Digital sovereignty means being able to demonstrate, using verifiable criteria, who controls a technological infrastructure, where data is processed, and under which jurisdiction. The Tech Sovereignty Catalogue is the European register that applies these criteria to market-ready digital solutions.
We are proud to announce that our two solutions, Radicalbit Platform and Radicalbit AI Gateway, have officially joined the Tech Sovereignty Catalogue, an initiative promoted by the European DIGITAL SME Alliance.
This represents a new milestone that consolidates Fortitude Group’s vision of building a resilient, secure, and independent European tech stack.
What Is the Tech Sovereignty Catalogue and Who Promotes It
The Tech Sovereignty Catalogue is promoted by the European DIGITAL SME Alliance, the organization representing over 45,000 digital small and medium enterprises across Europe, and is developed in line with the EuroStack framework. Its goal is to map, evaluate, and provide visibility to technologies developed and controlled within Europe that offer a concrete alternative to major non-EU players across five strategic pillars: Cloud, Artificial Intelligence, Cybersecurity, Connectivity, and Data Infrastructure. The Catalogue was created alongside two qualification labels, “Software made in Europe” and “Software hosted in Europe”, designed to offer technology buyers a quick way to distinguish what is genuinely European from what is European in legal headquarters only.
For AI strategy and infrastructure leads, the Catalogue is not a promotional showcase: it is a due diligence tool. It connects European technology developers with those responsible for selecting solutions, whether an enterprise decision-maker, a public administration procurement office, or a policymaker working on standards and public procurement. Being listed in the Catalogue means having passed an external verification process against criteria that a company would otherwise have to assess independently, vendor by vendor.
What Being “Sovereign by Design” Means
Inclusion in the Catalogue is not based on generic self-declarations. A solution must demonstrate commercial maturity at TRL 9 (Technology Readiness Level 9), meaning it is a product already proven at scale and ready for production deployment. It must have its headquarters and ownership within the European Union or EFTA countries, a condition that excludes exposure to extraterritorial regulations like the US CLOUD Act.
It must guarantee that data processing and storage take place within European borders under the exclusive jurisdiction of the GDPR and related EU laws which differs from simply hosting a data center in Europe while the parent company remains subject to foreign legislation. Finally, it must adopt open and interoperable architectures, allowing workloads to be moved from one cloud to another or returned on-premise without rewriting integrations from scratch.
These requirements shift the conversation from declared sovereignty to verifiable sovereignty. They do not measure a vendor’s intentions; they measure concrete operational conditions: where data resides, who has legal access to it, and what happens if the company decides to change course. This distinction is what makes the Catalogue a trusted benchmark for anyone evaluating enterprise AI vendors.
Why Digital Sovereignty Has Become an Enterprise Purchasing Criterion
Until a few years ago, digital sovereignty remained confined within regulatory debates, a political principle rather than an operational one. The mass adoption of Generative AI models has turned it into a measurable risk. The US CLOUD Act obliges American companies to provide US authorities with access to the data they manage, regardless of where that data is physically stored: a European company sending data to an American API exposes that data to US jurisdiction, even if the physical data center is located in Europe. The “sovereign” versions offered by major hyperscalers mitigate the issue, but they do not solve it at its root, because the jurisdiction over the parent company remains American.
In addition, three regulatory frameworks directly impact the tech supply chain of European enterprises: the AI Act, with operational obligations for GPAI systems taking effect in August 2025; NIS2, which extends cybersecurity accountability across the entire supply chain, including ICT suppliers; and DORA, which imposes strict third-party provider resilience requirements on the financial sector.
In November 2025, all EU member states signed the Declaration for European Digital Sovereignty, a political act explicitly recognizing technological dependence as a strategic risk rather than just an ethical concern. The practical outcome is that IT procurement departments no longer merely ask if a vendor is GDPR-compliant. They ask where training and inference data reside, who has access to audit logs, and what happens if the vendor changes policies, ownership, or jurisdiction.
The Tech Sovereignty Catalogue answers these questions through a publicly available, upfront verification process.
Radicalbit Platform in the Catalogue: Sovereign MLOps and Observability
Radicalbit Platform unifies Real-Time Machine Learning, MLOps, and AI Observability into a single environment to design, deploy, monitor, and scale AI and ML models in production.
The entry published in the Catalogue recognizes its sovereign by design nature: development, maintenance, and intellectual property remain under European jurisdiction. Its architecture supports containerized deployment on European sovereign clouds, hybrid systems, or on-premise data centers, keeping data processing, training, and inference entirely within the EU/EFTA area.
The technical highlight emphasized by the Catalogue is its approach to the “black-box AI” challenge: drift detection, data quality monitoring, and observability are embedded directly into production workflows rather than added as an external layer after the fact. For financial institutions, manufacturing companies, and public administrations, this means being able to demonstrate, not just declare, algorithmic fairness and compliance with European standards and the AI Act. In an audit, a drift detection log carries far more weight than a well-written policy document.
Radicalbit AI Gateway in the Catalogue: The Control Layer for Models
Radicalbit AI Gateway addresses a different and complementary challenge: the governance of traffic directed to Large Language Models (LLMs), regardless of the provider. The solution positions itself as an orchestration layer between enterprise applications and AI models, managing routing, rate limiting, semantic caching, and cost optimization across multiple providers simultaneously, without the application needing to know which specific model is handling the request.
In its Catalogue entry, the AI Gateway is described as a sovereign firewall for enterprise data assets: it applies real-time anonymization, automatic PII data masking, and policy-based access controls before any payload leaves the corporate perimeter. Audit logs, prompt metrics, and compliance databases remain strictly within EU/EFTA borders, even when the enterprise uses external models like GPT or Claude alongside European open-source models. It represents a key missing piece in the European sovereign AI stack, designed specifically for regulated sectors such as healthcare, defense, and public administration, where Generative AI adoption cannot come at the expense of strategic control over information flows.
Two Solutions, One Sovereign Stack
The two solutions operate at opposite ends of the same AI pipeline: while the Platform governs the lifecycle of production ML models, from training to continuous monitoring, the Gateway manages access and traffic to LLM models, including those hosted by non-EU providers.
Having both solutions listed in the Catalogue allows our team to cover the entire scope of the enterprise AI stack through two independent verifications, spanning from model development and observability to daily access control.
For data and infrastructure leads, this removes a step from the vendor selection checklist, as verifying European headquarters, data jurisdiction, and lock-in avoidance no longer needs to be conducted internally since it has already been performed publicly by a third party. For procurement teams in public administration and companies subject to NIS2 or DORA, this significantly reduces due diligence times for critical suppliers, while for policymakers and standardization experts, it adds two verified benchmarks to today’s map of available European alternatives.
Digital Sovereignty Is Built One Layer at a Time
No European enterprise achieves digital sovereignty through a single decision, such as migrating everything on-premise or abruptly abandoning American hyperscalers. Digital sovereignty is built by systematically reducing unnecessary dependencies while expanding control, observability, and portability — one stack layer at a time.
The inclusion of Radicalbit Platform and Radicalbit AI Gateway in the Tech Sovereignty Catalogue provides external validation of this approach: two independent verifications covering headquarters, jurisdiction, maturity, and interoperability, accessible to anyone seeking an AI provider without giving up control over their infrastructure.
In this landscape, Radicalbit Platform and Radicalbit AI Gateway deliver the architecture necessary to put these principles into practice. From continuous performance oversight to data flow governance toward Large Language Models, Radicalbit solutions transform AI management into a transparent and measurable process.
Request a personalized demo to evaluate these solutions on concrete use cases and see firsthand how to integrate them into your ecosystem.
