Trustworthy products for the public good: the USA.Gov AI chatbot experience
Client: GSA—Public Experience Portfolio | Year: 2025
Role:
Lead UI/UX design
Team:
1 UI/UX designer, 2 developers, 1 GSA lead (PM)
Tools:
Figma, FigJam, Slack, Trello
Goal:
Create a proof of concept (POC) utilizing Drupal AI and conversational UX to integrate with OpenAI for government. Highlighting initial flows of GSA unclaimed money and GSA auctions.
Navigation
Improving government services with AI
If you recently called a government call-center, you may have noticed increased wait times and difficulty in finding a “real” person on the other line. You may have also noticed:
- Being read a repeat of existing information already on the site
- Inconsistency in response/answer from one agent to the next
- Difficulty in understanding the agent, or frustration in communication style
Our customer service agents experience a high volume of calls that are solved by researching available contents on our site. They are also under constant pressure to perform and respond as accurately as possible within contractual obligations. We remove the call center option from the site navigation once we reach the quota. Since the contact center also operates within a limited time frame, every communication matters. If the contractual obligations are roughly 10,000 calls/chats a day, a simple answer may deny others a response to their more complicated issues; leaving a large percentage of people unserved.
How do we create a system that can use our customer service agents more effectively, reduce simple, searchable answers and still provide a service to the American people? Our solution was an AI chatbot designed to remove inaccuracies or room for interpretation with reliable and vetted information. Enabling the American public to get immediate answers in a natural language dialogue.
Project overview
Along with my team, we created a Proof of Concept (POC) that would use AI as a chat bot to communicate USA.gov information and highlight GSA specific offerings. At first instance, we explored the options of unclaimed money, financial hardship information and GSA Auctions. We also explored different technical and administrative constraints in a time of government layoffs and reductions in force.
My primary task was creating UI/UX flows for the chatbot, using principles of conversation design as well as creating the initial designs and handoffs for developers. The project lasted approximately two months with weekly sprints and demos.
Reflection & outcomes
Designed a new UX strategy applying AI for the USAGov chatbot with a focus on accessibility and intuitive user flows. We completed the design work and set it up for successful deployment, even though the project concluded before we could gather all the implementation metrics.
Taking advantage of Drupal AI, Ollama and multiple resources, we created a firm foundation based on USWDS web standards and USAGov brand identity that will improve user flows, create scalable UI components and reduce contact center call/chat volume.
Lessons learned
- In a project as complex as the USA.gov AI Chatbot, parameter chains played a critical role in breaking down information into manageable, visual components. Early mapping of controllable parameters aligned stakeholders, reduced ambiguity, and prevented scope creep as the system developed.
- The complexity of variables made it difficult to establish explicit AI control upfront. Without real user data, we relied on informed assumptions and strategic best-bets—critical tools that shaped early design decisions and guided user testing.
- I approached the chatbot as an AI agent, emphasizing conversational design principles to maintain a natural flow. This included offering users obvious ways to revisit previous topics, exit loops, or clarify intent—creating a circular, user-led experience that mimics real human interaction
- To support a smooth conversational experience, the chatbot also needed fallback responses for unsupported topics, along with multiple guided flows that could interpret user intent through keywords and phrases—ensuring users always had a clear path forward.
- While the long-term vision is for the chatbot to handle a wide range of user needs—including potential direct integration with the contact center—we recognized the importance of designing for phased improvements. Given possible contractual limitations, we planned for a fallback that directs users to the contact page, ensuring a clear handoff even in the worst-case scenario.
- Although I was familiar with the basics of conversational design, this project pushed me into a deep exploration of dialogue and interaction principles. It involved extensive research into conversational design best practices, making the experience both intellectually rewarding and genuinely enjoyable.
- As the project developed, maintaining consistency was critical. Implementing Figma variables and design tokens brought structure to the system, enabling us to scale the design more intentionally and efficiently.
- Although the USA.gov site does not currently implement dark mode, it is a consideration for future updates. Dark mode’s introduction would benefit chatbot users, enhancing accessibility and user experience.
