European Ecosystem of Trustworthy Generative AI Agents for Cybersecurity and Regulatory Compliance

AIGENTIC aims to develop a European ecosystem of collaborating Generative AI Agents designed to support cyber threat detection and response, operational resilience, and regulatory compliance. The project combines automation with human oversight, transparency, and data protection, strengthening the secure adoption and use of artificial intelligence by businesses and public-sector organisations.

Detailed Project Description

Modern cyberattacks combine multiple techniques, including social engineering, supply chain compromise, and the malicious use of artificial intelligence. AIGENTIC seeks to address these challenges through a modular and interoperable framework for the development, coordination, and secure operation of AI agents.

The main objectives of the project are:

• Development of a trustworthy five-layer technological framework, connecting computing infrastructure and secure data flows with AI agents, task orchestration, and governance mechanisms.
• Support for cybersecurity operations, through threat intelligence analysis, incident correlation, product security assessment, and assistance in response and recovery activities.
• Security and compliance by design, including data and model provenance tracking, explainable decision-making, and documentation supporting regulatory requirements such as NIS2, CRA, DORA, the AI Act, and GDPR.
• Maintaining human oversight, through controls and approval mechanisms for critical actions, as well as safe rollback and recovery mechanisms in the event of failures.
• Strengthening European digital sovereignty, by leveraging computing resources under European governance and secure data-sharing practices.

The project will be evaluated through nine use cases: six sector-specific use cases covering the automotive, railway, telecommunications, financial services, maritime, and smart cities sectors, as well as three cross-cutting use cases. The objective is to validate the solutions at Technology Readiness Level (TRL) 5, with selected applications aiming to reach TRL 6.

Type and Scope of Work to Be Provided

DOTSOFT S.A.’s expected contribution focuses on requirements definition, secure automation, and the evaluation of solutions within a smart city environment. More specifically, it includes:

• Use Case Requirements and Specifications (WP2): Leadership of the work package and of the activity related to use case definition, including the identification of involved systems, data, operational needs, and initial evaluation conditions.
• AI Agent Orchestration and Secure Operation (WP5): Participation in the development of executable workflows, simulation and testing environments, secure deployment procedures, and mechanisms for human oversight and recovery.
• Smart Cities Pilot (WP7): Provision of the use case through the SmartisCity/NOONiot platforms, leveraging data from mobility, parking, environmental monitoring, lighting, and energy services.
• Citizen Data Protection and Information Integrity: Testing of agents for the enforcement of privacy rules, detection of suspicious sensor data, incident correlation, and automated generation of compliance documentation.
• Dissemination and Exploitation of Results (WP8): Participation in communication, standardisation, and exploitation activities, supporting the future adoption of the solutions by municipalities and other organisations.

Through its participation in AIGENTIC, DOTSOFT aims to strengthen the security of digital city services and the protection of citizens’ data by leveraging trustworthy artificial intelligence with controlled automation.

.