Despite the growth of video as a medium, videos remain inaccessible to many people. Prior video accessibility research has focused primarily on blind and low vision or d/Deaf and hard of hearing audiences. However, the video watching experiences of people with ADHD are largely unexplored. Through semi-structured interviews with 20 participants self-identifying with ADHD, we uncovered video watching frustrations, current strategies for access, and desired accessibility features. Participants faced both overstimulation and understimulation from visuals and audio (e.g., flashing lights, slower speech), which impacted their attention, engagement, and information retention. Common strategies included altering video speed, using captions, and leveraging timestamps for skipping through videos. Participants desired adjustable sound channels for aiding focus, video summaries for retaining information, and warnings for preempting sensory discomfort.

We close by discussing (1) design recommendations for platforms and creators to support users in achieving their viewing goals and (2) ADHD-inclusive design principles.

Visual media is often made accessible to blind and low vision (BLV) people through audio description (AD), typically written by experts. Prior efforts to increase the scale of description output have involved sighted novices as describers or used generative AI (GenAI) to automatically convert images to text; however, description quality remains a concern. To support novice describers in writing high quality descriptions, we designed and developed a GenAI-powered online tool, "Guidedogs," featuring five dogs with unique names, images, and voices that provided immediate and varied feedback on draft descriptions. We piloted the tool during a large hackathon-style description workshop in 2024. Through 17 semi-structured interviews, we explored the efficacy of using metaphors as personas for AI assistants and gathered insights on participants' perceptions on using AI for accessibility purposes. We contribute preliminary insights on generative AI assistant personas in an accessibility context and share design considerations to guide future work.

Disabled people on social media often experience ableist hate and microaggressions. Prior work has shown that platform moderation often fails to remove ableist hate, leaving disabled users exposed to harmful content. This paper examines how personalized moderation can safeguard users from viewing ableist comments. During interviews and focus groups with 23 disabled social media users, we presented design probes to elicit perceptions on configuring their filters of ableist speech (e.g., intensity of ableism and types of ableism) and customizing the presentation of the ableist speech to mitigate the harm (e.g., AI rephrasing the comment and content warnings). We found that participants preferred configuring their filters through types of ableist speech and favored content warnings. We surface participants' distrust in AI-based moderation, skepticism in AI's accuracy, and varied tolerances in viewing ableist hate. Finally, we share design recommendations to support users' agency, mitigate harm from hate, and promote safety.