Usage Policy

Version 1, Effective: 1st October 2026

This Usage Policy sets out the terms governing acceptable use of the MedicDesk.ai platform, developed and operated byACE Ltd (“MedicDesk.ai”, “we”, “us”, “our”) and our affiliates.

The MedicDesk.ai platform provides two connected services to healthcare clinics and their staff: the Ambient Scriber, which supports clinical documentation through AI-assisted transcription and note drafting, and the AI Receptionist, which supports front-desk and call-handling workflows through an AI voice and messaging agent. Both are described further in §2 below.

This Usage Policy sits alongside — and does not replace — our Terms of Use, Data Processing Addendum, and Privacy Policy. Where this Usage Policy and the Terms of Use address the same subject, the Terms of Use govern the contractual and legal relationship between the parties; this Usage Policy governs what constitutes acceptable, in-scope use of the Services day to day. Capitalized terms not defined here have the meaning given in the Terms of Use.

By accessing or using the Services, users acknowledge that they have read and understood this Usage Policy and agree to abide by it, together with the Terms of Use.

1. Purpose

MedicDesk.ai’s mission is to give doctors and their staff back the time that charting and phone traffic take from patient care — not to replace clinical judgment, and not to practice medicine. The Ambient Scriber and AI Receptionist are administrative and documentation aids built to sit inside a clinician’s or clinic’s existing workflow, not to sit in place of it.

This Usage Policy exists to make that boundary concrete: what the Services are for, what they are not for, and what every user of the Services — Providers, Assistants, and Admins alike — is responsible for when using them.

2. AI Functionality

2.1 Ambient Scriber

The Ambient Scriber captures audio from a clinical encounter, transcribes it, and uses AI to generate a structured draft clinical note from that transcript. The Provider must review the transcript and draft note, correct anything that needs correcting, and sign the note before it carries any clinical weight or forms part of the Patient record. Final responsibility for the encounter’s documentation, and for every professional and medical decision relating to a Patient’s diagnosis, advice, or treatment, rests solely with the treating Provider — never with the Ambient Scriber.

2.2 AI Receptionist

The AI Receptionist uses AI to answer inbound calls, converse with callers, and — where enabled — propose appointments and send or place outbound reminders, confirmations, and recall communications. Where the AI Receptionist proposes booking, rescheduling, or cancelling an appointment, that proposal is held for the clinic’s own review and approval before it is finalized, consistent with the same reviewed-before-acted-on principle that governs the Ambient Scriber’s draft notes. The AI Receptionist does not make final clinical or scheduling decisions on the clinic’s behalf.

2.3 Draft, Not Final

Every output the Services generate — a draft clinical note, a tagged entity, a call summary, a proposed appointment, or an AI-drafted response to a caller — is a draft for a human to review, not a finished, authoritative record. This principle underlies every limitation in §3.

3. Limitations

3.1 No Replacement for Professional Medical Advice

The Services and their outputs must not be construed as professional medical or healthcare advice, a definitive diagnosis, or a treatment recommendation. They are not intended to directly assess, maintain, diagnose, treat, or improve the physical, mental, or emotional health of a Patient, and they are not a certified medical device.

3.2 Prohibited Uses

Users must use the Services consistently with their intended purpose and this Usage Policy. Without limiting the foregoing, users must not:

• intentionally input fake, misleading, or deceptive data or prompts into the Services, or attempt to prompt or manipulate the Ambient Scriber into producing a clinical diagnosis, differential diagnosis, or treatment recommendation of the kind described in §3.1;
• use a bot, script, or automated tool to access, scrape, harvest, index, copy, or systematically extract content, data, or other materials from the Services;
• reverse-engineer, decompile, or otherwise attempt to discover the source code, models, prompts, or underlying technology of the Services, except to the extent applicable law permits despite this restriction;
• circumvent, disable, or interfere with any rate limit, usage cap, security feature, or other technical restriction applied to the Services;
• use the Services or their outputs to develop, train, or improve a competing AI model or service;
• use the AI Receptionist for unsolicited bulk communications, robocalling, or spam-like activity, or to contact a Patient by a channel or at a time the Patient has opted out of or that falls outside applicable telemarketing or quiet-hours rules; or
• use the Services in violation of applicable law or in a manner that infringes any third party’s rights.

A breach of this section may result in immediate suspension or termination of access, as described in §3.3 and the Terms of Use.

3.3 Restrictions on Use

We may place reasonable limitations on use of the Services (including rate limits, session or usage caps, or restricted access to certain features) where we consider use goes beyond what is reasonable and expected, is inconsistent with this Usage Policy, or otherwise at our discretion — for example, the Ambient Scriber’s Free tier (currently 20 sessions per Provider, ongoing) and the platform-wide 14-day trial described on our website and in the Terms of Use.

Where we impose such a limitation, we may offer the option to move to a higher-tier plan to access expanded usage allowances instead. If we reasonably believe a user has breached this Usage Policy, or that suspension is necessary to protect MedicDesk.ai, our users, or a third party, we may suspend, restrict, or terminate that user’s access to the Services, in whole or in part, with or without prior notice.

3.4 Review and Verification by Users

Users are responsible for reviewing and verifying the accuracy of all Service outputs — including transcripts, draft notes, tagged entities, translated section headers, Ask AI answers, call summaries, and proposed appointments — before relying on them for clinical decision-making, patient care, or clinic operations. This responsibility applies in full even where the Services offer functionality designed to help a user verify an output (for example, entity highlighting or a side-by-side transcript view).

Providers must exercise independent professional judgment when assessing whether any Service output is clinically relevant, accurate, and appropriate for the Patient in front of them.

4. Disclaimer of Warranties

Except as otherwise provided in the Terms of Use, the Services are provided “as is” and without warranties of any kind, express or implied. We disclaim all warranties, including the implied warranties of merchantability, fitness for a particular purpose, and non-infringement. Users acknowledge that the Services may not be free from errors or inaccuracies and are not intended to provide medical advice. See the Terms of Use for the full warranty and liability terms governing the Agreement.

5. Multi-Language Support

The Ambient Scriber can transcribe a clinical encounter and generate a draft note directly in the language the encounter was conducted in, and lets a Provider confirm a translated version of a note template’s section headers for use in future visits in that language. This functionality is intended for Providers and staff who are fluent in, and able to independently verify the accuracy of, the language used — it does not replace a Provider’s own review of the transcript and draft note under §3.4, and it does not itself verify clinical accuracy across languages.

Users are encouraged to report inaccuracies or unexpected behavior in transcription or translation through the Services’ built-in Help/Feedback flow, described in §9.

6. Ask AI

The Ambient Scriber’s Ask AI feature lets a Provider ask for a note to be reworked (for example, shortened, restructured, or translated) or ask a general question, which may draw on web search for background information such as current guidelines or reference material.

6.1 Rework Requests

A rework request only changes the wording, structure, or language of an existing draft note — it does not add clinical facts beyond what the note or the Provider’s instruction already contains. The resulting note remains a draft subject to the same review-and-sign requirement as any other Ambient Scriber output (§2.1, §3.4).

6.2 Lookup Requests

A lookup request is general research support, not a Patient-specific clinical recommendation. It is not part of the Patient’s note, and using it does not itself change or add to the Patient record. Providers are responsible for independently verifying any information returned, including by checking original sources where possible, before applying it to a clinical decision — Ask AI must not be treated as a substitute for a Provider’s own clinical judgment or for consulting current, authoritative guidance directly.

7. AI Receptionist — Appointments, Emergencies & Outbound Communications

7.1 Appointment Handling

The AI Receptionist may propose booking, rescheduling, or cancelling an appointment based on a call, but does not finalize that change itself — clinic staff review and approve proposed appointments before they take effect, consistent with §2.2. Customer is responsible for reviewing proposed appointments promptly enough that patient care and scheduling are not adversely affected by the delay.

7.2 Emergency Calls

The AI Receptionist is not an emergency service and must not be relied on, configured, or presented as one. Customer is responsible for ensuring genuine medical emergencies are directed to local emergency services rather than handled by the AI Receptionist, and for configuring the AI Receptionist’s escalation behavior (including any decision rules and knowledge base content it draws on) to reflect this.

7.3 Outbound Communications, Consent & Opt-Out

Where Customer enables outbound text or voice communications (for appointment confirmations, reminders, or recalls), Customer is responsible for ensuring it has a lawful basis to contact each Patient by the channel and at the time used, and for complying with applicable telemarketing, do-not-call, quiet-hours, and consent laws. The Services provide configurable tools to record a Patient’s request not to be contacted or called and to respect configured contact-time windows; Customer remains responsible for keeping Patient contact preferences current and for the underlying lawfulness of each communication — see §3.2’s prohibition on robocalling and spam-like use, and the Terms of Use for the full allocation of responsibility.

7.4 AI Disclosure

Where required by applicable law — including Article 50(1) of the EU AI Act — the AI Receptionist discloses to a caller, at the outset of a voice interaction, that they are speaking with an AI system and on whose behalf it is operating. Customer must not disable, remove, or alter this disclosure.

8. Reliance on MedicDesk.ai & Compliance with Regulations

8.1 Reliance on MedicDesk.ai

Providers and staff using the Services must exercise caution and critical judgment when interpreting their outputs. The Services provide administrative and documentation support, not a basis for clinical decision-making — the ultimate responsibility for patient care and decisions always lies with the treating Provider.

8.2 Compliance with Regulations

We develop and operate the Services to support compliance with applicable healthcare data and AI regulation, including the GDPR (EU/UK) and, where applicable, the EU AI Act — see §7.4’s disclosure requirement, already implemented for the AI Receptionist’s inbound calls. Users are responsible for ensuring their own use of the Services complies with all laws, regulations, and professional standards applicable in their jurisdiction, including licensing, scope-of-practice, informed consent, record-keeping, and patient-confidentiality obligations.

9. Data Privacy, Security & Reporting Issues

9.1 Data Privacy and Security

Providers and staff using the Services must adhere to the data privacy and security practices described in our Data Processing Addendum and Privacy Policy, including our published retention practice: raw audio is retained for up to 14 days and then automatically deleted, and the resulting clinical note/transcript is retained for up to 7 years, consistent with standard medical-record retention, unless a Provider configures a different, consent-backed retention setting available in the platform. We do not use Customer Data to train AI models made available to other customers, except with Customer’s express, separate consent.

9.2 Reporting Issues

Any concern about the Services’ performance, functionality, or adherence to this Usage Policy — including a suspected inaccuracy, an unexpected AI output, or a possible security issue — should be reported promptly through the Services’ built-in Help/Feedback flow, or by emailing us, so we can investigate and resolve it.

10. General

We may amend or modify this Usage Policy at any time. Users will be notified of material changes, and it is each user’s responsibility to review and adhere to the updated policy. By using the Services, users acknowledge that they have read and understood this Usage Policy, together with our Terms of Use, and agree to abide by both.

11. Contact

For questions or further information regarding this Usage Policy, please contact us at:

MedicDesk.ai

Email: connect@medicdesk.ai