Summary The proposed Cloud and AI Development Act (CADA) establishes eight "grand challenges" in Annex I to drive large-scale, cross-sectoral innovation across Europe's digital ecosystem. As defined in Article 6(2), these challenges serve as the operational vehicle for the Cloud and AI Leadership Initiatives, addressing "major technological and industrial challenges of strategic relevance for the Union." The eight pillars are: (1) Environmental sustainability, performance and security of data centres; (2) Cloud stacks; (3) Frontier AI; (4) Physical AI; (5) Industrial AI; (6) Cooperative European Industrial Models; (7) AI Agents Platform; and (8) Public Sector AI. These are not merely research goals but actionable frameworks designed to bridge the gap between advanced research and sustainable market exploitation, supported by Union funding and aimed at strengthening technological sovereignty.

Detail

The Cloud and AI Development Act (CADA), as proposed in COM(2026) 502 final, introduces a structured mechanism to strengthen Europe's cloud and AI ecosystem. Central to this mechanism are the Cloud and AI Leadership Initiatives, which are designed to support research, development, and deployment activities across the Union. Article 6(2) of the proposal explicitly mandates that the operational objectives of these initiatives "shall be implemented through large-scale, cross-sectoral initiatives addressing major technological and industrial challenges of strategic relevance for the Union ('grand challenges'), as indicated in Annex I."

These grand challenges represent the strategic backbone of the proposal's supply-side measures. They are intended to coordinate public and private investment, guide the allocation of funding from Union programmes such as Horizon Europe and the Digital Europe Programme, and foster a competitive, resilient, and sovereign digital infrastructure. By focusing on "grand challenges," the proposal aims to move beyond fragmented national efforts, creating a unified European approach to overcoming critical bottlenecks in compute capacity, AI model development, and infrastructure sustainability.

Annex I of the CADA proposal details the following eight grand challenges, each with specific technical and strategic objectives:

1. Grand Challenge 1: Environmental sustainability, performance and security of the Union's data centres

This challenge addresses the urgent need to test and deploy technologies that surpass state-of-the-art energy and resource efficiency in data centres. It focuses on three key areas:

  • Lowering Power Usage Effectiveness (PUE): The goal is to achieve an average PUE of 1.15 across the Union. This involves enabling advanced cooling technologies, waste heat recovery, grid integration, and the development of next-generation energy-efficient technologies at operational scale.
  • Raising server utilisation rates: The initiative aims to raise average server utilisation rates towards 50%. This will be achieved by integrating AI-powered technologies for dynamic server utilisation management, runtime workload management, and scheduling to balance utilisation, energy costs, thermal constraints, and latency requirements.
  • Enhancing security and resilience: This involves integrating semiconductor and quantum technologies designed and manufactured in the Union to improve resistance to physical and cybersecurity threats, including targeted attacks, thereby strengthening the security of the data centre value chain.

2. Grand Challenge 2: Cloud stacks

This initiative focuses on building end-to-end hardware and software cloud stacks, including AI tools, infrastructure, services, and management layers. The primary objective is to bridge the Union's critical capacity gaps. Key activities include building AI servers powered by semiconductors and quantum technologies designed and manufactured in the Union for distributed and decentralised cloud and edge computing. Pilot programmes are expected to demonstrate the capabilities of these European open cloud stacks in strategically important sectors, fostering the development of secure, resilient, and performant alternatives to existing proprietary stacks.

3. Grand Challenge 3: Frontier AI

This challenge supports the development of the next generation of multimodal frontier AI models and systems, pioneering novel capabilities. The focus is on the architectural design and development of models that push the boundaries of current algorithmic capabilities. This includes achieving superior performance in advanced reasoning, cross-modal understanding, and agentic capabilities. The challenge also investigates novel approaches to model efficiency, cognitive modelling, and alternative computational structures. Potential applications span foundational science, scientific discovery, complex data interpretation, and the development of world models for improved reasoning, automated management simulation, and planning.

4. Grand Challenge 4: Physical AI

Physical AI refers to AI systems and models capable of perceiving the physical environment and executing complex actions within it. This grand challenge focuses on co-designing software and its underlying hardware architectures to deliver robust manipulation, navigation, and interaction capabilities with minimal human supervision. It combines frontier AI techniques with world models supporting physical reasoning. The potential applications include autonomous robots, industrial systems, and drones operating in dynamic real-world environments, addressing the need for robust AI in unstructured settings.

5. Grand Challenge 5: Industrial AI

This challenge aims to accelerate the development and deployment of European industrial AI across the Union's strategic sectors. It focuses on developing AI models and systems capable of serving high-value industrial applications that are adaptable to sector-specific use cases and enable secure deployment. Initiatives under this challenge will rely on specialised computing resources and testing facilities necessary to validate AI systems in real-world environments before supporting their large-scale deployment. Strategic sectors identified include automotive (for automated driving), manufacturing (for optimised production processes), healthcare, energy, agri-food, and defence.

6. Grand Challenge 6: Cooperative European Industrial Models

This initiative seeks to develop cooperative European industrial AI models and systems for strategic sectors by enabling collaboration at a European industrial scale without exposing commercially sensitive data between participants. The focus is on advanced confidentiality-preserving technologies. These mechanisms include federated and distributed training approaches where algorithms are brought to the data rather than data being transferred centrally, secure execution environments, encryption-based processing, anonymisation and pseudonymisation techniques, and protections against the extraction of commercially sensitive information from trained models. Strategic sectors benefiting from this include aerospace, pharmaceutics, cybersecurity, mobility, autonomous vehicles and drones, energy, and defence.

7. Grand Challenge 7: AI Agents Platform

This challenge focuses on developing a European AI agent orchestration framework, providing the essential middleware for the resilient and secure deployment of autonomous agents at scale. The focus is twofold: (i) exploring innovative technological paradigms that enable multiple AI agents to collaborate effectively, surpassing the capabilities of standalone systems while maintaining rigorous security standards; and (ii) creating resilient, cloud-based open platforms dedicated to the large-scale management of AI agents. Potential applications include healthcare (clinical decision support and research coordination), cybersecurity (threat detection and response), and foundational science.

8. Grand Challenge 8: Public Sector AI

This initiative targets the development of AI models and systems based on high-quality data from the public sector, targeting critical domains such as healthcare, public administration, law, and crisis management. The focus is on public service solutions expected to have a high positive impact on the most critical public services and shared across different levels of public sector organisations. A key target is to enable data sharing and frontier model development across national public services to increase the impact on the overall Union's public sector. This includes using privacy-preserving frameworks, such as federated learning and high-fidelity synthetic data generation, to train models without compromising the confidentiality of underlying datasets.

What this means for you

For technology leaders, CTOs, and innovation managers, the eight grand challenges in Annex I of the proposed CADA represent a definitive roadmap for where European public and private investment will be directed over the coming decade. These are not abstract policy statements but concrete criteria for accessing funding and participating in large-scale European projects.

Aligning R&D with Strategic Priorities If your organisation is developing technologies in energy efficiency, semiconductor design, or advanced AI models, aligning your roadmap with these grand challenges is critical. For instance, Grand Challenge 1 explicitly targets data centre operators and technology providers who can deliver PUE improvements below 1.15 or AI-driven workload management tools. Similarly, Grand Challenge 5 offers a clear pathway for industrial AI developers to access testing facilities and computing resources for validation in real-world environments.

The Shift to Open and Sovereign Stacks Grand Challenge 2 signals a decisive move away from proprietary, closed ecosystems towards open, interoperable, and sovereign cloud stacks. Architects and software developers should anticipate a market environment where public procurement and large-scale EU projects increasingly favour solutions built on European-designed hardware and open-source software. This presents a significant opportunity for SMEs and start-ups that can contribute modular components to these open stacks, particularly in areas like AI optimisation and middleware.

Collaboration Without Data Leakage Grand Challenge 6 addresses a major barrier to AI adoption in sensitive industries: the fear of data leakage. By promoting federated learning and secure multi-party computation, the proposal creates a framework for competitors in sectors like pharma and aerospace to collaborate on model training without sharing raw data. Organisations that can demonstrate robust confidentiality-preserving technologies will be well-positioned to lead or participate in these cooperative industrial models.

Preparing for Public Sector Demand Grand Challenge 8 indicates a surge in demand for AI solutions within the public sector, particularly in healthcare and crisis management. Providers of AI systems for these domains should prepare for a procurement environment that values privacy-preserving techniques and the ability to share models across borders. Early engagement with the emerging EuroCloud Federation and understanding the technical standards for public sector AI will be a competitive advantage.

Sovereignty as a Prerequisite While the grand challenges focus on innovation, they operate within the broader CADA sovereignty framework. Any solution developed under these challenges that aims to serve public sector bodies or critical infrastructure will likely need to comply with the Union assurance levels (Levels 1-4) defined in Article 16 and Annex II. This means that technical excellence must be paired with demonstrable operational autonomy, data localisation, and supply chain transparency.

Common misconceptions

Misconception 1: The grand challenges are legally binding mandates for all companies. The grand challenges outlined in Annex I are strategic priorities for the Cloud and AI Leadership Initiatives, not direct legal obligations for every cloud provider or AI developer. They do not impose a "grand challenge" compliance regime on the private sector. Instead, they guide the allocation of Union funding, the design of large-scale projects, and the strategic direction of the Cloud and AI Leadership Initiatives. However, compliance with the broader CADA sovereignty framework (starting from Article 16) will be mandatory for providers seeking to serve public sector bodies at higher assurance levels.

Misconception 2: Only large tech corporations can participate. While the grand challenges are "large-scale," the CADA proposal explicitly aims to support SMEs and start-ups. Article 33 encourages Member States to ensure that at least 25% of procurement for cloud computing services and AI systems is awarded to innovative SMEs. The grand challenges are designed to create ecosystems where SMEs can contribute specialised components, such as niche AI models for industrial applications (Grand Challenge 5) or privacy-preserving technologies (Grand Challenge 6). The "cross-sectoral" nature of these initiatives implies a need for diverse participants, not just dominant incumbents.

Misconception 3: "Sovereign" means only EU-owned hardware. Grand Challenge 2 focuses on "open cloud computing stacks" and technologies "designed and manufactured in the Union," but the sovereignty framework (Annex II) allows for nuanced assessments. It is not a blanket ban on non-EU technology but rather a set of criteria to ensure operational autonomy and data protection. The grand challenges aim to build competitive European alternatives, but the sovereignty framework provides a tiered approach to risk management, allowing for the use of third-country services under strict conditions and audits (e.g., Article 18).

Misconception 4: The grand challenges are purely theoretical research goals. The proposal explicitly links these challenges to "large-scale capacity" and "deployment" (Article 3). They are intended to bridge the gap between research and market exploitation. This means that successful projects will likely result in deployable infrastructure, certified software stacks, and operational AI systems, not just academic papers. The focus on "pilot lines," "test beds," and "real-world environments" underscores the practical, deployment-oriented nature of these challenges.

Related

This is general information about a draft EU regulation, not legal advice.