George Washington University

01/13/2026 | News release | Distributed by Public on 01/13/2026 13:07

GW Business Students Harness AI in Inaugural Chatbot Competition

GW Business Students Harness AI in Inaugural Chatbot Competition

FraudLens, a chatbot that determines why a transaction is flagged, judged winner by industry executives.
January 13, 2026

Authored by:

Mary A. Dempsey, Sher Delva

Team members Maryam Shahbaz Ali, Tyler Ordiway, Zihui Chen and Pingyi Xu gather for a photo with Professor Patrick Hall after winning an AI competition for their chatbot, FraudLens.

Impressing executives with an artificial intelligence (AI) project wasn't always part of the decision sciences curriculum at the George Washington University School of Business,but the industry has shifted. Now, students create AI software judged by business leaders as the program positions itself at the forefront of the rapidly growing industry. .

The new decision sciences practicum culminated in a case competition on Dec. 5, when executives from Capital One, EY, FI Consulting and Mizuho Financial Group gathered to evaluate student-designed AI projects. From helping users understand fraud to creating a time-saving travel chatbot, the projects showed how students can tackle real-world problems.

The winning team-Maryam Shahbaz Ali, Zihui Chen, Tyler Ordiway and Pingyi Xu -impressed judges with FraudLens, a chatbot that combines fraud detection and machine learning to help users understand why a transaction is flagged.

"Many machine learning systems flag transactions but don't explain the reasoning behind their decisions," said Ali, a member of the winning team and a business analytics graduate student. "FraudLens was built to bring transparency and accountability to that process by combining a responsible, explainable model with a chatbot that highlights the key factors behind each alert."

FraudLens makes it easy to understand complex fraud alerts, making it faster and easier for users to respond.

Teaching Assistant Professor of Decision Sciences Patrick Hall, who redesigned the practicum as a competition, said the goal was to give students hands-on experience with AI tools and real-world problem solving. Hall was recently named first chief AI officerat GW Business, where he is tasked with advancing human-centered and trustworthy AI within the school.

"Students used to meet with industry leaders to build prototypes for a specific company, but there has been a huge shift in the business analytics world because of AI," he explained.

He said Cassin Dyson, executive director of the F. David Fowler Career Center, suggested turning the practicum into a case competition, and the career center recruited industry leaders to serve as judges.

Hall described the competition as a crucible for students. For Ali, competition day was both exciting and overwhelming.

"When I first arrived at the school, I had moments of self-doubt about whether I could meet the demands of the program," she said. "Coming from a non-technical background pushed me to work harder, learn continuously and challenge myself. That motivation shaped how deeply I invested in this project."

And the moment her team was declared the winner?

"I was in tears," Ali said. "I had put so much into this project: my doubts, the late nights and the effort to prove to myself that I belonged. Seeing that work recognized was incredibly rewarding."

Ali said Hall provided students with critical support on the projects.

"He was always available to answer questions and offer guidance," she said. "Without that mentorship, FraudLens wouldn't have reached this level of impact."

Teams, guidance and real-world pressure

The practicum included 61 students divided among 14 teams. Students met regularly with Hall and the judges for guidance. Midterm peer reviews maintained accountability within teams. In addition to the competition, students were required to set up poster presentations explaining their projects then field questions from the public on the day of the case competition.

The student projects were underpinned by data from a set of industries: finance, housing, travel and real estate. Sydney Frelich and Edwin Ordonez, both seniors in their final semester at GW, worked on a team that developed a travel chatbot. The idea was spurred by Frelich's plan to backpack in Southeast Asia before she begins a job in a management rotation program at Capitol One.

"It's hard to pick a topic that's somewhat original because as AI develops there are so many chatbots," Frelich said. "We decided on an AI travel agent. We thought a lot of people would relate to a platform that consolidates everything for a trip."

In customizing a travel itinerary, the chatbot also asks users about their travel budgets and the activities they like.

Ordonez, who is graduating with a Bachelor of Science in Business Analytics, has a job lined up with Gainwell Technologies in Texas. He said the project worked as a resume booster because it encompasses communication with business professionals, project management and technical skills.

"We're literally building an AI agent, and our major doesn't do a lot of that. I've learned so much and my teammates are amazing," Ordonez said.

"It's definitely different from a lot of my curriculum," added Frelich, a business analytics major with a marketing minor. "I'm a hands-on learner, and I think it's a great class."

On case competition day, each team had 30 minutes to present their project to judges and answer questions. Students were graded on both project results and individual contributions. Members of the public visiting the poster presentations also voted on the best project.

For Ali, the AI project experience was as much personal as professional.

"Nothing is impossible," she said. "If you commit your time, effort and heart to a problem, you can solve it. This project showed me that. It gave me confidence that I can handle challenges in my future career."

Related Content

George Washington University published this content on January 13, 2026, and is solely responsible for the information contained herein. Distributed via Public Technologies (PUBT), unedited and unaltered, on January 13, 2026 at 19:07 UTC. If you believe the information included in the content is inaccurate or outdated and requires editing or removal, please contact us at [email protected]