ChatSDG: Harnessing AI for Measurable Societal Impact

At the heart of the United Nations 2030 Agenda for Sustainable Development are 17 Sustainable Development Goals (SDGs) that call on all countries to take action toward peace and prosperity. The SDGs range from creating equitable access to quality education to ensuring clean water and sanitation for all. ChatSDG is a tool for universities and academics to measure their research impact on these critical goals. 

The generative AI model is a collaboration between the SDG Dashboard team within Saint Joseph’s University’s Haub School of Business and Cabells Scholarly Analytics. David S. Steingard, PhD, associate professor of management, who leads up the SDG Dashboard team, stresses the importance of “research with impact.” 

“[ChatSDG] is a breakthrough application,” Steingard says. “The large language model can analyze research being done and tell those involved how that research does, or does not, make an impact on the accomplishment of the SDGs.”

Not only is the application itself a milestone in the quest for finding socially responsible and beneficial ways to use artificial intelligence, but the undertaking has accelerated a paradigm shift in the way academic research is perceived and judged. This is a much needed revelation if universities are to play a significant role in the progress made toward the SDGs.

“Right now,” Steingard explains, “research is measured on more of a scholar-to-scholar basis. Was the research published in a reputable journal? Was it peer reviewed? It’s all part of this ‘publish or perish’ mentality.”

In a paper under review for publication, Steingard and co-authors David J. Reibstein, PhD, of the Wharton School at the University of Pennsylvania, and Mark Normandin, MS ’26, graduate student in the MSBIA program, have dubbed this dynamic the “citation counting dysfunction” — the main aim of which is to increase citations, often at the expense of broader, interdisciplinary or socially relevant research.

To demonstrate ChatSDG’s ability to counteract this phenomenon, Steingard pulls up the application and feeds one of his papers on SDGs into the platform. In seconds, ChatSDG delivers a qualitative and quantitative analysis of the paper complete with an explanation of which SDGs the research supports and a score of real-world relevance. Steingard’s paper, not surprisingly, receives a score of 4.7 out of a possible 5 points.

ChatSDG’s usefulness is driven home when a second paper, chosen at random from an online repository, receives a score of 2.4.

Steingard recalls performing a similar demo at a recent conference. Academics were able to feed their own research into ChatSDG to see for themselves what the impact of their research was. Steingard politely describes the experience for many of these individuals as “eye-opening.”

“It's a collaboration between humans and AI,” Steingard clarifies. “[ChatSDG] can make humans better and more consistent. It's an accelerator.”

Steingard, no stranger to responsible research and ethics, will be the first to offer the caveat that any AI is only as ethical, responsible and accurate as those who build and maintain it. That, he offers, is what makes SJU the perfect home for ChatSDG.

“The Jesuit values at SJU are built on thinking critically, making ethical decisions, and pursuing social justice and environmental sustainability,” Steingard concludes. “We’re not alone on this, we’re part of something bigger. But SJU is making a really important contribution to this effort for a sustainable future.”