How Multi-Model Realities and Shared Infrastructure Demands Are Redefining Artificial Intelligence in Research and Education
Estimated reading time: 3 minutes
Artificial intelligence is transforming how institutions teach, conduct research, enhance the student experience, and manage administrative operations. As the research and education (R&E) community moves from pilot projects and experimentation to enterprise-scale adoption, new complexities are emerging around governance, infrastructure, and access.
In Campus Technology’s Tech Outlook 2026, Internet2 leaders Sean O’Brien and James Deaton offer forward-looking perspectives on the next AI inflection points for R&E. Ultimately, their insights signal the need for intentional collaboration to ensure institutions can keep pace — and advance together.
A Multi-Model AI Future Requires Early Governance
Sean O’Brien, associate vice president for NET+ cloud services at Internet2, predicts rapid AI model sprawl in the coming year. Faculty, researchers, and staff will increasingly move between multiple models and tools, guided by factors like cost, data access, and integration needs.
To prepare, he suggests leaning on recent lessons learned in the cloud space.
“One of the key lessons learned from research and higher education’s cloud adoption is that waiting too long to plan for multiple services creates governance, cost, and visibility challenges that are difficult to unwind later,” O’Brien said. “AI is at a similar inflection point.”
Sean O’Brien
He advocates that now is the time to apply those lessons.
“2026 represents a narrowing window for institutions to proactively establish governance, access controls, cost management, and visibility across multiple AI models. Those that act early will enable innovation while maintaining institutional oversight.”
Shared Infrastructure Can Help Close the Access Gap
James Deaton, vice president of network services at Internet2, emphasizes that meaningful AI adoption will depend on access to the network capacity and compute resources institutions need to scale.
“Much of the AI conversation in higher education has been about chatbots and content generation. That’s going to change in 2026 as institutions confront a more complex challenge: the infrastructure demands of AI-enabled research and instruction,” Deaton said.
Deaton cautions that navigating infrastructure demands in isolation risks widening gaps in AI access and capability. As a path forward, he points to the R&E community’s success in building shared regional and national infrastructure.
James Deaton
“The question for 2026 isn’t whether AI transforms higher education. It’s whether we seize the opportunity to apply our experience building shared infrastructure to collectively fulfill the AI-enabled education, research, and service missions of every university.”
Read the full story in Campus Technology, Tech Outlook 2026: What Higher Ed Tech Leaders Expect this Year
ICYMI