning AI output, please report it to: ai-feedback@mysafetherapy.com

8. Crisis-Related Safeguards

AI tools are not monitored in real time and do not provide crisis response.

Where AI detects language that may indicate distress:

• The system may display supportive messaging or crisis resources
• This does not constitute monitoring or emergency intervention
• Responsibility to seek help remains with the user

9. Your Rights

Under UK GDPR, you have rights relating to AI-assisted processing, including:

• Right to be informed about AI use
• Right to access relevant data
• Right to object to AI-based processing
• Right to request human review
• Right to withdraw consent
• Right to erasure, subject to legal obligations

To exercise these rights, contact: privacy@mysafetherapy.com

10. Opting Out

You can opt out of AI features at any time via your account settings, including:

• Weekly AI insights
• AI journaling prompts
• Navi chat assistant

Opting out does not affect access to therapy or core platform functionality.

11. Third-Party AI Providers

We work with carefully selected AI providers, including:

• Google AI – for analytical insights
• OpenAI – for conversational AI features

All providers:

• Are bound by contractual data protection obligations
• Do not use your data to train their general AI models
• Process data only on our documented instructions

12. Safety & Testing

We apply safeguards to AI features, including:

• Clear labelling of AI-generated content
• Output filtering
• Regular internal reviews
• Feature-level risk assessments

13. Regulatory Compliance

Our AI practices align with:

• UK GDPR and ICO guidance on AI
• The Data Protection Act 2018
• Principles of the EU AI Act (where applicable)

We review and update our AI practices as regulations evolve.

14. Changes to This Policy

We may update this policy as AI features develop. Material changes will be communicated via the platform or email.

15. Contact Us

AI-related questions or concerns:
📧 ai-feedback@mysafetherapy.com

Data protection queries:
📧 privacy@mysafetherapy.com