1. Introduction
These AI Terms of Use (“Terms”) govern your use of any Artificial Intelligence (“AI”) features and services provided by Clanwilliam Headquarters Limited and its Affiliates (“Clanwilliam”, “we”, “our”) across our software systems.
These Terms are supplemental to, and form part of, your existing software licence agreement, service contract, or end-user terms with Clanwilliam. In the event of any conflict, these Terms apply specifically to your use of AI Services.
Because AI technologies are evolving rapidly, and in response to changing regulations, best practices, and risks, we have taken the decision to provide these Terms via our AI Trust Centre. This ensures that you always have access to the most current and relevant version of our AI Terms, without being tied to a specific software release cycle.
Any updates or amendments to these Terms will be published at that address and users will be notified via email when material changes occur.
We embed compliance, transparency, and safety by design in all AI deployments.
By using or accessing any AI feature, you accept and agree to these Terms.
2. Updates and Change Management
Clanwilliam reserves the right to update, amend, or revise these AI Terms of Use at any time to reflect:
- Changes in technology, regulation, or legal requirements;
- New features or capabilities of the AI Services; or
- Evolving internal risk assessments or governance policies.
All changes will be published at our AI Trust Centre. When material changes occur — such as changes to data use, responsibilities, or limitations — users will be notified by email in advance of the effective date.
Your continued use of AI Services after such updates constitutes your acceptance of the revised Terms. You are encouraged to review the AI Terms of Use regularly to stay informed of any changes.
Where applicable, a changelog or archived version of previous terms will be retained for transparency and auditability.
3. Definitions
AI Services: Any tool, assistant, or functionality powered by artificial intelligence, including large language models (“LLMs”), summarisation engines, predictive analytics, and generative or assistive features embedded in Clanwilliam software.
Affiliates: Subsidiaries of Clanwilliam Headquarters Limited.
LLM: A machine learning model trained to generate, summarise, or respond using natural language.
User: Any authorised clinician, administrator, or staff member interacting with Clanwilliam products.
Sensitive Data: Personal data related to health, genetics, or other protected categories under applicable law.
High-Risk Use: Use of AI that may directly impact clinical care, diagnosis, patient safety, or regulatory compliance.
4. Purpose and Scope
These Terms apply to all AI features, tools, and services made available through Clanwilliam software platforms.
AI Services are intended to assist and enhance clinical, administrative, and operational workflows — not to replace professional judgment, human oversight, or legally required processes.
While these technologies can improve efficiency, accuracy, and insight, all outputs generated by AI Services must be reviewed, interpreted, and validated by a qualified human. The User remains fully responsible for decisions made.
5. Permitted Use and Restrictions
You are authorised to use Clanwilliam’s AI Services only in accordance with their intended functionality, as described in accompanying documentation and within the boundaries set out in these Terms.
Permitted Use
You may use the AI Services:
- As a support tool to enhance administrative efficiency, documentation, and workflow;
- In clinical or professional settings, provided outputs are reviewed and validated by a qualified professional before use; and
- In line with any applicable laws, clinical guidelines, or regulatory requirements in your jurisdiction.
Prohibited Use
You must not:
- Rely solely on AI-generated outputs for medical diagnosis, treatment, or clinical decision-making;
- Use AI Services to replace professional judgment or statutory clinical review obligations;
- Reverse engineer, disassemble, decompile, or attempt to extract source code or data models from the AI Services;
- Circumvent or disable security features;
- Use AI Services in any manner that violates laws, regulations, or third-party rights;
- Input or process any content that is unlawful, misleading, defamatory, or infringes the rights of others; and
- Represent AI outputs as independently verified facts or substitute them for legally required documentation.
Violations of these restrictions may result in suspension or termination of your access to AI Services and/or legal action.
6. Nature of AI Output
AI Services provided by Clanwilliam are based on LLMs and other probabilistic algorithms. As a result, the content produced may be:
- Non-deterministic (i.e. vary slightly each time depending on inputs and context);
- Incomplete, inaccurate, or overly confident in tone;
- Subject to bias or assumptions present in the training data; or
- Misinterpreted if not reviewed by a qualified human.
AI-generated summaries, drafts, suggestions, or prompts do not replace professional judgment and must always be reviewed, validated, and contextualised before being included in formal documentation.
By using these services, you acknowledge that the AI does not understand context as a human would, and its outputs are based on patterns and probabilities, not comprehension or intentionality.
7. Logging, Audit, and Monitoring
Clanwilliam maintains logging and monitoring protocols for all AI Services in order to:
- Ensure feature reliability, performance, and integrity;
- Investigate issues, bugs, or anomalies; and
- Comply with legal, contractual, and regulatory obligations.
What is logged:
Metadata such as time of use, features accessed, user session identifiers, and error messages may be logged.
What is not logged:
We do not log, store, or retain identifiable patient data, clinical content submitted to the AI, or any personal health information in usage logs. Our logging systems are designed to protect sensitive clinical inputs from retention or reprocessing.
Logs are retained only for as long as required to fulfil operational, diagnostic, or compliance obligations. All audit data is protected with appropriate access controls and is used exclusively for safety, compliance, and service improvement purposes.
8. Transparency and Disclosure
Clanwilliam is committed to making its AI Services understandable and trustworthy. We ensure that:
- All AI-generated content is clearly identified within the interface or system;
- Users are informed when outputs are generated by AI rather than a human; and
- Feature-level documentation is provided that explains what each AI tool does, how it works at a high level, and what its limitations are.
This approach supports our principle of Plain-Sight AI — ensuring that even the least technical user can understand when and how AI is being used.
Where feasible, in-product messaging or tooltips will indicate:
- That content is machine-generated;
- Whether the AI has been provided local records, external references, or structured data; and
- Any known caveats or reliability considerations.
9. Governance and Model Behaviour
Clanwilliam operates a centralised AI governance platform to ensure:
- Responsible and controlled development of all AI features;
- Version-controlled model deployment with audit trails;
- Risk-based evaluation of AI feature scope and usage context; and
- Technical and ethical review of new AI tools before launch.
We recognise that language models are probabilistic, not deterministic. This means:
- Outputs are based on patterns in data rather than verified facts;
- Responses may vary with different prompts, contexts, or phrasing; and
- Outputs can unintentionally include bias, outdated guidance, or hallucinated content.
As a result, all outputs from AI Services are provided with the explicit understanding that human validation is required before use.
Clanwilliam continually reviews performance, user feedback, and safety data to tune and improve AI tools over time.
10. Data Protection and Privacy
Clanwilliam processes data in accordance with the General Data Protection Regulation (GDPR) and the Data Protection Act 2018 (UK).
Key Data Protection Measures:
- All data is processed within private, secure, access-controlled environments;
- Personal data is never used to train external, public, or third-party models;
- No persistent storage of clinical inputs or outputs unless explicitly required for reconciliation or logging purposes; and
- All access and processing is role-based and logged.
Controller vs Processor Role:
Where AI Services are integrated into external systems (such as an Electronic Health Record [“EHR”]), Clanwilliam acts as a data processor under instruction from the data controller (e.g., the GP Principal or practice). We process data solely to deliver the agreed AI functionality and never for secondary purposes.
11. Opt-Out and Feature Management
If a customer wishes to disable any AI functionality they have previously enabled, this request should be managed directly through their practice’s EHR provider or system administrator.
Configuration options for enabling or disabling AI features may vary by product. In some environments, it may only be possible to activate or deactivate AI Services at a whole-practice or site level, rather than per individual user.
Clanwilliam will continue to offer configuration flexibility wherever technically feasible and will ensure that opt-out processes are documented and accessible via system support channels.
12. Incident Reporting and Misuse
If you encounter unsafe, misleading, or inappropriate behaviour from an AI feature — or suspect it is being misused — you must:
- Immediately cease use of the affected feature; and
- Report the issue to ai@clanwilliamgroup.com as soon as possible.
Clanwilliam will:
- Investigate the reported issue promptly;
- Take corrective action if a safety risk, legal breach, or violation of these Terms is confirmed; and
- Where necessary, suspend AI access for affected users or sites to prevent further risk.
Misuse includes, but is not limited to:
- Using AI-generated content without review;
- Attempting to bypass safety controls;
- Introducing prohibited or harmful prompts or content; and
- Using AI in unauthorised or non-clinical environments.
13. Intellectual Property
All rights, title, and interest in the AI models, training data, code, workflows, and related documentation are owned by Clanwilliam or its licensors.
You are granted a limited, non-exclusive, non-transferable, revocable licence to use AI Services only as part of your licensed Clanwilliam software and only for their intended use.
You may not:
- Copy, modify, or create derivative works of the AI Services;
- Extract or reuse any part of the AI models or underlying datasets; or
- Use the Services outside the scope of your licensed access.
You retain ownership of any original content or data you input into Clanwilliam software. However, any AI-generated outputs, feedback, or suggested edits are the intellectual property of Clanwilliam, and you are granted a right to use them only within the context of your service access.
14. Compliance and Law
These Terms are governed by the laws of Ireland, without prejudice to any mandatory protections that may apply under local law if you are based in another jurisdiction.
Clanwilliam’s AI Services are designed to comply with, and be interpreted in accordance with:
- The General Data Protection Regulation (GDPR);
- Applicable laws and guidance relating to clinical safety, information governance, and software as a medical device (SaMD), where relevant.
We proactively assess each AI feature for its regulatory classification and design compliance frameworks accordingly — including data minimisation, auditability, and human oversight. Where feasible, we apply the precautionary principle in cases of uncertainty, prioritising patient safety and professional accountability over speed of deployment.
15. Limitation of Liability
To the maximum extent permitted by applicable law, Clanwilliam disclaims all liability and responsibility for any harm, loss, or damage arising from the use of AI Services, whether direct, indirect, incidental, special, consequential, or otherwise.
This includes, without limitation:
- Any clinical errors, adverse events, or patient harm resulting from improper use of AI outputs;
- Lost revenue, data, or business disruption;
- Reliance on inaccurate, incomplete, or misleading AI-generated content; or
- Decisions made based on unvalidated outputs.
AI Services are provided “as is” and “as available” without warranties of any kind, whether express, implied, or statutory — including warranties of fitness for a particular purpose, accuracy, availability, or non-infringement.
You are solely responsible for validating the accuracy, appropriateness, and fitness of AI outputs before acting on them. Clanwilliam shall not be liable for any claims, losses, or liabilities arising from your failure to do so.
If you do not accept this limitation, you must not use the AI Services.
16. No Medical Advice
AI Services provided by Clanwilliam are intended solely to support administrative and workflow efficiency. They are not medical devices, and do not constitute clinical decision-making tools, diagnosis, or professional medical advice.
You must not rely on AI-generated outputs to independently assess, treat, or advise patients. Responsibility for clinical decisions remains solely with the qualified healthcare provider.
17. Indemnity
To the fullest extent permitted by law, you agree to indemnify, defend, and hold harmless Clanwilliam, its affiliates, officers, directors, employees, and agents from and against any and all claims, demands, liabilities, damages, losses, costs, or expenses (including reasonable legal fees) arising out of or in connection with:
- Your use or misuse of the AI Services;
- Any breach of these Terms;
- Improper reliance on AI-generated outputs without appropriate review or professional oversight; or
- Any violation of applicable laws, regulations, or third-party rights in connection with your use of AI Services.
This indemnity includes, without limitation, claims brought by patients, regulatory authorities, or third parties related to your decisions, actions, or omissions based on AI outputs.
18. De-identified Data Use
Clanwilliam may, from time to time, use de-identified and aggregated metadata generated through AI Services to improve functionality, monitor system performance, and enhance safety protocols. No identifiable health or patient information is used for this purpose.
