Security and Privacy in Artificial Intelligence (SenPAI)
Security and transparency of AI-based solutions
Machine Learning (ML) processes are applied in many different areas that demand the analysis of vast amounts of data. However, even ML-algorithms and trained neural networks are vulnerable, and an attack can result in the leakage of confidential personal data. One of ATHENE’s objectives is to improve the security of ML algorithms and systems, especially when considering the challenges in the area of data protection. Additionally, ATHENE explores the existing possibilities that ML technologies offer for the development of security solutions and adapts them for practical applications.
Prinicipal Investigators
Projects assigned to the research area Security and Privacy in Artificial Intelligence (SenPAI)
Detecting CSAM without the Need for CSAM Training Data (DeCNeC)
Forensic and OSINT Technology with Machine Learning (FROST)
LLM-Aided and Affected Authorship Verification/Attribution (LAVA)
Interactive Visual Cyber Analytics for Trust and Explainability in Artificial Intelligence for Sensitive Data (VCAXAI)
Protecting Privacy and Sensitive Information in Texts
Security in Large Language Models (SecLLM)
