„Identity becomes the new perimeter of security“

As Europe builds a Multi-Provider Cloud-Edge Continuum, identity becomes the new perimeter of security

Silvio Ranise of Fondazione Bruno Kessler on AI at the edge, Lab8ra and post-quantum readiness

October 2025

In Conversation with Silvio Ranise
Fondazione Bruno Kessler

Silvio Ranise is Director of the Center for Cybersecurity at Fondazione Bruno Kessler (FBK) and Professor at the University of Trento. In this conversation, he shares his perspective on how digital identity, AI at the edge, and the Lab8ra testbed are shaping Europe’s path toward a resilient and interoperable Multi-Provider Cloud-Edge Continuum.

What is FBK’s role within the 8ra Initiative?

“Fondazione Bruno Kessler is a multidisciplinary research institute with twelve centres, from ICT and sensors to quantum technologies and the humanities. Within the 8ra Initiative, three FBK centres work together within our project Cybersecurity & AI at the Edge: Cybersecurity, Digital Societies and Augmented Intelligence. The goal is to advance AI techniques and ensure they can be deployed in ways that are secure, trustworthy and privacy-preserving.”

Why does Europe need a Multi-Provider Cloud-Edge Continuum?

“Because choice builds resilience and proximity builds performance and trust. A Multi-Provider Cloud-Edge Continuum makes it possible to place workloads where they perform best and comply with regulation, close to where data is created, while keeping the freedom to adapt as needs evolve. If everything runs on just a few external infrastructures, control shifts away from Europe. When many providers interoperate, we gain flexibility and sovereignty.

This is also about readiness. We know from recent debates on cross-border data access how important it is to guarantee data locality and optionality. We have to prepare now, so that in a few years we can switch providers, move workloads and meet new requirements. Having more – and decentralising more – is key.

That is why, together with our partners, we contributed to the launch of Lab8ra. It is a testbed where we can validate our ideas in practice – from connectivity to edge nodes to the 8ra software stack – and make sure the architecture works under real conditions.”

What is Lab8ra and what are you currently working on?

“Lab8ra is a hardware-and-software testbed, coordinated with virt8ra, that allows us to test and validate what we are building. Europe’s “last mile” is the edge and we need to expand that layer to make the continuum fully effective. Together with partners such as TIM we are setting up a hardware infrastructure to support the Virt8ra stack. The infrastructure is complex, but progress is good. We are already gaining momentum.

What’s also important – Lab8ra is not a closed initiative. From the beginning, we have been open to other partners. The Italian government has organised events to bring in companies beyond the original consortium and I have personally been contacted by several interested organisations.

We also align with other European efforts such as the Common European Data Spaces and the NIS2 Directive. For us, it is important that knowledge flows in both directions – what we learn in IPCEI-CIS and the 8ra Initiative can strengthen other initiatives and insights from outside projects help us refine our work inside 8ra.”

Security is one of the cornerstones of the 8ra Initiative. How do you approach it?

“In a multi-provider world the old perimeter is gone – identity becomes the new perimeter. Every user, every device, every service and even every workload must prove who or what it is. Trusted digital identities are the foundation and on top of them we build authorisation and accountability so that we always know who did what, when and on which resource.

We also rely on cryptographic mechanisms and new approaches to digital identity management. This combination allows us to ensure not only privacy but also resilience – the ability to recover quickly from attacks and keep services running.”

What role can AI play in this architecture?

“AI can be extremely valuable – but it has to be applied in the right way. By running AI at the edge, we can detect anomalies earlier and respond faster, because processing happens close to where data is generated. Federated learning is crucial here, because it allows local learning while still sharing results more broadly.

We are not aiming for massive, general-purpose models. What we need are domain-specific, verifiable AI systems that are reliable and accurate. At the edge, resources are limited, so models have to be pruned to consume less energy and memory while maintaining accuracy. This approach also helps meet the objectives of the European Green Deal.

Another aspect is that AI is increasingly used to manage software itself – to code, deploy and configure. That gives us efficiency, but it also means we must strengthen governance and verification. We cannot simply hand over control. We need to understand the mechanisms and make sure automation improves resilience instead of creating new risks.”

How are you preparing for the post-quantum era?

“We have to act now. At some point, quantum computers will make the cryptographic mechanisms we use today obsolete. That creates a harvest-now, decrypt-later risk: adversaries could collect sensitive data today and decrypt it in the future.

That is why we are planning the post-quantum transition. We need strategies to replace existing cryptographic infrastructure with new, quantum-safe mechanisms, and we need to start testing them. It is a complex effort, but essential if we want to protect long-lived sensitive data.”

Which use cases best illustrate the value of the Multi-Provider Cloud-Edge Continuum?

“Smart cities and smart industry are particularly interesting. They are complex environments that push the continuum to its limits – with heavy demands on latency, capacity and above all, privacy.

Take smart mobility in a city. Geolocation data is highly sensitive because it can identify individuals very precisely. If you do not process it in a privacy-preserving way, you run into problems. The same goes for industrial IoT devices, which can track workers’ movements. We need strong anonymisation and privacy safeguards built into these systems from the start.

If we get it right, the benefits are significant: safer workplaces, more efficient production and urban services that improve people’s daily lives – from reducing commuting times to enabling more flexible working.”

If you could fine-tune one element of Europe’s approach, what would it be?

“We need a better balance between foundational and applied research. Europe is very strong in the “first mile” – basic science. Where we sometimes struggle is in the “last mile” – turning those great ideas into concrete infrastructure and services that benefit people and businesses.

But it is not about shifting everything to application and forgetting the fundamentals. Without strong foundational research, there is no innovation further down the line. We have to balance the two and initiatives like 8ra are important because they connect them.”

Silvio Ranise directs the Cybersecurity Centre at Fondazione Bruno Kessler (FBK) and is a Professor of Computer Science at the University of Trento. In 8ra, FBK runs the project “Cybersecurity & AI at the Edge” and also contributes to the cross-workstream security, combining AI at the edge with identity-centred, privacy-enhancing security to strengthen the Multi-Provider Cloud-Edge Continuum; key deliverables include Decepto (Kubernetes-based cyber deception), FLAD (federated DDoS detection), LUCID (deep-learning intrusion detection) and TLSAssistant (securing container-to-container/cloud-native traffic).