Artificial intelligence (AI) and machine learning (ML) are increasingly integrated into the functioning of physical and digital products, creating unprecedented opportunities for interaction and functionality. However, there is a challenge for designers to ideate within this creative landscape, balancing the possibilities of technology with human interactional concerns. We investigate techniques for exploring and reflecting on the interactional affordances, the unique relational possibilities, and the wider social implications of AI systems. We introduced into an interaction design course (n = 100) nine ‘AI exercises’ that draw on more than human design, responsible AI, and speculative enactment to create experiential engagements around AI interaction design. We find that exercises around metaphors and enactments make questions of training and learning, privacy and consent, autonomy and agency more tangible, and thereby help students be more reflective and responsible on how to design with AI and its complex properties in both their design process and outcomes.

Purpose: The purpose of this paper is to conduct research with children, who have different abilities from adults, in terms of language understanding and level of attention, is a challenging task, especially concerning novel interactive systems such as social robots. Consequently, self-reporting methods are often replaced or supplemented by observational methods that are usually carried out taking advantage of video recordings. However, some limitations make this approach challenging for studies conducted with groups of children in real-world environments, whose relevance is being addressed more and more frequently in human-robot interaction (HRI) research. Thus, there is a growing need for rigorous observation approaches in unstructured test environments. Design/methodology/approach: This paper presents an alternative analysis approach, in relation to an experimental child-robot interaction (CRI) application, which was developed at the Academy of Arts and Design, Tsinghua University, China. The proposed methodology is based on the analysis of video recordings of in-wild activities of children with a robot. The methodology has the aim of providing a framework to facilitate knowledge identification and structuring. It was implemented for experiment evaluation and validation purposes and to propose a reference structure for the organization of new experiments and the stimulation of new ideas and activities in the design process. Findings: This methodology provides a logical structure, which can be used to identify the effectiveness or limits of design choices, pertaining to such aspects as the morphology or movement of robots or the choice of their specific role in education, all of which play crucial roles in the design process and could be improved to achieve better results. This structured identification is a practical implication for the design process, above all when it is oriented toward social robots and their interaction with children or elderly senile people. In this case, the outcomes were the identification of important elements of an experiment (psychological profiles of the involved children and possible problems or risks) and their impact on the design process. Originality/value: The methodological approach, which structures and uses cognitive maps to elaborate multicriteria evaluation models, is not new to the operations research field (where it is defined as a multimethodology application of Soft OR), but it has not yet been applied in the field of HRI studies, to analyze children’s perception of a robot and to identify the factors that can affect a good CRI or to structure knowledge that can be shared to guide the design process of robots for the experience of children playing.

Artificial intelligence (AI) and machine learning (ML) are increasingly integrated into the functioning of physical and digital products, creating unprecedented opportunities for interaction and functionality. However, there is a challenge for designers to ideate within this creative landscape, balancing the possibilities of technology with human interactional concerns. We investigate techniques for exploring and reflecting on the interactional affordances, the unique relational possibilities, and the wider social implications of AI systems. We introduced into an interaction design course (n=100) nine 'AI exercises' that draw on more than human design, responsible AI, and speculative enactment to create experiential engagements around AI interaction design. We find that exercises around metaphors and enactments make questions of training and learning, privacy and consent, autonomy and agency more tangible, and thereby help students be more reflective and responsible on how to design with AI and its complex properties in both their design process and outcomes.