DATAIA Days « Safety & AI »
There are different levels of security and different degrees of responsibility, for which we are faced with compromises or alternative solutions. These choices can only be analysed if one considers both the theoretical and practical challenges of the engineering problem for AI safety. This view should cover a wide range of AI paradigms, including systems that are specific to a particular application, those that are more general and that may result in potential unanticipated risks. We must also bridge the gap between the short and long term, idealistic and pragmatic, operational and political, and industry and academia, to truly build, evaluate, deploy, operate and maintain truly secure AI-based systems.
« Safety & AI should not be an option, but a design principle. »
The DATAIA Days « Safety & AI » will provide a forum for thematic presentations and in-depth discussions on safe AI architectures, artificial intelligence security, safe human-machine interaction and security considerations in automated decision systems, in a way that makes AI-based systems more reliable. These workshops aim to bring together experts, researchers and practitioners from diverse communities, such as artificial intelligence, security engineering, ethics, standardization and certification, robotics, cyberphysics, safety-critical systems and communities of application such as automotive, health, manufacturing, agriculture, aerospace, critical infrastructure and trade.