Stony Brook University Libraries Named Finalist for International Innovation Award

Stony Brook University Librarieshas been named a finalist for the inaugural Clarivate Library Innovation Award, an international honor recognizing libraries that are advancing research and learning through innovative ideas and the creative use of technology.

The recognition cites SEARCH AI, an artificial intelligence tool developed by Stony Brook librarians that enhances how students and researchers interact with the library catalog. The system allows users to conduct natural language searches that are translated into optimized queries designed to retrieve more relevant results from the catalog.

The project places Stony Brook among three finalists chosen to represent North American academic libraries, along with University of Nebraska-Lincoln and University of Texas at Austin.

Presented by Clarivate, the Library Innovation Awards were created to recognize forward-thinking libraries that are redefining their impact within institutions and communities. The awards celebrate innovation, measurable outcomes and the use of technology and strategy to advance research, learning and engagement.

SEARCH AI was developed in response to changing research behavior. Library staff noticed that patrons were increasingly entering full questions or conversational phrases when searching the catalog, a pattern influenced by the growing presence of large language models and conversational search tools in everyday digital platforms.

Traditional library systems work best if users have familiarity with keyword searching, Boolean logic and controlled vocabulary. For many students, that can make it more difficult to find relevant sources.

SEARCH AI was designed to reduce that barrier by allowing users to describe their research topic in plain language while the system generates a structured search query behind the scenes. The tool expands a user's topic, constructs Boolean search strings and applies filters such as date ranges, material type and peer-reviewed status to help refine results.

"When I joined Stony Brook, I made building AI capacity a priority for the Libraries. I have been working in AI for more than a decade, well before the rise of large language models," said Karim Boughida, dean of University Libraries. "What made this possible is a culture where staff and faculty feel empowered to innovate."

SEARCH AI was developed by Matthew Hartman, senior lead for delivery services and library applications, and Timothy Kohn, resource sharing manager at Stony Brook University Libraries, with support from other colleagues.

"SEARCH AI grew out of a grassroots idea by Matt Hartman and Tim Kohn, who did an outstanding job in a short period of time addressing a practical challenge: helping users search the library catalog in natural language," Boughida said. "While vendors are introducing AI research assistants, we wanted to build something we control ourselves, an application layer that can sit above existing systems and integrate with multiple library platforms."

The system was built during the summer of 2025 and released in beta form on October 1 following internal testing. Since its launch, the tool has been used more than 3,000 times, demonstrating significant early engagement from the campus community.

Usage data suggests that the tool is already influencing research behavior. During the Spring 2026 semester, SEARCH AI usage reached roughly half the level of traditional advanced search activity despite limited formal promotion.

The tool has also improved search outcomes. Library data shows that SEARCH AI reduces zero-result searches by about 75 percent compared with basic catalog searches. The system also applies facets and filters three times more frequently than users typically do on their own, helping researchers refine their results and locate relevant materials more efficiently.

"We are seeing adoption from our campus community and students are reporting that they find the tool useful," Hartman said.

The system also provides librarians with a data dashboard that offers insights into how patrons interact with the catalog. It allows staff to analyze query patterns, monitor usage trends and identify opportunities to improve the research experience.

SEARCH AI has also been incorporated into the library component of select writing courses during the Spring 2026 semester. As students use the tool in their coursework, their feedback is helping to make user-driven improvements.

The software is open source and available on GitHub, allowing other libraries to adapt the system to their own discovery platforms.

"We are also committed to building AI tools using open source approaches so that other libraries can benefit from and build on this work. I am very proud of the team," Boughida said.

Since the SEARCH AI tool is built within the existing catalog rather than a separate system, the development team ensured that SEARCH AI could integrate smoothly with the current catalog while also allowing flexibility for future upgrades.

Finalists for the Clarivate Library Innovation Awards will be recognized at the 2026 ELUNA User Group Conference in Los Angeles on April 27. The overall winner will be announced at the conference, where finalists will also have the opportunity to present their work to peers from libraries around the world.

- Beth Squire