Inside Indiana Money Quest, a ‘responsive, reactive’ financial literacy program powered by AI

Last week, the Indiana Secretary of State’s office launched Indiana Money Quest, a digital “responsive, reactive” financial literacy and fraud-prevention program that can be personally tailored to each resident’s needs through the use of artificial intelligence.

The new program was built using technology from the Michigan software firm Venturit and tools like Google’s Vertex AI, to create and deliver financial literacy content in real time. Some of these lessons, which include practical guidance on savings, retirement and how to spot scams, will depend on a user’s needs or their “micro persona,” a data-informed profile that represents a narrow segment of residents that the program can use to tailor content and learning pathways in a more precise way than traditional educational materials.

Robert Fulk, the chief information officer for the Indiana Secretary of State’s office, said the goal is to move beyond static websites toward continuously evolving systems that can adapt to changes in financial trends, user behavior and fraud risks.

Fulk said the project was developed for the Secretary of State’s securities division, to replace its former financial-education materials. Fulk described them as “paragraphs of stuff on a website” or “something from the 90s” that residents really weren’t using.

He said Money Quest was designed as a revamp to the government’s typical approach to improving the public’s financial literacy and preventing fraud. While the curriculum was built by certified instructional designers, the AI helped to turn those lesson plans into “micro lessons” by generating short modules and media, like videos, infographics and audio, that are adaptive to the user’s device, such as a phone, tablet or computer. Additionally, the state says the content meets all web and mobile accessibility guidelines.

Fulk said the content was organized into six “micro personas”: K-12 students, young adults, working adults, people nearing retirement, seniors and educators.

“They all have different needs and wants. So we did this whole persona-based stuff. The idea is that we can really engage them in a different way, and to provide them resources they are really going to look at and learn from,” Fulk continued. “We really looked at these personas, and what do these people want and need out of a system. … And so their paths and their experiences, the learning pathways, are completely different.”

He said keeping AI away from core financial guidance and confining it to production work, like graphics, audio and video, helped to eliminate a number of risks, such as harmful or wrong advice, especially in high‑stakes areas like fraud and investments. He said it also reduced costs.

“We literally did this in about four weeks to completion. … We would never be able to do that without AI,” Fulk said. “We literally probably did this phase one of this thing for fifty grand, and this would easily [have] been a quarter million dollar project.”

The new platform also comes with new tracking tools, which Fulk said will allow the office’s securities division to monitor engagement and course completion, which are just a handful of the metrics it couldn’t measure previously. Fulk said his team and its partners at Google and Venturit are plotting the second phase of the program, which he hopes will help connect the state’s K-12 and collegiate educators, so they can bring these resources into the classroom.

Money Quest is the second educational program Fulk’s team at the Secretary of State’s office has revamped with generative AI. Last fall, his office updated the state’s notary training system, similarly replacing “outdated and clunky” coursework with AI-generated scripts, audio, video and interactive modules. This coming weekend, Fulk said, his office will launch phase two of the notary system’s upgrades, this time using AI to expedite the notary licensing process. The AI, Fulk said, will review notary applications, make recommendations and automatically create a queue for human review. This, he said, will “reduce the workload 60 to 70% in our back office.”