This research begins with a simple but essential question: In predictive vehicle maintenance, where should raw data from a car be stored? Some argue that it makes sense to keep it within the vehicle, managed directly by its control units. Others prefer sending it to a backend system for evaluation by larger models. We found that this decision cannot be made based solely on technical reasoning. The legal conditions surrounding personal data, cybersecurity, product responsibility, and AI-based systems in the European Union carry equal weight in shaping feasibility. These regulations not only impose constraints but also actively influence the design and application of such technologies in real-world scenarios. The findings suggest that legal barriers frequently limit implementation more than current technical capabilities. Addressing these constraints early in the design process is crucial, as they define not only what types of data processing are permitted but also how predictive maintenance services can be delivered to users. To support responsible development, this study proposes a legal-technical framework tailored to the European regulatory landscape. It aligns predictive maintenance systems with key European Union regulations, including the GDPR, Cybersecurity Act, Cyber Resilience Act, and AI Act.

No description provided