GCPProductionEvidence: Medium55/100

Indiana Secretary of State modernizes state services with Gemini AI and Vertex AI

The Indiana Secretary of State transformed over 20 million historical government records into searchable intelligence using Google Cloud's Gemini and Vertex AI, enabling rapid data insights and modernized legacy systems. They deployed bilingual AI-powered conversational agents Liz and Otto to provide 24/7 resident support in English and Spanish, significantly increasing team capacity and user access. Additionally, the office launched a Notary Learning Management System (LMS) powered by AI on Google Kubernetes Engine, training over 50,000 statewide notaries with interactive micro-learning, leading to an award for Best Application Serving the Public at the State Level.

Published
May 2026

Reported outcomes

Strategic outcomes

Better decisions & insightTransformed legacy records into searchable intelligenceCustomer experience & trustProvided bilingual 24/7 resident supportNew product / capabilityLaunched AI-powered notary training systemSpeed & agilityEnabled instant AI-assisted licensing approvals
Why do we believe this deployment?Customer identity, provider attribution, maturity, and source checks
Customer
Indiana Secretary of State
Provider
GCP
Maturity
Production

They deployed bilingual AI-powered conversational agents Liz and Otto to provide 24/7 resident support in English and Spanish, significantly increasing team capacity and user access

Customer identity supportedSource describes one deploymentMaturity supported

Primary read

Use case focus

Showing 3 of 3

  • 1AI-powered conversational agents
  • 2Data modernization
  • 3Secure AI training platform
  • State public records were digitized but locked in static, fragmented formats, making research slow and inefficient, traditionally requiring weeks for data retrieval.
  • The state aimed to modernize legacy data systems and improve resident services with AI-enabled solutions while ensuring security and compliance with government standards.
  • Used Google Cloud's Gemini models and Vertex AI to build intelligence layers on top of 20M+ legacy records, transforming them into accessible, searchable knowledge.
  • Deployed Gemini-powered bilingual AI agents for conversational resident support on the INBiz portal, enhancing accessibility and system reliability with ground data.
  • Built an AI-powered Notary Learning Management System (LMS) on Google Kubernetes Engine with interactive, AI-generated content to replace static manuals and improve training retention.
  • Implemented Zero Trust security architecture leveraging Wiz for full stack visibility, data encryption, and alignment with stringent NIST SP 800-53r5 standards.
  • Reduced research time drastically from weeks to minutes, increasing operational efficiency.
  • Bilingual AI agents increased support team capacity tenfold, providing 24/7 service for residents.
  • The Notary LMS trained over 50,000 professionals statewide, improving education accessibility and winning the 2025 Government Technology Award for Best Application Serving the Public.
  • The modernization paved the way for advanced AI workflows including agentic licensing with AI reviewing licensing materials for instant approvals, further streamlining state operations.
Architecture

Architecture leverages Gemini and Vertex AI for AI-driven data intelligence layers; BigQuery as data warehouse; Google Kubernetes Engine for scalable AI-powered LMS and AI agents; Cloud Run and Cloud Storage for platform hosting; and Wiz for advanced zero trust security compliance.

Sources & evidence1
Evidence: Medium55/100Evidence strength
  • Customer explicitly identified
  • Deployment status explicitly supported
  • Primary source available
  • Technical implementation details available
Type: Customer StoryPublished: May 10, 2026Publisher: Google Cloud Customer StoriesEvidence: PrimaryConfidence: High

AI-generated summary. Verify important details with the linked sources before relying on this case.

Explore related AI use cases

Was this useful?

Community

Comments

Loading comments...

Similar cases