The Smart Bench is a speculative design aiming to create critical awareness about the social implications of Data-driven Nudging in the Smart City. The project originated from a critical perspective on Smart City as a leading urban development paradigm. Many cities labelling themselves as smart cities competitively implemented advanced technologies like artificial intelligence (AI), pervasive sensor networks, or internet of things (IoT) for infrastructural or economic purposes (Kitchin et al, 2015; Townsend, 2013). This enabled ubiquitous extraction and monitoring of urban behavioral data, facilitating more efficient, resilient, and frictionless urban governance.Data-driven nudging emerged as vast aggregate of citizen behavioral insights became available to be implemented into the concept of ‘nudge’, a libertarian paternalistic strategy of behavior change (Ranchordás, 2019; Thaler & Sunstein, 2008). Citizen behaviors can be steered more elaborately through targeted and dynamic adjustment of choice architecture enabled by development of analytical algorithms. Pervasively embedded network of urban sensors has made this new type of nudge more invisible, thus more powerful. Hence, data-driven nudging is raising ethical concerns in terms of behavioral manipulation, ubiquitous surveillance, and urban depoliticisation (Kitchin et al, 2015; Lanzing, 2018; Ranchordás, 2019; Void et al, 2018; Yeung, 2016). Making matters worse, citizens’s level of awareness not only on data-driven nudging itself but also on smart city in general turned out to be significantly lacking (Government Europa, 2018; Jameson et al, 2019). The project addresses the lack of awareness of the presence of data-driven nudging and its social implications by designing a speculative experience of Smart Bench. The Smart Bench is an intelligent bench which is designed to nudge citizen behaviors according to a predetermined algorithm. Because smart technologies tend to reproduce pre-existing prevailing biases, the purpose of the design is to make them experiential to the general public. Participants are invited to try out the bench and a special interface where they could take a look at the software principles behind it and have a unique opportunity to adjust the nudging algorithm for others. As they proceed along the experience, participants can experience the discriminatory implications of data-driven nudging and openly discuss its application for urban governance. The evaluation results confirmed a fruitful capacity of speculative design in terms of materialising an abstract unfamiliar concept and evoking critical awareness on its potential implications. Most of all, the design provided participants with a space for open discussions about the necessity of data-driven nudging for urban governance and further reflect on pre-existing discrimination in our current society.