How might we create user experiences for AI? I wanted to learn more so I joined User Experience for AI, 3 credits by Halmstad University, spring 2025.
Diary Studies to learn about user experience
We was tasked with Diary Studies, to learn about our own everyday user experience with technology. After two weeks of diary and self-observation, I engaged in reflection. The activity of walking stood out as a particularly meaningful everyday experience. I used my noise-cancelling headphones during many of my daily experiences with headphones. I used them all day for all sorts of activities, running, exercising, studying, relaxing, travelling/commuting. But more than anything, I was using them for walking. An experience had almost become visible to me. Keeping a diary was helping me to spot this almost unknown daily habit.
After logging my own headphones and walking activities for two weeks, I made an additional survey. I surveyed 5 female respondents in a similar age group. I aimed to test and validate my diary study. This led to the development of the persona of an urban walker. I limited the experience to keep the work manageable, given the time constraint. The research shows headphone usage for commuting, workday, as well as personal home life with family.
The scenario of walking with headphones stood out to me
I adapted the scenario of the walking experience into a scenario of an urban lunchwalk. This adaptation was to fit in the goals and needs covered in the research. It created one user journey ”as is”. Doing so, I was able to highlight the biggest user pain point, toggling between different modes. I also made additional desk research to learn about headphone technology. I explored other trends and technologies apart from AI. They may affect our future listening and walking experience.
When and how do we use headphones today?
I discovered headphones was used throughout the entire day. We wear them when we commute and during work. We also use them during leisure time. We listen to music, podcasts, and news. We watch videos or take phone calls. We also use them to avoid disturbing sounds like engines. We use the headphones to shield ourselves from our surroundings and to disconnect from the environment. This helps us focus on work or find inner peace and relaxation in public environments. We also use them to avoid disturbing people near us. This allows us to be private with what we listen to or watch.
It’s important that despite the headphones, we can hear ambient sounds. These include traffic when cycling or walking. This ensures that neither others nor ourselves pose a danger. The headphones themselves must not jeopardise our hearing either. The problem users have is primarily being able to switch, or toggle between different modes. It’s important to sync with the right devices quickly and smoothly. It’s crucial to shut out sound in public environments. Hearing traffic is also essential.
How can AI support the experience?
From my perspective, the core strength of AI lies in human-AI collaboration. Inspired by Ben Shneiderman and Pontus Wärnestål, I view AI as a powerful design material. It augments human capabilities. The technolgies allows us to solve many problems more effectively. However, AI should enhance human decision-making instead of replacing it. It should deliver clear, transparent, explainable, and usable solutions. These solutions must be aligned with genuine human needs. In my design concept, AI-integrated noise-cancelling headphones dynamically enhance walking experiences by managing environmental awareness, emotional states, and productivity.
A critical moment juggling between different modes
The critical challenge in this scenario is effectively toggling between noise-cancelling and ambient awareness modes. This is essential to maintain user safety and provide mindfulness opportunities. Here, generative AI can help by interpreting real-time sensory data. This includes location, emotional and physical states, and calendar schedules. It crafts personalised walk suggestions. By proposing adaptive experiences, such as mindful walks, energetic routes, or social interactions, generative AI enriches the user’s everyday routines. However, integrating generative AI raises important questions about trust.
Trust in AI is challenging. These systems operate as probabilistic ”black boxes.” Their decision-making processes are often opaque even to developers. Trust-building, therefore, requires explicit transparency about the capabilities and limitations of the AI system. Calibrated trust is a key concept introduced in our course. It emphasis’s ongoing user engagement and clear communication of system actions. It also promotes easy options for users to intervene or override AI decisions. Practical interaction designs can significantly help users feel in control. Techniques such as voice commands or haptic feedback are useful, particularly at critical moments like switching between audio modes.
Bias represents another crucial concern. AI systems inherently reflect biases present in their training data, which can lead to unintended discrimination or inaccuracies. Awareness and proactive management of bias are achieved through diverse and inclusive user testing. It also involves transparent data sourcing and regular model recalibrations. These actions are essential. For instance, subtle biases in generative suggestions could unintentionally reinforce harmful stereotypes or neglect certain user needs. Ensuring continuous oversight and user feedback loops can mitigate these risks, fostering a more equitable and inclusive user experience. Relating this reflection directly to my headphone concept, trust and bias management are integral. Users must reliably trust the headphones to accurately switch modes without compromising safety or comfort. Mismanaged toggling could undermine user confidence or, worse, lead to dangerous scenarios.
Thus, a transparent, clearly communicated, and user-controllable AI interaction model is vital. In conclusion, thoughtful design and rigorous testing of AI-enhanced experiences can significantly elevate everyday interactions. We must address trust, transparency, and bias proactively. By ensuring these elements are central to design, AI can effectively complement human abilities. It can amplify these abilities, transforming routine activities like walking into meaningful, safe, and engaging experiences.
Persona: The Urban Walkers user experience
The persona would express thoughts and feelings such as, ”I love it when it’s smooth. It’s easy to toggle between modes with my headphones.” Needs include the ability to toggle between modes. It is important to secure safety goals. For instance, one can unwind and recharge by muting the outside world. It is also important to create context awareness of the surrounding environment. This includes situations like traffic or being at work. Focus can then be placed on interactions or deep work.
User Journey for an urban walking tour
The user journey grew over time. Resaarch, eflections, and information were added to new assignments tasks in swimlanes. These included ”future” opportunities. Here we stop, but it was a true joy and interesting work.
Now I am so excited to continue my own AI journey. I am learning how to craft human centered experiences for AI driven services. Welcome to connect with me at LinkedIn!
References:
Lectures User Experience for AI Module 1 and 2
Wärneståhl, P. (2022) Designing AI-Powered Services. Studentlitteratur
https://www.forbes.com/quotes/11375/
https://www.gov.uk/government/publications/futures-toolkit-for-policy-makers-and-analysts/the-futures-toolkit-html#introduction
https://www.sciencedaily.com/releases/2025/05/250520122024.htm
https://www.youtube.com/watch?v=ArGKgodEUSo&t=84s
https://www.audiophileon.com/news/the-future-of-headphones-trends-innovations-and-challenges
https://www.audiophileon.com/news/bone-conduction-headphones
https://www.audiophileon.com/news/best-bone-conduction-headphones
https://naenka.com/blogs/products/the-future-of-bone-conduction-across-the-globe-naenka?srsltid=AfmBOoquFg1tnDKtbmF_AjBZgpLY2Y2awEsXsPef0oyBIeY8J3Zop-0k
https://www.lifewire.com/bone-conduction-headphones-audio-ar-8762755
https://hpoal.com/the-future-of-ai-in-hearing-aid-technology/
https://www.zdnet.com/article/this-new-bone-conduction-headset-comes-an-ai-trainer-plus-health-tracking-and-more/
https://www.youtube.com/watch?v=ArGKgodEUSo&t=84s
https://www.audiophileon.com/news/the-future-of-headphones-trends-innovations-and-challenge
https://nielseniq.com/global/en/landing-page/webinar-recording-state-of-consumer-tech-in-2025/?submissionGuid=16144f84-9539-47eb-b5df-dc14e6f0ffec
https://www.unite.ai/sv/ai-h%C3%B6rlurar-l%C3%A5ter-dig-lyssna-p%C3%A5-en-person-i-en-folkmassa/
https://programs.sigchi.org/chi/2024/program/content/147319
https://www.unite.ai/sv/b%C3%A4sta-ai-ljudf%C3%B6rst%C3%A4rkare/
