LEGAL
ClinicalPearls Data License Agreement
Version v1 — Effective 2026-05-24
This Data License Agreement (the "Agreement") governs the use of any dataset or data bundle (the "Dataset") licensed by ClinicalPearls ("Licensor") to a buyer ("Licensee"). By accessing the Dataset, Licensee agrees to be bound by this Agreement.
1. License Grant
Subject to the terms of this Agreement and Licensee's payment of the agreed fees, Licensor grants Licensee a non-exclusive, non-transferable, worldwide, perpetual license to use the Dataset for:
- (a) training, fine-tuning, evaluating, benchmarking, and red-teaming machine learning models, including large language models and multimodal models;
- (b) internal research, including derivative analyses, statistical summarization, and the creation of derived datasets that are not substantially reconstructive of the Dataset; and
- (c) using outputs of models trained on the Dataset in Licensee's commercial products and services.
2. Restrictions
Licensee shall not:
- (a) redistribute, resell, sublicense, or publish the Dataset or any substantial portion of it in raw or substantially-reconstructive form;
- (b) attempt to re-identify any individual, whether patient, author, or reviewer, referenced in or associated with the Dataset;
- (c) use the Dataset to provide clinical decision support to identified patients, or as a substitute for licensed medical advice; or
- (d) remove or alter the dataset manifest, license file, or schema documentation included in the Dataset bundle.
3. Representations and Warranties
Licensor represents that:
- (a) each case in the Dataset is accompanied by an immutable per-case author attestation, the cryptographic hash of which is recorded in the bundle manifest;
- (b) each case has been reviewed and verified by a ClinicalPearls reviewer whose user identifier, role, and verification timestamp are recorded in the bundle manifest; and
- (c) Licensor has obtained from each author the rights necessary to grant the license in Section 1.
THE DATASET IS PROVIDED "AS IS." LICENSOR DISCLAIMS ALL OTHER WARRANTIES, EXPRESS OR IMPLIED, INCLUDING WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, AND NON-INFRINGEMENT.
4. Updates and Versioning
Bundles are versioned by build hash and schema version. Licensor may, but is not obligated to, provide updated bundles. If an Update Right is included in the order form, Licensee is entitled to receive new bundles published during the Update Right window without additional fee.
5. Confidentiality
The Dataset and the terms of this Agreement are confidential. Licensee shall not disclose either to any third party except to its employees and contractors who have a need to know and who are bound by confidentiality obligations at least as protective as this Section.
6. Indemnification
Licensor shall defend, indemnify, and hold harmless Licensee against any third-party claim alleging that the Dataset, as delivered, infringes a U.S. copyright or contains protected health information of an identifiable individual, provided Licensee gives prompt notice and reasonable cooperation. This is Licensor's sole liability and Licensee's sole remedy for such claims.
7. Limitation of Liability
EXCEPT FOR LICENSOR'S OBLIGATIONS UNDER SECTION 6 AND EITHER PARTY'S BREACH OF SECTION 2 OR SECTION 5, NEITHER PARTY SHALL BE LIABLE FOR ANY INDIRECT, INCIDENTAL, OR CONSEQUENTIAL DAMAGES. EACH PARTY'S TOTAL LIABILITY UNDER THIS AGREEMENT IS LIMITED TO THE FEES PAID OR PAYABLE BY LICENSEE IN THE TWELVE MONTHS PRECEDING THE CLAIM.
8. Termination
This Agreement terminates automatically upon Licensee's material breach of Section 2. Sections 2, 3 (last paragraph), 5, 6, 7, and 9 survive termination.
9. Governing Law and Venue
This Agreement is governed by the laws of the State of Delaware, without regard to conflict of laws principles. Exclusive venue lies in the state and federal courts of New Castle County, Delaware.
Order forms reference this Agreement by version. The version delivered with the Dataset bundle governs that delivery.
Questions: support@clinicalpearls.ai
