Atos Initiates Research Program on AI Security
Atos, in collaboration with German research institutes, has announced the launch of a program aimed at securing artificial intelligence against manipulations.
Establishment of the 'Forensics of Intelligent Systems' Program
Atos has signed a contract with the Agency for Cybersecurity Innovation to establish the 'Forensics of Intelligent Systems' program. This project, located in Halle, Germany, is conducted in collaboration with the Fraunhofer Institute for Applied and Integrated Security, the Fraunhofer Institute for Intelligent Analysis and Information Systems, the Institute for Internet Security at the University of Westphalia, and the University of Cologne. The goal is to develop methods and prototypes to detect manipulations of AI systems in a legally reliable manner.
Simulation Environment and Forensic Software Testing
The program includes the creation of a simulation environment in a research laboratory to test forensic software methods. AI models will be subjected to simulated attacks to identify legally exploitable traces and to adjust algorithms for early detection. According to Boris Hecker from Atos, this initiative aims to provide security authorities with advanced tools for accurate forensic analysis of AI systems. Prof. Dr. Christian Hummert from Cyberagentur emphasizes the importance of making intelligent systems legally secure and trustworthy.
Enhancing Digital Security and Trust in AI Systems
The 'Forensics of Intelligent Systems' project is part of Atos' strategy to enhance digital security and trust in AI systems. Atos focuses on transparency and European sovereignty with a portfolio that includes solutions for securing data and cloud infrastructures. This program meets the expectations for reliability and economic security, while strengthening the integrity of systems against technical uncertainties in the field of artificial intelligence.
