This Acceptable Use Policy (“AUP”) sets out the rules for using CliniVoice AI. It forms part of your agreement with CliniVoice AI Ltd and should be read alongside our Terms of Service. Violations may result in suspension or termination of your account.
1. Scope
This AUP applies to all users of CliniVoice AI, including individual healthcare professionals and staff members of organisations with a CliniVoice AI subscription.
By accessing or using the Service, you agree to comply with this AUP. If your Organisation has deployed CliniVoice AI, your Organisation’s own information governance policies also apply and should be followed alongside this AUP.
2. Permitted Use
You may use CliniVoice AI to:
- Record and transcribe your own clinical dictation for the purpose of creating clinical correspondence (e.g. clinic letters, discharge summaries, referral letters, GP letters)
- Use AI-assisted letter generation to draft clinical documents, provided you review and approve all output before use
- Store transcriptions and letters within the Service for your own clinical practice
- Export generated letters for inclusion in patient records systems (e.g. via copy/paste or document download)
- Use template and quick-phrase features to improve efficiency
- Access the Service from approved devices in accordance with your Organisation’s IT and information governance policies
3. Prohibited Use
You must NOT use CliniVoice AI to:
- Create fraudulent or fabricated clinical records. All dictated content must reflect genuine clinical encounters.
- Process patient data without a lawful basis. You must have appropriate authority (e.g. direct care relationship, explicit patient consent) before dictating patient-identifiable information.
- Share your account credentials with any other person. Each user must have their own account for audit trail purposes.
- Use the Service on unsecured or shared public devices (e.g. library computers, shared kiosks) where patient data could be exposed.
- Attempt to extract, reverse-engineer, or copy the AI models, algorithms, training data, or source code of the Service.
- Use automated scripts, bots, or scraping tools to interact with the Service.
- Circumvent security controls, including authentication, access controls, or rate limits.
- Upload malware, viruses, or malicious content through the Service.
- Use the Service for any non-clinical or commercial purpose not authorised under your subscription (e.g. legal transcription, market research, journalism).
- Use AI-generated output without review as a final clinical document. All output must be checked by a qualified clinician.
- Record patients, relatives, or colleagues without their knowledge or without following your Organisation’s recording consent policies.
- Process data relating to children (under 18) without additional safeguards as required by your Organisation and applicable law.
4. Patient Data & Consent
4.1. You are responsible for ensuring you have a lawful basis for processing any patient-identifiable data through CliniVoice AI. In most clinical settings, this will be Article 6(1)(e) (public task) or Article 6(1)(b) (contract) combined with Article 9(2)(h) (health/social care provision).
4.2. If your Organisation requires explicit patient consent before using AI-assisted dictation tools, you must obtain and document that consent before recording.
4.3. Where possible, minimise the amount of patient-identifiable information included in recordings. Use hospital numbers or NHS numbers rather than full names where your clinical workflow permits.
4.4. You must inform patients if you are recording your clinical consultation for dictation purposes, in line with GMC/NMC guidance on transparency and your Organisation’s policies.
5. Clinical Use Obligations
By using CliniVoice AI, you confirm and agree to the following:
- You are a registered healthcare professional (or working under the direct supervision of one). CliniVoice AI is not for use by the general public for self-diagnosis or self-management of health conditions.
- You will review all AI-generated output before using it in patient care or adding it to any patient record. AI Output is a draft; you bear clinical responsibility for its content.
- You will not process data relating to patients you are not treating or responsible for. Processing patient data without a legitimate care relationship and lawful basis is a breach of this AUP and UK GDPR.
- You will not share your CliniVoice AI account credentials with any other person. Each clinician must have their own account for audit trail and information governance purposes.
- You will not use AI-generated clinical letters without review as final documents. Every letter must be checked for accuracy, completeness, and freedom from hallucinations before use in patient care.
- You will comply with your Organisation’s information governance policies when accessing the Service on NHS or other organisational devices and networks.
6. Clinical Responsibility
5.1. CliniVoice AI is a productivity tool, not a clinical decision support system. It does not diagnose, recommend treatments, or provide medical advice.
5.2. You retain full clinical and professional responsibility for all documents you produce using the Service. The AI assists with transcription and formatting; the clinical content and accuracy are your responsibility.
5.3. You must review every AI-generated letter for:
- Factual accuracy (patient details, clinical findings, medications, dosages)
- Completeness (no missing information from the consultation)
- Absence of AI hallucinations (fabricated details not present in the original dictation)
- Appropriate clinical tone and language
5.4. If you identify errors in AI-generated content, you must correct them before the document enters the patient record. Report persistent or significant errors to us at support@clinivoice.com so we can improve the Service.
7. AI Output Responsibilities
6.1. AI-generated text may contain inaccuracies, omissions, or “hallucinations” (plausible-sounding but incorrect information). This is an inherent limitation of current AI technology.
6.2. You must not rely on AI Output as a substitute for your own clinical assessment or professional judgement.
6.3. You must not present AI-generated content to patients, colleagues, or regulatory bodies as if it were solely your own work without disclosing that AI assistance was used, where your Organisation’s policy requires such disclosure.
6.4. CliniVoice AI Ltd accepts no liability for clinical outcomes, regulatory sanctions, or professional conduct complaints arising from your use of AI-generated content without adequate review.
8. Security Obligations
You must:
- Use a strong, unique password for your CliniVoice AI account
- Enable two-factor authentication if available
- Lock your device when leaving it unattended
- Only access the Service on devices that comply with your Organisation’s security policies
- Report any suspected security incident or unauthorised access immediately to security@clinivoice.com and your Organisation’s IT team
- Not download or store patient data from CliniVoice AI onto personal devices unless explicitly permitted by your Organisation
- Log out of your session when using shared or borrowed devices
9. Reporting Violations
If you become aware of any violation of this AUP, whether by yourself or another user, you should report it to:
- CliniVoice AI: security@clinivoice.com
- Your Organisation: Your line manager and/or information governance team
We take all reports seriously and will investigate promptly. Reports can be made anonymously where necessary.
10. Enforcement
9.1. Violations of this AUP may result in:
- A written warning
- Temporary suspension of your account
- Permanent termination of your account
- Notification to your employing Organisation
- Reporting to relevant regulatory or law enforcement bodies where required by law
9.2. We will take proportionate action based on the severity, intent, and impact of the violation. We will provide you with an opportunity to respond before permanent action is taken, except where immediate action is necessary to protect data subjects or system security.