FCA authorised · FRN 724952 0117 325 0027Quote & buy →
Aipedia: Doctors and medical practices
A private clinic chatbot that answers symptom or treatment questions can amount to clinical advice, and the clinic is responsible for what it says. Whether medical malpractice, general PI or neither responds depends on the policy wording and who supervised the bot.
Part of: AI in doctors and medical practices
In short
Private clinics use AI chatbots on websites and messaging apps to answer enquiries, book appointments and respond to symptom or treatment questions. Once a chatbot suggests whether a patient needs treatment, or what a symptom means, it is giving advice the clinic may be liable for. Clinics registered with the CQC are inspected on safe care, and chatbot conversations are health data under UK GDPR. One defence organisation's published position on AI excludes fully autonomous systems and data breaches, so a clinic may need medical malpractice, technology PI and cyber cover working together.
A chatbot on a clinic's website or messaging account can answer questions about prices, opening hours and procedures, take bookings and send reminders. Some now use generative AI to answer free-text questions, such as whether a symptom is serious, whether a treatment suits someone, or what to do after a procedure. Some clinics connect chatbots to their booking and record systems so the bot can see appointment history.
Information about the clinic and its services is low risk. The line is crossed when the chatbot interprets a patient's symptoms, recommends or rules out treatment, gives aftercare advice for a complication, or tells someone they do not need to be seen. At that point the clinic is effectively giving clinical advice without a clinician. The GMC expects doctors to use professional judgement and discuss AI use with patients, which is hard if the bot responds unsupervised. If the chatbot's answer amounts to a significant decision based solely on automated processing of health data, the safeguards in the Data (Use and Access) Act 2025 apply, including the right to human intervention.
Illustrative scenario: a patient messages a private clinic's chatbot late at night describing swelling and pain after a procedure. The bot, trained on routine aftercare information, reassures them that this is normal and suggests a cold compress. The patient waits until morning and is then admitted with a serious complication. The patient complains that the clinic's advice delayed treatment.
Other risks include chatbots that promise outcomes, give pricing or suitability statements the clinic would not stand behind, or leak conversation data through a poorly configured integration.
Chatbots built on general-purpose generative AI can also produce confident answers outside their intended scope, such as dosing advice or comments on another provider's treatment. A clinic may not know what its bot has said unless it reviews transcripts. If a supplier updates the underlying model, behaviour can change without warning. The National Commission's work, accepted by the government on 6 October 2026, includes plans for guidance on other categories of AI, so expectations of clinics using patient-facing tools are likely to rise.
It depends on the wording. A medical malpractice policy usually covers clinical services provided by or under the supervision of named practitioners, and may not respond to advice generated autonomously by software. One defence organisation's published position excludes fully autonomous systems and data breaches. A technology or general professional indemnity policy may cover errors in information provided by the clinic, but may exclude clinical advice. A data breach usually needs cyber cover. See AI chatbot insurance, healthcare clinic insurance and cyber insurance and AI.
See also AI triage and online consultation, does medical indemnity cover AI and the doctors AI hub.
Apex arranges indemnity for private clinics. A broker reviews your cover and how each policy treats a patient-facing chatbot.
It can, but the clinic is then responsible for that advice. Safer designs limit the chatbot to administrative questions and send any symptom or treatment question to a clinician, with clear routes to emergency care for urgent symptoms.
Not necessarily. Malpractice cover usually focuses on care by or under the supervision of named practitioners, and one defence organisation's published position excludes fully autonomous systems. Check the wording and tell your insurer or broker what the chatbot does.
Yes, in practice. The GMC expects doctors to discuss AI use with patients, and the government has accepted that patients have a reasonable expectation of being told about AI. Data protection transparency rules also apply to chatbot conversations.
Tell us about your practice and the work you do. A broker will review how your current cover treats AI-assisted work, or arrange new cover. No fee for a quote. Or call 0117 325 0027.
Start an indemnity quote Request a callbackApex Insurance Brokers Limited is authorised and regulated by the Financial Conduct Authority. Registered in England and Wales, company number 07014570. This page is general information about how AI is used in the profession and how insurance may respond. It is not legal, regulatory or insurance advice on your circumstances, and it does not guarantee that cover will be available or on what terms.
