Future power systems are often envisioned to be decentrally organized. I will start by sharing some high-level thoughts on the current status and recent trends with respect to decentral energy systems. I will also discuss possible motivations for decentral organization, proposing that resilience might be a key driver for decentral intelligence. This initial discussion will motivate our work on a distributed scheme for secondary frequency control, that has no single point of failure. After introducing this scheme, I will present a first analysis of the proposed algorithm’s robustness against malicious attacks. A comparison against the classic central secondary control scheme that is used in many power system worldwide is made. I finally present an overview over the topics of my new group at TU Darmstadt, where resilience is a major topic.