The perfect idea for an AI application that saves lives or reduces CO2 emissions—but instead of failing because of the code, it fails due to fears of legal violations? In Baden-Württemberg, this very scenario is set to become a thing of the past. With the recently published white paper “Ethical and Legal Requirements for an AI Data Platform,” the “Ethics & Law” task force of the AI Alliance—led by the University Medical Center Freiburg, the University of Tübingen (IZEW), and the FZI Research Center for Information Technology—is laying the foundation for a digital ecosystem in which trust is already firmly built in.
Artificial intelligence opens up new opportunities for business, society, and government—from smart mobility to medical research. But data-driven innovation requires more than just algorithms: it requires trust, legal certainty, and clear rules.
With the recently published white paper “Ethical and Legal Requirements for an AI Data Platform,” the “Ethics & Law” task force of the Baden-Württemberg AI Alliance lays the groundwork for exactly this. The result is a framework that outlines how a cross-sector AI data platform can be designed responsibly—from a technical, legal, and organizational perspective.
A data platform as a strategic building block for Baden-Württemberg
With the development of the AI data platform, the AI Alliance is creating a key component for promoting data-driven innovation in the state. The goal is to systematically facilitate access to high-quality, legally compliant, and ethically sound AI assets—that is, data and models. At the heart of this effort is trustworthy, public-interest-oriented digitalization that enables innovation while safeguarding the rights and interests of all stakeholders.
The platform’s architecture deliberately breaks with the traditional concept of a central data repository. Instead of centralizing data, the AI Alliance relies on a federated broker model: data ownership remains with companies, hospitals, or research institutions. The platform makes metadata discoverable, comparable, and usable in a legally compliant manner.
Carolin Heizmann of the University Medical Center Freiburg, a co-author of the white paper, emphasizes: “Technologies are not neutral. Platforms, in particular, have a normative impact. That is why ethical and legal principles must be an integral part of the architecture from the very beginning.” Data sovereignty thus becomes not just a buzzword, but a technical design principle.
Ethics-by-Design as a Driver of Innovation
The white paper takes an integrated approach that combines ethics-by-design, compliance-by-design, and embedded ethics. Ethical considerations and legal requirements are not addressed as an afterthought, but are embedded early on in development processes and organizational structures.
Normative guidelines and concrete recommendations for action were developed through interdisciplinary exchange forums and workshops held in Tübingen and Freiburg. Technology, law, and ethics were consistently considered together—for example, through practical scenarios drawn from mobility, industrial quality assurance, and personalized medicine.
Dr. Thomas Usländer, co-author and project manager of the AI data platform at Fraunhofer IOSB, sums up the strategic dimension: “Innovation, ethical principles, and legal safeguards are not mutually exclusive. Integrating regulatory requirements into technical specifications at an early stage creates planning certainty—especially for small and medium-sized enterprises.”
This is a key advantage for small and medium-sized enterprises. Many SMEs do not have dedicated legal or AI departments. An infrastructure that systematically integrates requirements from the GDPR, the AI Act, or the Data Governance Act reduces uncertainty—and lowers barriers to market entry.
Values as a guide for our operations
The white paper translates core value principles into concrete design and governance requirements. Autonomy, security, transparency, fairness, sustainability, and accountability are not discussed in abstract terms, but are operationalized—for example, through:
- Role-based access systems and auditability
- Mechanisms for bias detection
- Interoperability with European initiatives such as GAIA-X or the European Health Data Space
- Sustainability criteria for infrastructure and data centers
This does not result in a static set of rules, but rather in a dynamic framework. The document lays the foundation for the sustainable further development of the AI data platform and, at the same time, identifies areas where further work is needed to permanently embed ethical and legal principles into its operations.
Framework and Source of Inspiration
This white paper is explicitly intended to be more than just an internal guideline. It serves as a framework for the ongoing integration of ethical and legal principles into data-driven AI innovations—in Baden-Württemberg and beyond.
At the same time, it is intended to provide impetus for similar platform initiatives and serve as a guide for all stakeholders involved in the development, use, or governance of the AI data platform.
As a cooperative initiative and central hub for AI in the state, we have a clear goal: AI not for its own sake, but as a solution to real-world challenges facing business, government, and society. The AI data platform is a strategic component of this mission. It provides guidance in the fragmented data market, facilitates access to quality-assured AI assets, and aligns regional needs with European standards.
The white paper clearly outlines the principles guiding the design of this infrastructure. Trustworthy technology design is not merely an afterthought—it is firmly embedded in the architecture.
Curious?
The white paper is publicly available and is aimed at stakeholders from the fields of technology, law, politics, and business—ranging from developers and small and medium-sized enterprises to policymakers.