UXAI: Bridging ML Complexity and User Trust
Making AI explainability accessible to the people who can advocate for it most: designers and product teams.
Context
UC Berkeley School of Information
MIMS Program graduation capstone
Role
Product Designer
in a group of four
Timeline
December 2019 – May 2020
Overview
A foundational resource for product teams new to AI, built during graduate school, still serving designers five years later.
UXAI is an explainable AI framework for early-career professionals, resource-constrained teams, and students. It's not expert-level material for seasoned AI practitioners. It's a starting point for those building their first AI products and learning to advocate for transparency.
I led the website design and helped conduct industry research with practitioners at Google and IBM. Our team also produced academic research on trust calibration between AI systems and different user types.
Note
This project predates LLMs and generative AI. The framework hasn't been updated since May 2020, yet it continues to draw ~1,000 monthly visitors as an educational resource.
The Challenge
Explainability knowledge was trapped in academic papers and ML research.
In 2019, "Explainable AI" was a growing field, but most resources targeted researchers and engineers. Product teams building AI features had few practical frameworks for thinking about when and how to make AI decisions transparent to users.
This gap hit hardest for teams without dedicated ML expertise: early-stage startups, small product teams, and students learning to design for AI.
Goal
Make explainability actionable for product teams just starting out, not just those with ML resources and expertise.
Research
Learning from academic literature and industry practitioners.
Our team conducted research across two tracks: academic literature review on trust calibration between AI and users, and practitioner interviews at Google, IBM, and other organizations building AI products.
Insight
Explainability was rarely prioritized early in product development. Designers and PMs were often unaware of what was technically feasible or why it mattered. Smaller teams without AI specialists felt this gap most.
Relevant industry resources at the time
The Gap
The people best positioned to advocate for user-facing explainability (designers and PMs) lacked the frameworks to do so. The people with explainability expertise (ML researchers) weren't involved in product decisions. This problem was amplified for smaller teams without access to AI specialists.
How might we equip product teams to advocate for explainability without requiring ML expertise or dedicated AI resources?
Design Principles
Meet teams where they are
Don't assume ML knowledge. Translate academic concepts into product language accessible to those just starting out.
Make it actionable, not just informational
Provide tools for real design workflows, not just reference material that requires expertise to apply.
Position designers as advocates
Frame explainability as a product decision, not just a technical one, so teams without ML specialists can still prioritize it.
Results
Explainable AI Framework
When and why explainability matters.
A structured approach to understanding when explainability matters and what forms it can take. Written for product teams without ML backgrounds, designed to be referenced during product development.
Example to advocate for explainable AI
Design Strategy Guide
Positioning designers as explainability advocates
Practical guidance on integrating transparency into product decisions. Aimed at helping designers and PMs make the case for explainability even when their teams lack dedicated AI expertise.
Strategy for "Who are your users and why do they need an explanation?"
Strategy for "When do users need an explanation?"
Brainstorming Toolkit
Cards for AI design sessions.
Card-based tools for product teams to use during design sessions. Prompts and frameworks for thinking through AI transparency without needing ML expertise. Low barrier to entry for teams new to AI product development.
Scenario cards for AI products
Explanation types for corresponding scenarios
Multiple explanation with examples
Impact
A starting point that keeps starting conversations.
~1,000 monthly visitors five years after launch, with no updates or active promotion. The continued traffic suggests the resource fills a gap for those entering the AI design space.
Retrospective
Enduring Principles in a Shifting Landscape
This project taught me that frameworks outlast features. The specific AI landscape has changed dramatically since 2020, but the core question, "how do we help users calibrate trust in AI systems," remains central to my work. Starting simple, for teams just starting out, clarified principles that scale to complex, high-stakes environments.
