CybersecBRAIN

Experimental development of a novel collective intelligence platform for cybersecurity.

The CybersecBRAIN project addresses the experimental development of the application of algorithms, techniques and technologies based on artificial intelligence for security management.

CybersecBRAIN focuses on two areas of innovation through the application of solutions based on artificial intelligence: automatic learning in certain tasks of a SOC (Security Operations Center) and the application of collective intelligence through federated learning algorithms or swarm learning.

Currently, the management of SOC activity is a question that MSSPs (Managed Security Service Providers, systems that can collect cybersecurity data from different organizations) handle in a non-automated way through manual processes, although some progress is being made through the integration of functionalities based on artificial intelligence.

When they reach a high volume of activity, the effectiveness of the SOC is reduced exponentially, also affecting their effectiveness and the quality of their services, operating on non-scalable models and their businesses are on the brink of profitability in the face of high prices and scarce resources, hence the need to integrate technologies that increase the degree of automation and autonomy, reducing response time.

Eurecat’s IT&OT Security Unit participates in the CybersecBRAIN project by researching and experimentally validating distributed learning techniques and algorithms, such as federated learning and swarm learning, as well as the collaboration between both approaches through techniques such as stacking, bagging and boosting, among others, with the aim of solving the existing limitations in collaborative environments.