BRACE-LLM - Blue- and Red-Agent-based Cyber Security with Large Language Models

The project Blue- and Red-Agent-based Cyber Security with Large Language Models (BRACE-LLM) aims to use generative AI to support SMEs preventively in the configuration and maintenance of IT infrastructure as well as in the (proactive) detection of attacks and selection of countermeasures.

BRACE-LLM is a joint research endevor between imbus AG, Hochschule für angewandte Wissenschaften Coburg and Julius-Maximilians-Universität Würzburg.

The objectives of the project can be summarized as follows:

1. Integration of domain knowledge to analyze security vulnerabilities and configuration errors
2. Development of methods for explainable attack detection
3. (Partially) automated auditing based on red- and blue-teaming AI agents
4. Development of a dialog-based assistance system