AI & Clinical Disclaimer

Cavitech AI incorporates artificial intelligence across two categories of features:

Clinical Decision Support features include automated radiograph analysis (teeth detection, pathology detection, bone level assessment), AI-generated findings and annotations, treatment plan suggestions, clinical report drafting, CBCT 3D guidance projections, and TMJ movement and audio analysis. These features produce AI Output that relates to clinical observations and must always be treated as preliminary suggestions requiring professional validation.

Operational features include ambient scribe (voice-to-SOAP-note transcription), conversational AI chat (normal, search, and thinking modes), appointment scheduling assistance, patient intake automation, and insurance pre-authorisation support. While these features streamline practice workflows, their output should still be reviewed for accuracy before being relied upon in a clinical or administrative context.

All clinical AI Output is subject to a mandatory approval workflow. Every AI-generated finding must be individually reviewed and explicitly approved or declined by a qualified dental professional before it is included in any report, treatment plan, or patient communication. No AI Output is acted upon automatically.

To ensure clarity regarding the scope and limitations of the Service, the following distinctions are essential:

Cavitech AI is NOT:

• A diagnostic device. It does not render diagnoses.
• A substitute for clinical examination. AI analysis of radiographs cannot replace a thorough intraoral and extraoral clinical examination, patient history review, or additional diagnostic tests.
• A replacement for professional training, education, or clinical experience. The Service assumes that its users possess the qualifications and judgment necessary to evaluate AI Output in context.
• An autonomous decision maker. The system never initiates treatment, prescribes medication, or takes any clinical action independently.
• A certified medical device under the FDA (United States), SAHPRA (South Africa), MHRA (United Kingdom), EU MDR (European Union), or any other regulatory authority. Cavitech AI has not been cleared, approved, or certified as a medical device in any jurisdiction.

Cavitech AI IS:

• A second opinion tool that surfaces observations for professional consideration.
• A clinical decision support system that assists, but never directs, clinical workflows.
• A practice management platform that streamlines scheduling, documentation, patient communication, and administrative tasks.
• A tool that augments the capabilities of dental professionals. It never replaces them.

While Cavitech AI strives for the highest possible accuracy, all AI systems have inherent limitations that users must understand:

Confidence scores. AI-generated findings may include confidence scores or probability indicators. These scores reflect the model's statistical certainty and should not be interpreted as clinical certainty. A high confidence score does not guarantee accuracy, and a low confidence score does not necessarily indicate an incorrect finding.

False positives and false negatives. Like all AI systems, Cavitech AI may produce false positives (identifying pathology where none exists) and false negatives (failing to identify pathology that is present). The approval workflow exists specifically to mitigate these risks through mandatory professional review.

Image quality dependency. The accuracy of AI analysis is directly dependent on the quality of the input radiograph. Factors such as exposure, positioning, resolution, artefacts, patient movement, and image compression can all affect AI performance. Poor quality images may yield unreliable results.

Absence of clinical context. AI analysis is performed on the radiographic image alone. The system does not have access to the patient's full medical history, current medications, symptoms, clinical presentation, or other diagnostic information that a treating dentist would consider. AI Output must always be interpreted within the broader clinical context that only the treating professional possesses.

The treating dental professional bears sole and complete responsibility for all clinical decisions, diagnoses, and treatment plans, regardless of whether AI Output was consulted in the process.

Specifically, you are responsible for:

• Reviewing and approving all findings. Every AI-generated finding must be individually assessed against your own clinical judgment. You must approve or decline each finding before it is incorporated into any clinical documentation.
• Conducting an independent clinical examination. AI radiograph analysis does not replace a comprehensive clinical examination. You must perform your own assessment, including visual inspection, probing, vitality testing, and any other diagnostics indicated by the clinical presentation.
• Obtaining informed consent. You must ensure patients are aware that AI-assisted tools are used as part of their care and that all findings are reviewed and confirmed by a licensed dentist. Informed consent procedures must comply with the requirements of your jurisdiction.
• Maintaining compliance. You are solely responsible for ensuring that your use of the Service complies with all applicable laws, regulations, professional standards, and ethical guidelines in your jurisdiction, including but not limited to scope-of-practice requirements, record-keeping obligations, and data protection laws.

When AI-assisted tools are used as part of patient care, dental professionals should ensure transparent communication with patients:

• Clearly communicate to patients that AI is used as a supplementary analytical tool within the practice, not as the primary basis for diagnosis or treatment planning.
• Explain that all AI-generated findings are reviewed, verified, and approved by their treating dentist before any clinical decisions are made.
• Ensure patients understand they may ask questions about how AI is used in their care and that they have the right to request that their treating dentist explain any AI-assisted findings in plain language.

Transparent patient communication reinforces trust and aligns with professional ethical obligations across all jurisdictions in which Cavitech AI operates.