Recent advances in artificial intelligence (AI), including generative models and socially interactive agents, are reshaping the design of interactive systems and raising new challenges for Human–Computer Interaction (HCI). While AI research has traditionally focused on improving performance and scalability, HCI emphasizes usability, transparency, and the broader social implications of technology, all aspects that imply the application of proper human-centered design methods. Bridging these perspectives is increasingly critical as AI moves from backend functionality to a central role in user interaction. This paper reports on the workshop Future Horizons in Human–AI Interaction: Joint Perspectives from HCI and AI, held at AVI 2026. It synthesizes key insights on emerging challenges and opportunities in designing human-centered AI systems, emphasizing the importance of integrating perspectives from HCI and AI, to advance the design of interactive AI systems that are both technically robust and aligned with human values.

Background A main challenge in many types of physical rehabilitation is patient adherence to recommended exercises. Vestibular rehabilitation is the most effective treatment for the symptoms of dizziness, vertigo, imbalance, and nausea caused by vestibular disorders, but adherence levels are particularly low as the rehabilitation program calls for many short exercise sets during the day, which can worsen symptoms and impair balance in the short term. Technological tools have the potential to increase adherence, but to date, there has been no comprehensive analysis, in the context of vestibular rehabilitation, of the specific needs from technology, of its limitations, and of concerns regarding its use. Objective The aim of the study is to identify the main features required from technology for vestibular rehabilitation, as perceived by patients with vestibular disorders and by vestibular physical therapists, using a socially assistive robot as a test case. We seek here to provide practical information for the development of future vestibular rehabilitation technologies which are based on human-computer interaction (HCI) and human-robot interaction (HRI). Methods We conducted a qualitative study with six focus groups (N = 39). Three groups of patients with vestibular disorders (N = 18) and three groups of physical therapists (N = 21) participated in this study. The participants answered structured questions on technologies for vestibular rehabilitation, watched a presentation of two videos of a socially assistive robot (SAR), and completed an online survey. Thematic analysis with a mixed deductive and inductive approach was used to analyze the data. Results Participants preferred phone applications or virtual/augmented reality platforms over an embodied robotic platform. They wanted technology to be adaptive to the different stages of rehabilitation, gamified, easy to use, safe, reliable, portable, and accessible remotely by the therapist. They reported that the technology should provide feedback on the quality and quantity of exercise performance and monitor these factors while considering the tolerability of the ensuing disruptive symptoms. Participants expected that using technology as part of the rehabilitation process would shorten exercise sessions and improve clinical outcomes compared to standard care. SARs for vestibular rehabilitation were perceived as useful mostly for children and patients with chronic vestibular disorders, and their potential use for rehabilitation raised concerns regarding safety, ethics, and technical complexity. Conclusions Although SARs can potentially be used to increase exercise adherence, a phone application appears to be a more suitable medium for this purpose, raising fewer notable concerns from users. We provide a summary of perceived advantages and disadvantages of technology for vestibular rehabilitation, as well as a set of specific requirements from it, which may inform the future development of specific supportive technologies. In addition, the focus group methodology employed demonstrates the importance of participatory design in the development of rehabilitation-supportive technology as participants were able to identify the likely inappropriateness of SARs for the specific case of vestibular rehabilitation.

Parkinson's disease (PD) is a neurodegenerative disorder that affects millions of people worldwide, and its incidence is expected to rise due to the aging of the global population. Individuals with PD present with adverse motor and non-motor symptoms that affect their well-being, quality of life, and relationships with their caregivers. It is thus important to explore new technological interventions, such as social assistive robots (SARs), to support those with PD. This study aimed to understand the needs, expectations, and attitudes of individuals with PD and their family members toward SARs. Six focus groups (n = 46) were held, including three comprised of individuals with PD and three that included family members. Qualitative thematic analysis of the information obtained from these focus groups was applied to the following main themes: (1) Key difficulties of people living with PD; (2) Ideas for assistive technologies for individuals with PD; (3) Perceived advantages of SARs; (4) Suggestions regarding SAR design, operation, and interactive functionality; (5) Concerns regarding the use of SARs; and (6) Attitudes towards SARs. Importantly, we found points of agreement and disagreement between the two population groups based on their perspectives regarding the effects of PD on individuals. This study provides a participatory-design foundation for the creation of new technologies to benefit individuals with PD.

To explore how socially assistive robots (SARs) may assist the specific needs of individuals with Parkinson's disease (IwPD) , we conducted three focus groups with 12 clinicians who treat IwPD. We present a thematic analysis of their perceptions of the needs of the patients, and their own expectations, perceived advantages, disadvantages and concerns regarding the use of SARs for IwPD. Clinicians were positive towards using SARs for IwPD, if used in the patient's home, for motor, communication, emotional, and cognitive needs, especially for practice and for help with activities of daily living. They were concerned that a SAR might be used to replace clinicians’ work, and stressed it should only augment the clinicians’ work. They thought a SAR may relieve some of the burden experienced by informal caregivers, and identified specific applications for SARs for PD. We asked 18 stakeholders (nine IwPD, nine family members) to rate their level of agreement with the clinicians’ statements. The greatest divergence between their views and those of the clinicians was on the topic of using a SAR as a companion, or as a feeding assistant, to which they objected. This work may be used as a basis for future studies designing SARs for IwPD.

Socially Assistive Robots (SARs) and immersive Virtual Reality (iVR) are interactive platforms that promote user engagement, which can motivate users to adhere to therapeutic frameworks. SARs use social presence to create affective relationships with users, leveraging the human tendency to be driven by social interactions. iVR uses spatial presence to provide an intense multisensory experience that submerges users in a virtual world. We adapted two such platforms – a SAR and an iVR – to deliver cognitive training (CT), by integrating established cognitive tasks in gamified environments that convey a strong sense of presence. Sixty-four participants underwent CT with both platforms. We tested: (1) their perception of both platforms; (2) whether they preferred one over the other in the short term; (3) their projected preferences for long-term training; and (4) whether their preferences correlated with personal characteristics. They preferred the virtual experience in the short term across age and gender. For long-term CT, there was equal projected preference for both platforms. It may be that a combination of social and spatial presence might yield engagement in long-term training.

When robotic systems are developed for individualized training during rehabilitation, providing effective feedback to the users is essential. We aimed to create a rule-based set of guidelines for the desired hierarchy, timing, content, and modality of feedback such a system should provide to users, such that they receive the necessary feedback, at the relevant time, in a way that enhances their performance, and does not encumber it. We conducted four focus groups with 20 stroke clinicians. The clinicians described the guiding principles they use when giving feedback to patients, and noted different output should be provided to the patient versus the clinician by the rehabilitation system. They delineated a hierarchy for providing feedback during the exercise set: success on the task is the primary goal, and feedback should be given on this aspect first. Once success is achieved, feedback should be given on the quality of the movement. Only when the task is successfully completed, with no compensatory movements (i.e., high quality), feedback should be given on movement speed. Using a follow-up survey and the member-checking approach, it was revealed that this hierarchical structure applies to early stages in the rehabilitation process, and that quality of movement becomes paramount as the rehabilitation process progresses. The clinicians expressed their desire to receive a full report on the patient's performance in each exercise session, which is more comprehensive than the feedback provided to the patient in real time. We conclude with a set of guidelines for developing automated feedback for patient populations.

Insights from social and cognitive neuroscience should inform the design of socially assistive robots for neurorehabilitation as novel roles emerge for them in human-human interactions.

We developed a novel gamified system for post-stroke long-term rehabilitation, using the humanoid robot Pepper (SoftBank, Aldebaran). Here, we present a participatory-design study with insights from both expert clinicians and from stroke patients who underwent a long-term intervention with the robot. We first present the results of a qualitative study with expert clinicians (n=12) on the compatibility of this system with the needs of post-stroke patients, and then the preliminary results of a long-term intervention study with post-stroke participants (n=4) in a rehabilitation facility. Both the clinicians and the patients found the robot and the gamified system engaging, motivating and meeting the needs of upper limb rehabilitation. The clinicians gave specific recommendations that may be applicable to a wide range of technologies for post-stroke rehabilitation.