GenAI x Mobile Voice Assistance Generative Study 

• Problem Space: In 2023, generative AI (GenAI) and large language models (LLMs) were rapidly advancing, with major tech companies investing in in-house LLM development. However, a key challenge was identifying real-world, on-the-go use cases for this technology.

• Product Context: As part of the Echo Frames team (Amazon’s wearable smart glasses), the business sought to explore how an LLM-backed AI assistant could enhance mobile and wearable experiences.

• Research Goal: My research aimed to map potential use cases and user expectations for an on-the-go AI assistant, helping UX designers and product teams identify high-value opportunities for future software integrations.

• What expectations do users have for an LLM-backed AI assistant in mobile or wearable contexts?

• Which potential use cases resonate most with users? (Based on an initial list ideated by UX Designers.)

• How do expectations differ across user demographics (e.g., age, gender, prior experience with LLMs, and smart device usage)?

• What concerns or barriers might prevent adoption of this technology?

• Mixed-Methods Research:

  • Survey (Quantitative & Qualitative Data Collection)

    • Designed and launched a Qualtrics survey to collect demographic data, smart device usage, prior LLM experience, and preferences for potential AI use cases.

    • Incorporated Q-sort ranking to assess user preferences for different AI-powered tasks and gauge enthusiasm or skepticism.

    • Included open-ended qualitative questions to capture nuanced expectations and concerns.

  • Follow-Up Interviews (Qualitative Deep Dive)

    • Selected participants based on extreme high and extreme low ratings for AI usage to explore both enthusiasm and skepticism.

    • Conducted semi-structured interviews to understand users’ mental models, imagined scenarios, and potential adoption barriers.

•     Analysis & Synthesis:

  • Conducted statistical analysis in R to identify trends and correlations across demographic segments.

  • Visualized survey findings with data graphics produced in R.

  • Thematically analyzed interview data to extract key insights and user narratives.

• End-to-End Research Execution:

  • Led the project from initial stakeholder alignment to study execution, analysis, and presentation.

  • Met with Senior UX Designers to refine research questions and align on key business and design needs.

• Survey Design & Data Collection:

  • Designed the survey and collaborated with a fellow UX researcher to implement Q-sort ranking in Qualtrics.

  • Recruited participants internally through Amazon’s employee network via Outlook.

• Interviewing & Qualitative Analysis:

  • Scheduled, conducted, and transcribed all semi-structured interviews.

  • Thematically analyzed interview responses to surface key behavioral patterns and concerns.

• Data Analysis & Visualization:

  • Used R for statistical analysis and to create clear, visually compelling graphics of survey findings.

• Stakeholder Communication & Impact:

  • Presented findings at the Amazon Echo Frames / Devices & Services org-wide capstone presentation, documenting key insights to inform future AI integration strategies.

(Under NDA – Generalized Summary Below)

• Delivered foundational insights on user expectations and adoption barriers for wearable AI assistants.

• Identified high-priority use cases and scenarios where users see the most value in on-the-go AI assistance.

• Provided demographic-based insights that informed design considerations for AI interactions in wearable devices.

• Research findings were internally documented, serving as a reference for future AI product roadmaps within Amazon’s Devices & Services organization.