2,038 students asked 8,217 questions in 37 days. The university runs the full ibl.ai stack on its own Google Cloud under a perpetual source-code license.
NEW YORK, NY, UNITED STATES, August 19, 2026 /EINPresswire.com/ — Syracuse University has published usage results from the first registration season of Clementine, the artificial intelligence platform the university owns rather than licenses. Built on the ibl.ai Agentic AI Operating System and deployed inside Syracuse’s own Google Cloud environment, Clementine sits on the university’s list of approved AI tools alongside Claude Enterprise, Microsoft Copilot and Google Gemini. On Syracuse’s own comparison of the four, Clementine is the only one that lets a user switch between vendor models in one place.
Clementine Class Search, built on Anthropic’s Claude, launched in Spring 2026 and went into Fall 2026 registration immediately. Between March 25 and April 30, 2026, 2,038 students asked 8,217 questions across 2,401 conversations. The five-day undergraduate registration window of April 9 to 15 accounted for 41 percent of all questions in the period. On April 14 alone, students sent 1,055 questions across 313 conversations. The longest single session ran 56 turns.
Two patterns in the data speak to what an always-available advising agent changes. More than a third of all activity, 37 percent, occurred outside weekday business hours, when advising offices are closed. And two out of three conversations went past a single question, with students returning days later to keep refining a schedule.
Students queried the course catalog in plain language rather than search syntax. Recorded examples include “no 8am and nothing past 5pm, and no fridays” and “build a 14 credit schedule, avoiding Fridays.”
“We wanted full ownership of our AI platform, the code, the data, the infrastructure,” said Jeff Rubin, Senior Vice President and Chief Digital Officer at Syracuse University. “ibl.ai delivered exactly that. We’re running on our own GCP, with our own models, fully integrated into our campus systems.”
The deployment reflects a procurement decision rather than a product selection. Syracuse received the complete source code under a perpetual license and stands up the entire stack within a university-controlled Google Cloud project, where the institution’s own cloud team manages access, networking and compliance as it would for any other campus system. The platform integrates with university single sign-on and maps role-based access to existing university roles, alongside a custom interface matching Syracuse branding.
Cost was a material factor at scale. Per-seat AI licensing at $20 per user per month would price a 30,000-student population at roughly $7.2 million annually before customization. Under a usage-based model, Syracuse pays for actual language-model consumption. The university has reported approximately 85 percent lower AI costs than the per-seat alternatives it evaluated.
Model choice sits with the institution rather than the vendor. The platform is built on Google Vertex AI and runs inside the university’s own Google Cloud, which Syracuse describes as giving it enterprise-grade models with the freedom to connect any large language model and swap freely as pricing changes. Of the four approved tools on the university’s comparison, Clementine is the only one offering access to different vendor models from a single place.
The clearest signal from the first registration season was students asking degree-path questions, such as whether a course counts toward a requirement. Because Syracuse holds the code, connecting the agent to each student’s degree audit is a decision the university schedules on its own timeline rather than a feature request submitted to a vendor.
Syracuse University Information Technology Services has published its own account of the deployment, the approved tool list, and the registration-season usage report at ai.syracuse.edu.
About ibl.ai
ibl.ai is an Agentic AI Operating System that organizations deploy on their own infrastructure with full source-code and data ownership. The platform is model-agnostic, supporting commercial and open-weight language models interchangeably, and is priced on usage rather than headcount. It is in production with more than 1.6 million users across more than 400 organizations. ibl.ai is SOC 2 Type 2 certified and deployable to private cloud, on-premise and air-gapped environments. The company is family-owned and operated from New York.
Miguel Amigot
ibl.ai
+1 571-293-0242
miguel@ibl.ai
Visit us on social media:
LinkedIn
YouTube
X
Other
Legal Disclaimer:
EIN Presswire provides this news content “as is” without warranty of any kind. We do not accept any responsibility or liability
for the accuracy, content, images, videos, licenses, completeness, legality, or reliability of the information contained in this
article. If you have any complaints or copyright issues related to this article, kindly contact the author above.
Media gallery
