AI Chatbot: Blood Center
The Arizona State University Artificial Intelligence Cloud Innovation Center, powered by Amazon Web Services (AWS), collaborated with a Blood Center to design a public-facing chatbot that demystifies blood supply. The solution helps people understand how donors save lives, learn about blood supply realities, and explore advocacy actions that strengthen access to blood products for every patient in need because informed clicks can lead to lifesaving outcomes.
Problem
Across the U.S., the public relies on a constant, safe, and equitable blood supply—but few people understand how the system works, where blood comes from, how it reaches patients, or how policy decisions can impact access. This lack of awareness can limit donor engagement, weaken public advocacy, and leave key parts of the healthcare infrastructure vulnerable during shortages or emergencies.
They identified an opportunity to “start the conversation” with an approachable tool that can answer common questions (like eligibility and donation basics), direct users to the right resources (like finding a donation location), and point people to timely updates from the blood centers newsroom when they ask about recent news.
Student Spotlight
Role: Cloud Developer | Connect with Ridham on LinkedIn
The most exciting part was implementing real-time streaming responses where users can see the AI's answer appear word-by-word instead of waiting for the full response. I also enjoyed being among the first to use AWS Bedrock's new Data Automation feature for multimodal document parsing, which extracts text, images, and tables from PDFs automatically. Seeing the chatbot accurately answer blood donation questions and help people find nearby donation centers was really rewarding.
I worked as a full-stack developer handling both backend and frontend. On the backend, I built the serverless infrastructure using AWS CDK setting up Lambda functions, Bedrock Knowledge Base with three data sources, OpenSearch for vector search, and Step Functions for automated syncing. On the frontend, I developed the React chat interface with bilingual support (English/Spanish), streaming responses, and an admin dashboard for monitoring conversations and managing data sources.
I learned how to build AI-powered applications using AWS Bedrock and RAG (Retrieval-Augmented Generation) architecture. I gained hands-on experience with Lambda Function URLs for streaming, Step Functions for workflow orchestration, and OpenSearch Serverless for vector embeddings. I also improved my skills in React development, Cognito authentication, and designing cost-effective serverless solutions. Most importantly, I learned how to build technology that genuinely helps people connecting blood donors with donation centers and providing accurate health information.
Role: UI/UX Designer | Connect with Ashik on LinkedIn
The exciting part of the Blood Centers project was making complex information about the blood supply easy for anyone to understand. It was meaningful to design a tool that could educate people and also encourage real actions like donating blood or supporting advocacy, which can help save lives.
I was the User Experience designer for this project, and my responsibilities included understanding the problem, defining the user experience, and designing the interface for the chatbot. I worked on structuring the content, creating user flows, and ensuring the design was clear, accessible, and aligned with America’s Blood Centers’ goals.
I learned how to simplify complex and sensitive information so it is easy for a general audience to understand. I also learned how important trust, clarity, and accessibility are when designing tools for public health, and how thoughtful UX decisions can encourage users to take meaningful real-world actions.
Approach
The CIC team built a serverless, citation-first chatbot on AWS with three main flows: Public Chat, Admin Dashboard, and Knowledge Refresh/Ingestion.
- AI/Knowledge grounding: Curated knowledge sources and blood centers fact sheets/FAQs/issue briefs to enable citation-backed answers.
- Knowledge refresh: Blood Supply status content refreshed daily via automated workflow scheduled for 2 PM EST (keeps supply-status information current).
- User experience: Bilingual English/Spanish support; donation navigation routes “where to donate” questions to Find a Blood Center page; news routing points “recent updates” questions to blood centers newsroom content.
- Admin & analytics: Internal dashboard surfaces questions asked and chat logs to support monitoring and continuous improvement.
- Infrastructure (AWS): API Gateway (secure endpoints), Lambda (chat + refresh logic), EventBridge (2 PM EST schedule), S3 (knowledge + refreshed content storage), DynamoDB (admin data store), CloudWatch (logs/metrics).
Industry Impact and Problem Solving
This chatbot helps blood centers “start the conversation” with the public making complex information about the blood supply easier to access, understand, and share. By pairing approachable answers with source citations, the experience supports trust and transparency while guiding people toward meaningful actions: donating blood, learning eligibility requirements, and engaging in advocacy that strengthens patient access to blood products.
Potential for Wider Application
The solution framework, an AI chatbot with multilingual support, citation-backed responses, and a maintainable knowledge base can extend beyond this use case to other public-serving organizations that need to communicate complex, high-impact information clearly and consistently.
This approach can be especially valuable in healthcare and nonprofit settings where accuracy, transparency, and accessibility are essential.
Supporting Artifacts
| Github Link | Contact us for link: [email protected] |
Next Steps
The codebase was prepared for open-source release, enabling other blood centers and public-serving organizations to adapt the framework for their own communities.
About the ASU CIC
The ASU Artificial Intelligence Cloud Innovation Center (AI CIC), powered by AWS is a no-cost design thinking and rapid prototyping shop dedicated to bridging the digital divide and driving innovation in the nonprofit, healthcare, education, and government sectors.
Our expert team harnesses Amazon’s pioneering approach to dive deep into high-priority pain points, meticulously define challenges, and craft strategic solutions. We collaborate with AWS solutions architects and talented student workers to develop tailored prototypes showcasing how advanced technology can tackle a wide range of operational and mission-related challenges.
Discover how we use technology to drive innovation. Visit our website at ASU AI CIC or contact us directly at [email protected].