Where Does Your Institution Actually Stand on AI?

Most higher education leaders know they should be doing something with AI. Far fewer know where they actually stand.

*This content originally appeared on LinkedIn.

Three years after ChatGPT's release, AI is no longer optional for postsecondary institutions. The technology has moved from an emerging trend to an embedded reality. Yet when we interviewed leaders from 33 institutions across the country, we found a striking disconnect: while enthusiasm for AI runs high, clarity about institutional progress remains elusive.

One provost captured what we heard repeatedly: "We're still a little bit exploratory in some areas... and in others, we're more siloed where people are doing things, but not necessarily connecting all the dots... That's what happens when you run really fast... you don't take the time to say, 'Wait a minute, where are we?'"

The 17% Problem

Here's the finding that should give every president and CIO pause: only 17% of the institutions we interviewed have achieved strategic AI integration. The remaining 83% are still experimenting with small pilots or beginning to scale in targeted areas, often without the policies, infrastructure, or coordination needed for sustainable adoption.

This isn't a critique. It's a diagnosis. And it reveals something important: the challenge isn't a lack of willingness. As one leader told us, "We have a bigger issue of too many people gathering information and making moves than too few. It's not a matter of trying to get the coalition of the willing. We've already got the willing. We need to design the structures so that all of this makes a lot more sense."

The willing are there. The structures often aren't.

What We're Seeing Across the Landscape

Our research revealed three distinct phases of institutional AI adoption: Experimental/Emerging, Scaling, and Transforming. The differences between them are less about specific tools and more about coordination, governance, and strategic intent.

At many institutions, AI work happens in pockets. A handful of enthusiastic faculty experiment with ChatGPT in their courses. An IT leader pilots a chatbot for student services. An admissions team tests automated outreach. These individual efforts are valuable, but they rarely connect into a coherent institutional strategy.

What separates institutions moving toward transformation from those stuck in experimentation? It often comes down to five factors: adoption coordination, policy development, leadership engagement, resource commitment, and technology deployment. The institutions making real progress have addressed all five. Those struggling typically have gaps in several.

The Resource Reality

Our analysis also revealed a hard truth about resources and outcomes. Among the well-resourced institutions in our study, 100% had achieved transforming-level adoption. Among resource-constrained institutions, that number dropped to zero.

But here's what's important: 40% of resource-constrained institutions had still managed to reach scaling status. Resources matter, but they don't determine destiny. What matters more is knowing where you stand so you can make strategic decisions about where to invest limited time and money.

One leader at a resource-constrained institution put it bluntly: "I barely can make it through the day. So the training and all that's involved to try to really leverage this powerful tool that we are underutilizing... we can't afford it." This captures a truth many institutions face: the cost of AI adoption extends far beyond software licenses. It includes the human and time costs of training, implementation, and ongoing support.

Why Diagnosis Comes Before Strategy

Several leaders we interviewed emphasized the value of starting with what they called a "discovery" or diagnostic phase, resisting the pressure to just buy into the hype and apply AI ad hoc. One institution described focusing on fundamental questions: "Where we are" and "what do we have."

Another leader shared why this matters: "It tells you what the right problems are you need to solve."

This diagnostic approach requires restraint. When vendors are knocking, when peer institutions are announcing AI initiatives, when board members are asking questions, the pressure to act can feel overwhelming. But institutions that pause to assess their current state before committing resources tend to make better decisions about where to focus.

What's Next

On February 10, we're hosting a webinar that goes deeper into these findings. We'll walk through the adoption framework in detail, share anonymized examples of institutions at each phase, and provide practical guidance for assessing where your institution stands today.

This isn't a sales pitch for more AI tools. It's a research-grounded look at what's actually happening across the sector, and what it takes to move from scattered experimentation toward strategic integration.

If you're responsible for AI decisions at your institution and you're not sure where you actually stand, this session is designed for you.

Register for "Inside the AI Adoption Landscape" on February 10.

This blog post draws on findings from T3 Advisory's national study on AI adoption in higher education, conducted in partnership with Complete College America. The study examined AI integration across 33 postsecondary institutions, with a focus on practical guidance for institutional transformation.