Artificial intelligence (AI) and digital twins are transforming industries – from predictive maintenance and supply chain optimisation to sustainability and compliance. But their success depends on one critical ingredient: access to high-quality and diverse data.
The role of data spaces
To unlock the real potential of AI, companies must change their view on their current business models and the role of their data.
They need to understand that competitive advantage won't come from owning data, but from sharing it.
Organisations aiming to harness data and AI collaboratively often encounter two key challenges:
- Sharing data across boundaries while maintaining control and trust. Only if companies remain in full control over their data, are they willing to share it with others.
- Data is structured and interpreted differently across organisations. Without a shared understanding of data, even the most advanced AI models can produce misleading results.
Data spaces enable collaboration by providing a governed environment that supports secure, sovereign data exchange and help harmonise inputs by introducing common data models and semantic standards. Data spaces are structured ecosystems, built on principles of data sovereignty, interoperability, and trust, that allow multiple organisations to share and access data under clearly defined rules, ensuring that each participant retains control over their data while contributing to a shared pool of knowledge. Right now, many data spaces are being established in many different industries – from automotive and aerospace to energy, healthcare and chemicals.
Data spaces are no longer a niche concept – they’re reshaping entire industries. From automotive and aerospace to energy, healthcare, and chemicals, ecosystems are forming at speed. The International Data Space Association’s radar1 tells the story: by the end of 2024, data spaces surged from 78 to 152. Over half (55%) are already in implementation, and 11% are operational. And that was just the beginning. The growth continues as companies across sectors recognize the transformative power of shared data. The momentum is undeniable – data spaces are becoming the backbone of digital collaboration.
Data spaces hold transformative potential, enabling collaboration among thousands of stakeholders across and within industries, from manufacturers to OEMs, organisations and SMEs.
In this article, we explore how digital twins and AI models can benefit from a collaboration via data spaces.
Federated learning: privacy-preserving intelligence
Federated learning (FL) enables AI models to be trained across multiple decentralised datasets – without moving the data itself. Instead of sending data to a central server, the model is sent to where the data resides. It learns locally, and only the updated model parameters – not the raw data – are shared back. These updates are then aggregated to improve the global model.
This approach is particularly valuable in regulated industries like healthcare, finance, and manufacturing, where data privacy and compliance are paramount. It allows organisations to collaborate on AI development without exposing sensitive information.
Data spaces contribute to this approach with their data models that ensure data is interpreted in the same way in all contributing companies. Data spaces also ensure compliance by enforcing data sovereignty, privacy, and security policies and manage trust by providing means for authentication, authorisation and audit logging. The data space's governance and operating model also ensure trust in the service providers orchestrating the federated learning process as well as in the validity of the contributed data, respectively the trained model updates.
Digital twins: simulating reality for smarter decisions
Digital twins are virtual representations of physical systems, assets, or processes that mirror their real-world counterparts in real time. They continuously ingest data from sensors, systems, and operations to simulate behaviour, predict outcomes, and optimise performance.
In the context of data spaces, digital twins allow organisations to collaborate on shared models without compromising proprietary data.
For example, a manufacturer can simulate the lifecycle of a product across suppliers, logistics partners, and recyclers – without exposing sensitive internal data. Data organised via digital twins in a data space can then be used to train AI models.
Constructing a digital twin of an aircraft within a data space framework would allow different users to access different data from the same digital twin model. For example, an aircraft’s digital twin can provide tailored access: OEMs see full fleet data, airlines access their own planes, and suppliers view only relevant component data.
Strategic implications for decision makers
Strategically, data spaces are a possible link between digital transformation and revenue transformation. They enable new monetisation models – such as data-as-a-service and outcome-based offerings – while embedding trust and regulatory alignment into the core of data exchange.
For executives, the imperative is clear: participating in and orchestrating data ecosystems will increasingly define competitive advantage. Data spaces are not just technical infrastructure – they are strategic enablers of growth, resilience, and ESG alignment.
1) Source: https://www.dataspaces-radar.org/radar/
Die Autoren
