AI Feature Disclaimer

Idicine may use automated, algorithmic, machine learning, or AI-assisted features to support parts of the Service.

These features may help organize information, process uploads, create match signals, summarize user-generated content, detect safety issues, or improve the user experience.

This AI Feature Disclaimer explains what these features are and what they are not.

1. AI Features Are Not Medical Advice

AI or automated features on Idicine do not provide medical advice, diagnosis, treatment, prescriptions, dosing instructions, clinical recommendations, genetic counseling, or emergency support.

AI outputs are for informational, organizational, peer-support, and product functionality purposes only.

Do not rely on AI outputs to make medical decisions. Always talk to a qualified clinician before changing medication, dosage, supplements, treatment, or health routines.

2. AI May Be Used For Product Features

Idicine may use AI or automated systems to: extract text from uploaded files; identify possible match signals; convert information into privacy-safe categories; summarize Bio Chat discussions; suggest conversation titles or prompts; detect unsafe or policy-violating content; rank or organize Bio Chats; support search; assist customer support; or improve the Service.

These systems are tools. They are not clinicians.

3. AI Can Be Wrong

AI-generated or automated outputs may be: inaccurate; incomplete; outdated; misleading; biased; overconfident; irrelevant; unsafe; or not applicable to your situation. AI may misunderstand uploads, misread text, miss context, misclassify information, or generate information that sounds plausible but is incorrect.

You are responsible for reviewing AI outputs carefully and seeking professional guidance where appropriate.

4. No Clinical Validation

Unless we expressly state otherwise, AI features on Idicine are not clinically validated medical devices. They are not intended to diagnose, treat, cure, prevent, or mitigate disease. They are not intended to replace professional judgment. They are not intended to determine whether a medication, supplement, product, dosage, test, or treatment is safe or appropriate for you.

5. Upload Processing

If you upload bloodwork, DNA files, biomarker reports, lab documents, or other files, AI or automated systems may help process those uploads. This may include: extracting text; identifying possible biomarkers; identifying possible genetic or biological signals; creating privacy-safe match signals; summarizing information; or preparing data for Bio Chat matching.

Upload processing may be inaccurate. You should not treat extracted or summarized information as a verified medical record.

When you upload a file for analysis, the file is processed by AI software running on Idicine-controlled servers. No file content is transmitted to external AI providers. We delete the raw file from our servers after processing is complete.

6. Bio Matching

AI or automated systems may help calculate or support bio match scores, match signals, and Bio Chat visibility. A bio match score is a product feature used to organize peer-support conversations. A bio match score is not a medical, clinical, diagnostic, genetic, safety, treatment, or scientific conclusion.

A high bio match does not mean: you have the same condition as another user; another user's experience applies to you; a product is safe or effective for you; a medication or supplement is appropriate for you; or you should make any health decision.

7. Bio Chat Summaries

Idicine may in the future use AI to summarize Bio Chats or user experiences. If and when this feature is introduced, summaries may omit important context, misrepresent user experiences, or combine information in a way that does not apply to you. A summary is not a recommendation. A summary is not proof that a supplement, medication, test, routine, or product works. A summary is not a substitute for reading the original conversation or talking to a qualified clinician.

For user-generated content displayed or organized through the Service, Idicine relies on the protections of Section 230 of the Communications Decency Act (47 U.S.C. § 230) and corresponding state and federal law. AI-generated summaries, however, may constitute content created or materially shaped by Idicine and therefore may not be covered by those protections in the same way. Before any AI summary feature is enabled in production, we will (i) add prominent in-feature disclaimers warning users that AI summaries are not medical advice and should not be relied on for medical decisions, and (ii) implement reasonable safeguards designed to reduce the risk of summarizing dangerous instructions or harmful content. Even with those safeguards, AI-generated summaries are not medical advice, and you should never act on them without consulting a qualified clinician.

8. Safety and Moderation

Idicine may use AI or automated tools to detect content that may violate our Terms of Use or Community Guidelines. These systems may make mistakes. They may fail to detect harmful content or may incorrectly flag harmless content.

9. No Emergency Use

AI features on Idicine are not designed for emergencies. Do not use Idicine AI features to evaluate urgent symptoms, overdose, allergic reactions, suicidal thoughts, severe pain, breathing problems, chest pain, medication emergencies, or any other urgent condition.

Call 911 or seek immediate medical care if you may have a medical emergency. If you are in emotional distress or thinking about self-harm, call or text 988 in the United States or seek immediate professional help.

10. Product Mentions and Affiliate Links

AI may help organize or display product mentions, user experiences, or affiliate-linked content. This does not mean Idicine recommends, endorses, prescribes, or medically approves any product. Product mentions are not recommendations for your genes, blood type, biomarkers, symptoms, diagnosis, or health situation.

11. Your Responsibility

You are responsible for how you use AI outputs. You agree not to rely on AI outputs as medical advice. You agree not to use AI outputs to make medication, supplement, dosage, treatment, or emergency decisions. You agree to consult qualified professionals for health decisions.

12. Changes to AI Features

Idicine may change, limit, suspend, remove, or discontinue AI features at any time. We may update models, prompts, providers, workflows, or output formats. AI behavior may change over time.

13. Data Use

AI features may process information you provide, including uploads or user-generated content, as described in our Privacy Policy and our Genetic, Biomarker and Bio-Matching Consent. AI processing runs on Idicine-controlled infrastructure.

14. Backend Pattern Discovery (Internal Research)

Idicine operates an internal research program — the Bio-Outcome Intelligence Engine — that may use automated, statistical, and AI-assisted techniques to identify population-level associations between bio-signatures (combinations of blood type, biomarker, and genetic features) and self-reported outcomes mentioned in Bio Chats. This program is strictly opt-in and is controlled by two granular toggles in Settings: research participation and research partner sharing.

The outputs of this program — including any predictions, scores, associations, hypotheses, or interpretations — are not returned to individual users through the Service. The Service does not present personalized predictions, health forecasts, risk scores, treatment suggestions, or clinical interpretations to you. This design is deliberate and serves two purposes: (i) to keep the program outside the regulatory definition of clinical decision support and software-as-a-medical-device, and (ii) to keep medical decisions in the hands of users and qualified clinicians.

The only product-facing effect of the program is improved internal ranking quality for Bio Chat surfacing, which is a peer-support feature and is not medical advice. All compute occurs on Idicine-controlled infrastructure. No data covered by research consent is processed by any third-party AI provider. Aggregate outputs shared with research partners pass through an internal pseudonymization and de-identification pipeline (HIPAA Safe Harbor methodology, informed by the principles of 45 C.F.R. § 164.514(b) Expert Determination, with k-anonymity ≥ 10) before any external transfer. Operational details are described in our Privacy Policy §32 and our Genetic, Biomarker and Bio-Matching Consent §27. The contractual obligations applicable to research partners are described in the Data Processing Agreement.

Even where AI or statistical methods produce associations that appear strong, those associations are hypotheses, not clinical conclusions. They may be biased, confounded, or inapplicable to any individual. Idicine does not represent that any internal model output is a medically validated finding, and you should not interpret any aspect of the Service as such.

15. No Warranty

AI features are provided “as is” and “as available.” Idicine disclaims all warranties regarding AI outputs, including accuracy, completeness, safety, reliability, usefulness, fitness for a particular purpose, and non-infringement. You use AI features at your own risk.

To the maximum extent permitted by law, Idicine is not liable for any harm, health outcome, medical decision, injury, or other consequence arising from your use of or reliance on AI-generated or automated outputs from the Service.

16. Contact

If you have questions about AI features, contact us at:

Idicine
5214F Diamond Hts Blvd
San Francisco, CA 94131
support@idicine.com
privacy@idicine.com