Last Updated: January 1, 2026

Data Use Policy

Overview

This Data Use Policy governs SoDNAscan's use of artificial intelligence to process your data. AI-powered features require your explicit, separate consent. If you do not opt in to AI processing, it will not be applied to your data in any way. You may use your SoDNAscan account to store and view your uploaded data without enabling AI features.

This policy should be read together with our Privacy Policy and Terms of Service.

1. AI Provider

SoDNAscan uses Anthropic's Claude API to power AI analysis features. Anthropic is a US-based AI safety company. The specific model used is from the Claude model family.

• Provider: Anthropic, PBC
• Product: Claude API (commercial tier)
• Infrastructure: Anthropic operates on Amazon Web Services (AWS) and Google Cloud Platform (GCP) infrastructure
• Location: United States
• Anthropic Privacy Policy: https://www.anthropic.com/privacy

2. What Data Is Sent to the AI

When you enable AI features and initiate Health Book generation, the following data is transmitted to Anthropic's Claude API:

Personal and demographic data:

• Name, age, sex, height, weight, ethnicity

Health profile:

• Current health status (free-text)
• Health history (free-text)
• Family health history (free-text)
• Current supplements (free-text)
• Health goals (free-text)
• Lifestyle notes and preferences (free-text)

Genetic data:

• Matched genetic variant data, including rsid identifiers, chromosome, position, alleles, associated gene, and biological system category

Blood work data:

• Confirmed biomarker values, including marker name, value, unit, reference range, and status flag
• Lab name and test date

Wearable health metrics:

• Aggregated health statistics including heart rate, HRV, SpO2, respiratory rate, sleep stages and duration, exercise metrics, body composition, and recovery scores

This data is transmitted via encrypted HTTPS connections.

3. What the AI Does

The AI performs two functions:

3.1 Blood Work Extraction

When you upload a blood work PDF or paste lab results as text, the AI extracts structured biomarker data (marker names, values, units, reference ranges) from unstructured text. You review and confirm the extracted values before they are used in your Health Book.

3.2 Health Book Generation

The AI generates 10 analytical reports covering different biological systems, which are compiled into your personalized Health Book. Each report analyzes your genetic variants, blood biomarkers, and wearable metrics in the context of your demographic and health profile.

The AI provides wellness insights and educational information — not diagnoses, treatment recommendations, or clinical findings.

4. Data Isolation

• Per-user processing: Each API call contains only your data. Your data is never combined with, compared against, or enriched by other users' data.
• No cross-user learning: The AI does not learn from one user's data to inform another user's results. Each Health Book is generated independently.
• Session isolation: Each API call is a stateless request. The AI does not retain memory of previous calls or users between requests.

5. No Model Training

Anthropic does not use API data to train AI models. This is guaranteed in Anthropic's commercial API terms.

Specifically:

• Your genetic data, health information, and AI outputs are not used to train, fine-tune, or improve Anthropic's AI models
• Your data is not incorporated into training datasets
• Your data is not used for model evaluation or benchmarking

6. Data Retention by Anthropic

• Default retention: Anthropic retains API inputs (your data) and outputs (AI-generated analysis) for 7 days for trust and safety purposes, then deletes them
• Safety classifiers: Anthropic's automated safety systems may process API data to detect policy violations. Classifier results (not your raw data) may be retained separately
• Zero Data Retention option: Anthropic offers a Zero Data Retention (ZDR) tier under enterprise contracts that eliminates all input/output storage except safety classifier results. SoDNAscan currently operates on the standard commercial API tier with 7-day retention. If we upgrade to ZDR, we will update this policy accordingly
• No indefinite retention: Unlike some AI providers, Anthropic's commercial terms specify a fixed 7-day retention window, not indefinite or unspecified retention

7. Consent and Opt-In

7.1 Explicit Consent Required

AI features require your explicit opt-in consent before any data is sent to Anthropic. This consent is:

• Separate from your agreement to the Terms of Service — you can agree to the Terms without consenting to AI processing
• Informed — this policy describes exactly what data is sent, to whom, and for what purpose
• Recorded — we store the timestamp and policy version of your consent

7.2 Granular Consent

You consent to the following AI processing purposes, each independently:

• AI analysis of genetic data — your matched genetic variants are sent to the AI for analysis
• AI analysis of blood work — your confirmed biomarker values are sent to the AI for analysis
• AI analysis of wearable data — your parsed wearable health metrics are sent to the AI for analysis
• AI extraction of blood work data — your uploaded blood work PDF or pasted text is sent to the AI for structured data extraction

7.3 Withdrawal

You may withdraw consent for AI processing at any time. Withdrawal:

• Takes effect immediately for future processing
• Does not affect the lawfulness of processing that occurred before withdrawal
• Does not delete your account or uploaded data — you retain access to your data without AI features
• Does not affect previously generated Health Books that have already been delivered

To withdraw consent, contact info@sodnascan.com or use the consent management settings in your account.

8. Limitations of AI

You should understand the following limitations:

• AI may produce inaccurate information. Large language models can generate plausible-sounding but incorrect statements. Always verify health information with qualified professionals.
• AI does not have access to your complete medical context. It only knows what you upload and enter — it does not have access to your full medical records, imaging, or clinical assessments.
• Genetic science is evolving. Many gene-health associations are based on population studies that may not apply to all individuals. New research may contradict current understanding.
• AI outputs reflect the training data of the AI model, not a comprehensive review of all medical literature.
• Confidence levels are informational, not clinical. When the AI indicates confidence in an association, this reflects pattern strength in available data, not clinical certainty.
• AI cannot account for gene-gene interactions, epigenetic factors, or environmental influences that may significantly affect health outcomes.

9. Not Medical Advice

AI-generated content in your Health Book is wellness and educational information only. It is not:

• Medical advice
• A diagnosis or indication of any disease or condition
• A recommendation for treatment, medication, or lifestyle change
• A substitute for professional medical judgment

Always consult a qualified healthcare provider before making health decisions. If you have concerns about a health condition, seek professional medical attention regardless of what your Health Book contains.

10. Human Oversight

SoDNAscan does not employ physicians or genetic counselors to review individual Health Book outputs before delivery. AI-generated content is delivered directly to you.

You are the human in the loop. It is your responsibility to:

• Evaluate AI outputs critically
• Consult healthcare professionals for medical interpretation
• Not act on AI-generated content without professional guidance when health decisions are involved

11. Automated Decision-Making (ADMT)

Under California's ADMT regulations (effective January 2026) and GDPR Article 22:

• SoDNAscan's AI processing constitutes automated decision-making technology that processes sensitive personal information
• You have the right to opt out of AI-powered processing while retaining access to the non-AI features of the Service
• AI outputs do not produce decisions with legal or similarly significant effects — they produce informational wellness content
• You may request information about the logic, significance, and envisaged consequences of the automated processing by contacting info@sodnascan.com

12. EU AI Act Transparency

In accordance with EU AI Act Article 13 transparency requirements (applicable August 2, 2026):

• System description: SoDNAscan uses a large language model (Anthropic Claude) to analyze genetic, blood work, and wearable health data and generate personalized wellness reports
• Capabilities: The system can identify genetic variant associations, interpret blood biomarker values in context, analyze wearable health trends, and synthesize findings into narrative health content
• Limitations: The system cannot diagnose diseases, replace clinical judgment, account for gene-gene interactions or epigenetic factors, guarantee accuracy of any individual finding, or incorporate information not explicitly provided by the user
• Human oversight: Individual outputs are not reviewed by medical professionals before delivery. The user serves as the point of human oversight
• Risk management: SoDNAscan maintains internal documentation of risk assessment and mitigation measures related to AI processing of health data

13. International Data Transfers

When AI features are enabled, your data is transferred from SoDNAscan to Anthropic's servers in the United States.

For EU/EEA/UK/Swiss users, this transfer is governed by:

• Standard Contractual Clauses (SCCs) included in Anthropic's Data Processing Addendum
• The DPA is automatically incorporated into Anthropic's commercial terms
• Supplementary measures include encryption in transit (TLS 1.2+) and the 7-day retention limit

See our Privacy Policy, Section 7, for full details on international data transfers.

14. Changes to This Policy

We may update this Data Use Policy to reflect changes in our AI processing practices, Anthropic's terms, or regulatory requirements. When we make material changes:

• We will update the "Last Updated" date
• We will notify you by email
• Material changes to data processing may require renewed consent

15. Contact

For questions about AI processing of your data:

• Email: info@sodnascan.com