Goal
Increasing IT security for business and IT processes in critical infrastructure.
Solution
AI-based security system for automated real-time anomaly detection.
Result
The system also recognizes previously unknown attack patterns and relies on the combination of human and artificial intelligence.
Challenge
In times of increasingly specific threats and the increasing shift of IT services to the cloud, complex IT landscapes are a potential target for attack – especially in critical infrastructure. In the increasingly digital energy sector, attacks on IT infrastructure can have massive social and economic consequences. Improving IT security is therefore essential.
Goal
To avoid power outages and critical system conditions, IT attacks must be detected as early as possible and defensive measures initiated. The central building block of IT security should be the secure and robust real-time detection of anomalies using artificial intelligence.
Solution
This development of efficient AI-based IT security measures is precisely the focus of the PROTECT research project, which partners Fraunhofer IOSB-AST, EAM Netz GmbH, Zittau-Görlitz University of Applied Sciences, and eoda.
The key to increasing IT security is a holistic approach to process data, such as measured values or control commands, and network data from IT systems. Together with Fraunhofer IOSB-AST, eoda is developing automated real-time anomaly and attack detection – including the identification of attack type and cause.
Unlike conventional security systems, this AI-based security system also detects previously unknown attack patterns, such as zero-day attacks.
The high performance of AI methods allows them to learn network behavior even with extensive protocols and services and use this knowledge for anomaly detection.
To further increase the security of the system, eoda relies on an Explainable AI approach and thus on intuitively understandable analysis models. This allows experts to understand and verify the algorithms, and to evaluate the identified anomalies. This allows the actual risk potential to be assessed even more accurately.
To combine human and artificial intelligence and thus continuously improve the quality of the algorithms, eoda relies on a specially developed analytics platform.
Result
With the self-learning AI firewall, energy suppliers will have an IT security system at their disposal in the future that can increase and ensure long-term IT security in business and process IT, both independently and under the guidance of human experts.
In addition, the automated detection and processing of IT attacks helps relieve the responsible employees of routine tasks and creates more time to address critical anomalies.
Get started now:
We look forward to exchanging ideas with you.
Your expert on Data-Science-Projects:
Lutz Mastmayer
projects@eoda.de
Tel. +49 561 87948-370
