Backed by a $2.8 million grant from the NIH’s National Cancer Institute, University of Maryland communication professor Xiaoli Nan and a cross-disciplinary team are creating a personalized, AI-powered chatbot designed to address parents’ concerns and communication styles around the HPV vaccine.

“The vaccine is safe, effective and prevents multiple cancers, but many parents remain hesitant,” Nan explained. “Our goal is to use AI to meet parents where they are and provide support in making informed decisions.”

The project brings together experts including Cheryl Knott and Min Qi Wang from the School of Public Health; Philip Resnik, an MPower Professor in linguistics and computer science; and Clement Adebamowo and Shana Ntiri from the University of Maryland School of Medicine.

Although physicians recommend HPV vaccination for adolescents as early as age 9 to protect against cancers of the cervix, throat, anus and more, adoption has lagged due to misinformation, stigma, and distrust, Knott noted. Some parents delay because of the vaccine’s link to a sexually transmitted infection. Uptake is especially low among men, people with limited education, and those with less access to healthcare.

Traditional tools like brochures and websites often fall short because they don’t respond to individual concerns. This chatbot uses natural language processing to simulate conversation while tailoring messages to each parent’s beliefs and communication preferences. Unlike simple scripted systems—or open-ended generative AI—the chatbot relies on large language models in a controlled, “middle ground” approach. It generates potential replies that are pre-vetted by experts, then selects the most relevant response in real time.

The first phase emphasizes user-centered design, gathering feedback from parents, healthcare providers, and community members. The team will then conduct two randomized controlled trials: one online with a national parent sample, and another in Baltimore pediatric clinics.

According to Wang, this approach represents a “paradigm shift” in public health communication, offering tailored, timely, and actionable information beyond the reach of traditional methods.

Beyond improving HPV vaccine uptake, the researchers believe this work could model how AI helps public health systems respond more quickly and inclusively during future crises. “This project really makes us think about what’s possible,” Nan said. “Can we improve public health communication with speed, scale, and empathy? I believe the answer is yes.”

Source: https://sph.umd.edu/news/umd-researchers-use-ai-help-boost-hpv-vaccine-confidence