Biobanks, Data Sovereignty and Trade Secret Risk

The most valuable assets are now harder to detect, harder to prove and easy to transfer indirectly.

Recent media coverage has focused on China’s rapid expansion of genomic biobanks and rising geopolitical tension around cross‑border data flows. Although these developments are often framed as national security or public health issues, the more immediate risk for life sciences and technology companies is commercial: the erosion of proprietary advantage through the quiet loss of data, models and workflows that no longer fit neatly within traditional intellectual property regimes.

As value shifts away from patentable inventions and toward large‑scale datasets and artificial intelligence (AI)‑enabled insights, companies are increasingly reliant on trade secret protection. At the same time, global collaboration, distributed research teams and complex vendor ecosystems make those secrets difficult to define, control and enforce. The result is a widening gap between where value resides and how well it is protected.

What the Media Is Highlighting

Media attention has centered on China’s aggressive investment in population‑scale genomic databases, often supported by state funding, national health initiatives and integrated data infrastructure. In parallel, Western governments have moved to tighten access to sensitive health and genomic data through export controls, privacy regulations and localization requirements.

These developments are frequently presented as opposing approaches to innovation. In reality, they reflect a broader global shift toward data sovereignty—the idea that certain categories of data are strategic assets subject to heightened control. For companies operating internationally, this means that data access, transfer and use are increasingly shaped by regulatory friction and political scrutiny.

A Shift in What Counts as IP

The most consequential change may be economic. Competitive advantage in the life sciences is moving away from discrete, patentable inventions and toward data ecosystems:

• Longitudinal, deeply curated genomic datasets
• Proprietary labeling, cleaning and harmonization processes
• Trained machine‑learning models and feature representations
• Embedded institutional knowledge about how to generate insights from data

These assets are notoriously difficult to patent, both because of eligibility constraints and because disclosure undermines their value. As a result, companies rely primarily on trade secret protection, even though trade secrets were never designed to carry this much strategic weight.

Core Risk Vectors

Several recurring pathways account for most losses of proprietary advantage in this space:

• Cross‑border collaboration. Joint ventures, research partnerships and academic collaborations can expose sensitive datasets or model outputs to regimes with different disclosure norms and enforcement mechanisms.
• Model laundering. Even when raw data never leaves a jurisdiction, trained models or derived parameters may encode valuable information that can be reverse‑engineered or repurposed elsewhere.
• Talent mobility. Researchers and engineers often carry tacit knowledge—workflows, heuristics and experimental judgment—that is difficult to document and even harder to restrict after departure.
• Vendor exposure. Cloud providers, CROs, data processors and AI vendors may have technical access to assets that exceed what is contractually intended, creating silent leakage risks.

Why Enforcement Is Difficult

Trade secret enforcement in the biobank and AI context is uniquely challenging.

First, attribution is hard. When value is embedded in models or workflows, it can be difficult to prove misappropriation as opposed to independent development. Second, jurisdictional limits complicate investigation and discovery when alleged misuse occurs abroad. Third, damages can be difficult to prove, particularly where the harm is loss of future competitive position rather than immediate revenue.

These challenges mean that even well‑founded claims often fall short of delivering meaningful remedies if raised too late.

Litigation Trends

Reflecting these realities, litigation strategies are evolving. Plaintiffs increasingly emphasize:

• Contract‑based claims (NDAs, licensing agreements, data use restrictions) rather than pure statutory trade secret claims;
• Hybrid theories combining trade secret misappropriation, breach of fiduciary duty and unfair competition; and/or
• Early injunctive relief, aimed at freezing models, data use or collaborations before irreversible dissemination occurs

Success in these cases depends less on dramatic courtroom revelations and more on whether the company can show disciplined asset governance from the outset.

How We Can Help

We work with companies at the intersection of life sciences, data and AI to help them anticipate and manage these risks before they crystallize into disputes. Our work includes:

Structuring research collaborations and data‑sharing arrangements to preserve trade secret protection, including:

• Asset segmentation. Clearly separate crown‑jewel datasets, derived datasets and expendable data, with controls calibrated accordingly.
• Model governance. Track how models are trained, what data they ingest and where model artifacts are deployed or shared.
• Cross‑border architecture. Design data flows and compute environments to minimize unnecessary exposure while still enabling collaboration.
• Contractual rigor. Align contractual rights, audit provisions and termination consequences with the realities of data‑driven value.
• Auditability. Build internal documentation and access logs that can support early injunctions if disputes arise.

Advising on global data governance and localization strategies aligned with business objectives.

Stress‑testing existing contracts and workflows for hidden exposure points.

Preparing for high‑stakes disputes by building evidentiary records that support rapid injunctive relief.

Our goal is not only to defend assets in litigation, but to ensure they are structured to remain defensible in the first place.

---

This document is intended to provide you with general information regarding the intellectual property law implications for biobanks. The contents of this document are not intended to provide specific legal advice. If you have any questions about the contents of this document or if you need legal advice as to an issue, please contact the attorneys listed or your regular Brownstein Hyatt Farber Schreck, LLP attorney. This communication may be considered advertising in some jurisdictions. The information in this article is accurate as of the publication date. Because the law in this area is changing rapidly, and insights are not automatically updated, continued accuracy cannot be guaranteed.