Skip to main content

Chapter 6 — Intellectual Property

All intellectual property developed for Vinova Lab and its products should belong to Vinova Lab, subject to applicable law and properly executed agreements.

The document intelligence engine and the domain knowledge accumulated within it represent the company's primary and most strategically significant intellectual property. Unlike code — which can be rewritten — the engine's accumulated domain intelligence, classification models, and training data are the product of real operational use and cannot be replicated by a competitor starting from zero.

This includes:

  • the document intelligence engine and its architecture;
  • domain intelligence models, classification schemas, and extraction patterns;
  • anonymised training datasets and the pipeline that generates them;
  • document type taxonomies built through real-world processing;
  • software and source code;
  • models, configurations, and fine-tuned components;
  • processing pipeline and orchestration logic;
  • APIs and database schemas;
  • designs and user-interface components;
  • documentation;
  • brands and trademarks;
  • internal frameworks;
  • knowledge bases;
  • reusable integration components.

Open-Source and Third-Party Components​

Vinova Lab's products are built in part on open-source models, frameworks, and libraries.

The use of open-source components must be managed carefully:

  • the licence terms of every open-source component must be evaluated before adoption;
  • licences that impose obligations on derivative works — including copyleft licences and model-specific licences common in the open-source model ecosystem — must be reviewed before fine-tuning or integration into proprietary products;
  • open-source components must be tracked, documented, and attributed appropriately;
  • the boundary between open-source components and Vinova Lab's proprietary additions must be clearly maintained.

Proprietary improvements, fine-tuning, and domain-specific adaptations built on open-source foundations remain the intellectual property of Vinova Lab, subject to the licence terms of the underlying component.

Customer Projects​

Customer contracts should distinguish among:

  • customer-owned data and documents;
  • customer confidential information;
  • deliverables created exclusively for the customer;
  • Vinova Lab background intellectual property;
  • third-party and open-source intellectual property;
  • generic and reusable Vinova Lab engine capabilities.

Customer data, confidential information, or proprietary business logic must never be reused improperly.

Generic engine capabilities developed or improved during a customer project may remain part of the Vinova Lab platform only where this is legally, ethically, and contractually permitted.

Anonymised structural signal. When documents are processed as part of a customer engagement, structural and semantic patterns may be extracted in fully anonymised form — stripped of all personal data, sensitive information, and customer-identifiable content. These anonymised patterns, which bear no resemblance to the original documents or their specific content, may be used to improve the engine. Customer contracts must address this explicitly, making clear what is extracted, how it is anonymised, and what it is used for.

Contributors​

Founders, employees, contractors, and external collaborators contributing to Vinova Lab products should execute appropriate confidentiality and intellectual-property assignment agreements.

Final legal provisions must be prepared and reviewed by qualified legal professionals in the applicable jurisdiction.

The Vinova Lab Blueprint — Version 0.1 — Confidential Working Draft