Large Language Models (LLMs) have rapidly become ubiquitous, demonstrating remarkable proficiency in generating text and responding to prompts. Despite their potential, concerns about privacy and trust persist, yet methods for engaging end-users in designing privacy-conscious AI remain limited. This paper presents a replicable co-design methodology for engaging end-users in privacy-conscious LLM design. Through workshops with 36 participants across three speculative scenarios—mental health applications, travel assistants, and workplace assistants—we demonstrate how combining speculative scenarios with established frameworks (Grice’s Maxims, Schaub’s privacy design space) and trained facilitators enables meaningful participation from users without technical expertise. The co-designed dialogues reveal that users desire dynamic, context-sensitive privacy communication that leverages LLMs’ conversational capabilities. We provide practitioners with a replicable methodology comprising: (1) scenario creation methods, (2) framework scaffolding approaches, and (3) facilitator training guidance, alongside design implications for privacy-conscious conversational AI.

Progressive digitisation and adoption of artificial intelligence (AI) are reshaping welfare services in ways that risk compounding inequalities for disabled people. Globally, many governments present these reforms as beneficial–streamlining processes, reducing costs and eliminating delays. Yet digitisation and automation of welfare decision-making can deepen exclusion and erode human accountability. In response, this paper foregrounds the lived experiences of people with the communication disability aphasia in navigating digitised welfare and their perspectives on AI-automated futures. We report findings from a four-stage participatory design study involving eight workshops with 42 recruited co-designers. Reflexive thematic analysis identified five challenges: the cost of performing disability, geographies of inequity, navigating digital bureaucracy, the accessibility paradox and hostile design. Co-designers voiced concerns about AI-automation but envisioned inclusive future alternatives: AI dialogues that are patient, multimodal and supportive; welfare systems that are compassionate, transparent and retain human recourse; and infrastructures that are open, publicly governed and truthful.

Widespread abandonment continues to hinder the long-term adoption of augmentative and alternative communication (AAC) devices among communities living with aphasia. Challenges with AAC are complex - ranging from poor personalization and stigma to prohibitive costs. Critics argue that many AAC tools reflect top-down, medicalized models shaped by ableist assumptions and commercial agendas. In response, this paper presents bottom-up perspectives from a four-month qualitative study of Zuzenna’s DIY AAC: a self-made communication diary. Far from a prescribed device, her diary is a bespoke, organically evolving tool that empowers her daily life and recovery after stroke. We offer a re-imagining of AAC design grounded in lived experience, highlighting how future tools might: (a) foster genuine communion through share-ability, (b) uphold agency by empowering self-expression, (c) value personal labour and support growth. Zuzenna's diary reveals that AAC does not have to be techno-centric or expensive to be powerful -- it just needs to be personal.

Cerebral palsy (CP) affects millions globally, many of whom rely on technology to support daily living and communication.Yet, there is a notable absence of formal autoethnographic research exploring the lived experiences of people with CP, particularly those who use augmentative and alternative communication (AAC) devices.This paper presents a longitudinal autoethnographic study of Tianchi, an international PhD student with CP as he navigated the final stages of his doctoral journey.Using a structured, questionnairebased diary method, we captured daily reflections on his everyday life, use of assistive technologies and communication experiences.Thematic analysis of the data revealed three key findings: Tianchi demonstrates resilience through technical proficiency and creative adaptations of mainstream technologies; the hidden labour of communication challenges complicate his access to community support and his reliance on both care and technology exposes fragile interdependencies.We further discuss: the importance of reducing input exertion for people with CP, the high costs of being heard and recognize the unanticipated benefits of non-assistive technologies.Ultimately, this study underscores the need for more participatory, first-person research approaches with communities living with CP.

Augmentative and alternative communication (AAC) technologies are typically fixed-form factor mobile devices, which can be physically obtrusive and visually identifiable – perpetuating public stigmas and limiting non-verbal communication. These qualities are undesirable for many AAC-using communities including stroke survivors, who can have complex communication needs onset by aphasia and hemiplegic body paralysis. Furthermore, underrepresented communities such as people with aphasia are often unconsidered in the design of emerging technologies. We contest these trends by asking what these emerging and future technologies might mean for communication support. To do this, we envision and prototype mixed reality and discreet assistive technologies that support communication with people living with aphasia. We report results from the co-design process, including participants’ perspectives on nascent mixed reality, diegetic technologies, and outcomes from low-fidelity AAC prototyping. These activities culminated in three high-fidelity prototypes with different characteristics and form-factors: Pico-project AAC, Prompt AAC and Holo AAC, which embody projection, worn audio devices, and headset driven mixed reality. From subsequent focus groups, we present findings from an evaluation of all prototypes, reflecting on the possibilities of present and future mixed-reality technologies to augment communication.

People living with complex communication needs employ multimodal pathways to communicate including: limited speech, paralinguistics, non-verbal communication and leveraging low-tech devices. However, most augmentative and alternative communication (AAC) interventions undermine end-users’ agency by obstructing these intuitive communication pathways. In this paper, we collaborate with 19 people living with the language impairment aphasia exploring contextual communication challenges, before low-fidelity prototyping and wireframing wearable AAC displays. These activities culminated in two low-input wearable AAC prototypes that instead, scaffold users’ pre-existing communication abilities. Firstly, the InkTalker is a low-power and affordable eInk AAC smartbadge designed to discreetly reveal invisible disabilities and usable as a communication prop. Secondly, WalkieTalkie is a scalable AAC app that converts smartphones into a feature-rich public display operable via multimodal input/outputs. We offer results from communication interactions with both devices, discussions and feedback responses. Participants used both AAC devices to interdependently socialise with others and augment pre-existing communication abilities.

Traditionally, Western philosophies have strongly favoured a dualist interpretation of consciousness – emphasising the importance of the ‘mind’ over the ‘body’. However, we argue that adopted assistive technologies become embodied and extend intentionality within environments. In this paper, we restore an embodied view of the mind to theoretically enhance: understandings of assistive technology and human-human communication. Initially, we explore literature on: phenomenological theories of human experience, post-phenomenological accounts of technology, embodied accounts of assistive technology and participatory design. We then present a case study demonstrating the generative and disruptive effects of the embodied framework for co-designing AAC with people living with aphasia. Our findings show that the embodied framework supports a more multidimensional account of experience and suggests a shift from AAC devices that seek to ‘repair’ users’ speech. Reflecting on our case study, we then outline concerns with nascent technologies that could disembody and limit accessibility.

With an ageing population and increased prevalence of people living with complex communication needs there is a growing need to design scalable high-tech augmentative and alternative communication (AAC) apps to support agency and social participation. For end-users it is currently difficult to regulate the prominence of most mainstream high-tech AAC devices and tablet-based apps – they are socially conspicuous, offer poor portability, are aesthetically unconsidered, and obstruct vital non-verbal communication pathways. In response to this, we leverage participatory design techniques to design and evaluate two discreet and inconspicuous AAC smartwatch apps. We engage with a community of people living with the language impairment aphasia, to collaboratively build and iterate both a smartwatch app for ‘public’ communication: Watch Out and ‘private’ cognitive support: Watch In. Following this, we evaluate both apps during an experience prototyping workshop with an actor and subsequent focus group. We report results from communication interactions with both apps, interviews and feedback responses. Participants were not only successful in using both AAC smartwatch apps but, critically, the wearable and discreet intervention did not restrict users’ agency and non-verbal communication.

High-tech augmentative and alternative communication (AAC) devices can offer vital communication support for those with complex communication needs (CCNs). Unfortunately, these devices are rarely adopted. Abandonment has been linked to many factors – commonly, stigma resulting from the visibility of the device and its intrusion into other essential modes of communication like body language. However, visible AAC is strategically useful for setting conversational expectations. In this work, we explore how we might envision AAC to address these tensions directly. We conduct user-centred design activities to build three high-fidelity AAC prototypes with different communities with CCNs, specialists and stakeholders. The prototypes demonstrate different form factors, visibility and modes of input/output. Subsequently, we conduct two qualitative focus groups using convergent and divergent co-design methods with people with the language impairment aphasia – supporting ideation of seven discreet and wearable low-fidelity AAC prototypes and critique of the three high-fidelity prototypes.

People with complex communication needs (CCNs) can use high-tech augmentative and alternative communication (AAC) devices and systems to compensate for communication difficulties. While many use AAC effectively, much research has highlighted challenges – for instance, high rates of abandonment and solutions which are not appropriate for their end-users. Presently, we lack a detailed survey of this field to comprehend these shortcomings and understand how the accessibility community might direct its efforts to design more effective AAC. In response to this, we conduct a systematic review and taxonomy of high-tech AAC devices and interventions, reporting results from 562 articles identified in the ACM DL and SCOPUS databases. We provide a taxonomical overview of the current state of AAC devices – e.g. their interaction modalities and characteristics. We describe the communities of focus explored, and the methodological approaches used. We contrast findings in the broader accessibility and HCI literature to delineate future avenues for exploration in light of the current taxonomy, offer a reassessment of the norms and incumbent research methodologies and present a discourse on the communities of focus for AAC and interventions.