Spectrum Alliance is a closed security initiative of the eMěsta project. Our mission is to transform cyber defence of the public sector from a reactive model to a proactive one. Using advanced Spectrum AI technology, we continuously verify the integrity of critical software and infrastructure that form the backbone of the modern state.
We do not wait for an incident and eliminate risk before it arises. We regularly verify the integrity of critical software and infrastructure.
A fleet of advanced AI agents identifies vulnerabilities with an efficiency that surpasses the best human experts.
Each partner brings strength in its own domain. Their combination creates a defence capacity that no single player can achieve alone.
The security environment has fundamentally changed. The impulse for launching this alliance was the Glasswing project by Anthropic. The research has clearly shown that modern AI models already possess the ability to identify and exploit "zero-day" vulnerabilities with an efficiency that surpasses human experts.
European public administration faces a critical challenge: systems that manage the data of millions of citizens are now exposed to the threat of automated, AI-driven attacks. While attackers are already adopting these technologies, the state's defence often remains stuck in outdated methods. We feel an obligation to close this gap.
These are not hypotheses or distant risks. During the initial audits and scans conducted by Spectrum AI, in the software on which the daily functioning of European public administration relies, we have uncovered high-severity vulnerabilities.
These weaknesses are not just a technical problem — if exploited, they open the way to direct intrusion into sensitive registers, data manipulation or the complete paralysis of key state services. In other words: systems that millions of people rely on can fail in a single moment.
The severity of the situation is so high that we cannot afford to publish technical details or the specific names of the systems affected. In accordance with responsible disclosure principles, we are now in direct dialogue with the relevant institutions and developers — because the priority is to prevent these weaknesses from being exploited before they are fixed.
The rise of autonomous cyber threats is one of the greatest tests that digital society is facing. No institution can manage this challenge in isolation. Spectrum Alliance is open to cooperation with:
Companies supplying software for public administration, whose systems run cities and state authorities.
Private technology leaders and innovators who develop cutting-edge security innovations for AI, cloud and network infrastructure.
Authorities and local governments that want to proactively protect both their own data and citizens' data using modern technologies.
Independent and corporate experts who want to actively participate in shaping security standards.
Only by joining forces with the best in the field can we fully harness the potential of AI to protect public administration.