Throughout the summer, I’ve been focusing on how AI should be best used on our campuses. The conversation first looked at existing risks and then also looked at guiding principles to set direction for teams when using AI.
I’ve said for some time now that I think 2026 and 2027 will be the years that governance comes to the AI conversation on our campuses. Having lived through the web and social proliferation, the first few years were wide open and then guardrails were established. Following that same timeframe, 2026 and 2027 will likely bring a new wave of guardrails.
The two main approaches to governance are guiding principles and operational strategy. I shared about guiding principles a few weeks ago, and today I’ll be looking at operational strategy.
Today’s Partner
This content is proudly sponsored by Enrollify.
Understanding Operational Strategy
When looking at AI, operational strategy makes the shift from what we believe to what we do in situations. Operational strategy helps to clarify how we use AI (or other tools) on a daily basis and in what situations. It’s a more tactical approach, as opposed to the values-based approach of guiding principles. Additionally, it’s rooted in experience, so it will be based in specific use cases that are important to the organization. The key to operational strategy is specificity in how AI is used. This means the operational strategy is going to vary more across campuses, while guiding principles look fairly similar.
I did a deep dive into operational strategy at dozens of universities and found three prevailing approaches. While the approaches are fairly different, the main similarity is that each clarifies when AI use is appropriate.
Examples of Operational Strategy
In the first approach, the university is divided into institutional hierarchy and use is clarified at each level. In this approach, university-wide uses look different than uses in a particular department. Additionally, appropriate use in each course is also clarified. Trinity University has this kind of approach and you can see how each level looks slightly different in the example below.
I like this approach because it offers a broad approach to AI use at the institution, while still giving academic leaders the opportunity to define use in their particular courses. I think this is a model many higher education institutions will take.
Another approach is defining the type of work that is being completed (whether by a student or an administrator) and then articulating what level of AI use is appropriate for that particular type of work. The University of Houston example below shows how this approach looks.
I like this approach because it acknowledges the situational nature of AI. AI use is nuanced, and this approach better accounts for that. As a leader, I can see this being something that many MarComm and other units could use to articulate AI expectations on projects and use this in a project brief.
While a little more detailed, I also like how the U.S. Department of Commerce is thinking about operational strategy in the example below. They have reviewed the specific tasks in a project lifecycle and then listed the appropriate AI activities in each project phase.
As higher education becomes more sophisticated in its AI use, I can see this model being built out further because it offers a great deal of specificity about AI use. I could see this being used as a way to onboard new projects or even to consider new tools and technology.
Looking Ahead
This post examines a few examples of operational strategy that higher education institutions can use in their work. Next month I’ll dig into using decision frameworks to help make decisions about AI and what is appropriate.
If you don’t want to miss that post, be sure to subscribe, so that post goes directly to your inbox.
Governance in AI is important to help manage innovation in a way that aligns with institutional values, and operational strategies are one way to do that.