Cyberagentur Starts Major Research Program On Secure Machine Learning

The Agentur für Innovation in der Cybersicherheit GmbH (Cyberagentur) has launched a five-year research project to increase the reliability and security of various AI approaches. The project, called “Robust and Secure Machine Learning” (RSML), aims to advance different innovation approaches to make machine intelligence less vulnerable. Examples of vulnerabilities include the dependence on training data and the intransparency of trained models, which can lead to unexpected and potentially catastrophic results. The project seeks to stimulate research and development of new approaches to increase the robustness and security of AI in various domains.