“There’s a mix of trepidation and cautious optimism,” O’Brien says, based on what he’s hearing from the research and education (R&E) community.
“I worry about a long-term divide between universities who can afford AI and those who can’t,” he adds. That’s where the value of community and collective influence comes into play.
Drawing on lessons learned from cloud adoption in the R&E space, he recommends institutions take proactive steps to mitigate risk before costs spiral out of control.
Establish AI governance now, before sprawl worsens. Faculty, researchers, and staff are already moving among multiple models and tools, guided by factors such as cost, data access, and integration needs. Governance frameworks should accommodate that multi-model reality by bringing together the right stakeholders to ensure AI adoption aligns with institutional policies, procedures, and values.
And institutions don’t have to do it alone. The Internet2 community provides a forum for collaboration across institutional boundaries, where peers benefit from sharing approaches, lessons learned, and emerging best practices around AI adoption.
Build cost visibility and budget guardrails together. A centralized token pool purchased in bulk and made available campuswide can save money, but only if it’s managed carefully. Without guardrails, institutions risk difficult decisions and budget uncertainty. “Are you cutting people off?” asks O’Brien. “Where is the money going to come from, if you outspend what you planned for tokens?”
According to O’Brien, institutions that act early will be better positioned to maintain necessary oversight without stifling innovation.
ICYMI
< Back to Internet2 News