Medical Disclaimer
The honest limits of AI health guidance — what our automated features can and cannot do, when to stop reading and call someone, and where responsibility sits.
Effective date: 7 August 2026 · Last updated: 7 August 2026
1. The Services do not provide medical advice
All content produced by OnlineCare AI's automated features — voice conversations, text chat, symptom analysis, mental-health screening, health insights and medication information — is provided for general educational and informational purposes only.
It is not medical advice. It is not a diagnosis. It is not a treatment plan or a prescription. It is not a substitute for examination, testing, judgement or care by a qualified healthcare professional who knows your history.
2. No doctor–patient relationship with OnlineCare AI
Using the AI features does not create a doctor–patient, therapist–client or any other healthcare provider relationship between you and OnlineCare AI. We are a technology company. We do not practise medicine.
Where you book a consultation, the clinical relationship is between you and that independently licensed clinician, governed by their professional obligations and the law of the state in which they are licensed.
3. Emergencies and crisis situations
Do not use OnlineCare AI in an emergency. The Services do not monitor your input in real time, cannot summon help and cannot guarantee any response time.
- Medical emergency in the US — call 911 or go to the nearest emergency department.
- Suicidal thoughts, self-harm or a mental-health crisis in the US — call or text 988, the Suicide & Crisis Lifeline. It is free, confidential and available 24 hours a day. You can also chat at 988lifeline.org.
- Domestic violence in the US — the National Domestic Violence Hotline is available on 1-800-799-7233.
- Outside the US — contact your local emergency number.
Warning signs that mean stop and seek help now include chest pain or pressure, difficulty breathing, severe or uncontrolled bleeding, sudden weakness or numbness on one side, sudden confusion or difficulty speaking, a seizure, a stiff neck with severe headache, a rash that does not fade under pressure, or any thought of harming yourself or someone else.
4. What AI cannot do
Our AI features are built carefully and reviewed against curated clinical reference material, but they have real and permanent limits:
- They cannot examine you, listen to your chest, feel your abdomen or look in your throat.
- They cannot order or interpret tests, imaging or bloodwork.
- They cannot observe you over time or notice that you look unwell.
- They depend entirely on what you choose to tell them. Incomplete or inaccurate input produces incomplete or inaccurate guidance.
- They can produce output that is confident and wrong, outdated, or inapplicable to your specific circumstances.
- They do not know about drug recalls, local outbreaks or guideline changes published after the underlying reference material was last updated.
5. A screening score is not a diagnosis
The mental-health tools use questionnaires adapted from recognised screening instruments. These instruments are designed to flag when a professional conversation may be worthwhile — nothing more.
A high score does not mean you have a condition. A low score does not mean you do not. Only a qualified clinician who can assess you in context is able to diagnose a mental-health condition. Please do not make decisions about medication, therapy, work or study based on a score alone.
6. Symptom checker
The symptom checker organises what you report into a structured picture and suggests a level of care. It is deliberately cautious: where the picture is unclear it will recommend the safer route rather than the statistically likelier one. That means it will sometimes send you to a clinician you did not strictly need to see. That trade-off is intentional.
It does not produce a list of likely diagnoses, and it should never be used to rule a serious condition out.
7. Medication information
Information about over-the-counter products is reference material only — not a prescription, not a dosage instruction, and not a recommendation that you take anything.
- Always read the product label and follow it.
- Check the active ingredient on every product to avoid accidentally taking the same one twice under different brand names.
- Speak to a pharmacist or clinician before taking anything if you are pregnant or breastfeeding, treating an infant or young child, managing a chronic condition, or already taking other medication or supplements.
- Interaction checks are based on what you have recorded in the app. If your list is incomplete, the check is incomplete.
8. Wearable and tracking data
Consumer wearables and phone sensors are wellness devices, not medical instruments. Step counts, sleep stages, heart-rate readings and derived metrics can be inaccurate, and they are not cleared for diagnostic use. Insights generated from them are observations about your own trend data, not clinical findings. Never make a medical decision based on a wearable reading alone.
9. Time-off notes
A self-reported time-off note records what you told the app on particular dates. It contains no clinical opinion and is not a medical certificate. Acceptance is entirely at the discretion of your employer, school, university or insurer, and may not satisfy legal or policy requirements that call for a clinician's certificate.
10. Clinician consultations
Consultations are real clinical encounters with independently licensed professionals, and the clinician's advice — not the app's — is what you should follow. Telehealth has its own limits: a clinician cannot physically examine you, and may determine that an in-person visit, testing or emergency care is required. Follow that direction.
11. Special populations
Take particular care, and involve a clinician earlier, if any of the following apply:
- Pregnancy or breastfeeding
- Infants, young children, or anyone under 18
- Adults over 65
- Chronic conditions such as diabetes, heart disease, kidney or liver disease, or asthma
- A weakened immune system, or current cancer treatment
- Multiple prescription medications
- Recent surgery, hospital admission or injury
12. Your responsibility
You are responsible for your own health decisions and for those of anyone you care for. By using OnlineCare AI you accept that:
- You will not use it for emergencies
- You will not delay or disregard professional advice because of something the app said
- You will consult a qualified professional about your actual medical decisions
- You use the AI features at your own risk
This disclaimer forms part of our Terms & Conditions.
Questions about this document? Email legal@onlinecareai.com. For privacy requests use privacy@onlinecareai.com. For anything urgent about your health, contact a licensed clinician or your local emergency number.
