Reflections on UCF’s “All of the Above” Framework

In Blog by Florida A.I. Learning Consortium (FALCON)Leave a Comment

Resources shared during the September 8, 2026, webinar: UCF’s All of the Above Framework: Meeting the Dual Needs of Integrating and Resisting AI

Blog By Dawn M. Hale
State College of Florida, Manatee-Sarasota

During “All of the Above: Meeting the Dual Needs of Integrating and Resisting AI,” Kevin Yee, Director for UCF’s Center for Teaching & Learning, offered a framework that makes room for what can sometimes feel like competing responsibilities. We need to prepare students for a world in which AI will increasingly be part of their work. At the same time, we need to protect opportunities for students to think, write, reason, create, communicate, and struggle through learning for themselves. Rather than asking higher education to choose between those priorities, the conversation invited us to consider how we might intentionally make room for both. Here are a few of the ideas that stayed with us and some further thoughts on those ideas.

When participants were asked to share one word they associate with AI, the responses reflected something many of us are experiencing across higher education: there is no single story about AI. And perhaps there shouldn’t be.

I am struck by how binary the responses to AI are. There is always a positive but a corresponding negative; Useful/Disruptive, Collaborator/Isolator, Overwhelming/Calming, Opportunity/Hinderance, Originator/Terminator. AI is without a doubt a real time-saver but can also be crippling to the learning process, a true double-edged sword. Even as I begin to type out these words I am aware of how frequently I turn to AI for routine tasks and the strain of bringing forth my original thoughts without my ever-present co-worker. The irony was not lost on me of how determined I was, not to use an AI tool, to post a blog about the uses of AI tools.

One phrase from Dr. Yee’s talk that stood out to me was that “Learning Requires Friction”.  AI, for the most part, reduces friction. So, a logical conclusion might be that it could reduce learning. Our students come to us expecting to be prepared to participate within their fields and to feel secure in their status as a college graduate to earn a living. As we know, there are many reasons students can use AI to lessen the load. If the Amazon layoffs, blamed on AI, have something to show our students, it is that the type of work that can easily be accomplished by letting AI complete tasks for you, is not work that is valuable to an employer. In other words, if you are the kind of student who lets AI complete the assignments that were to prepare you for the workplace, then you missed the chance to learn what is needed in the workplace.

It is not, however, just a discussion about learning to be more effective in the use of AI. The discussion spoke to leaving room for both sides, integrating, and resisting. I think this is exactly where the struggle needs to take place. We often speak of critical thinking, and perhaps AI takes critical thinking to the next level. Not only must our students think critically about the problems they face and the learning they are attempting, but they must also think about the processes they are using to engage with these problems. Our students need direct instruction about how to consider their own ways of thinking and learning. The act of engaging deeply in metacognition about their own use of AI can create fluency with the AI tool, however, AI literacy alone may create habits that leave our students as bystanders, only able to use AI to complete tasks for them.

Employers expect students who can use AI, and students who do not know how to use it may be disqualified from the workforce. Only knowing how to use AI to do your work for you could quickly lead to being replaced by that AI.

In a recent conversation with a young engineer this exact problem was highlighted. His firm invested in a proprietary AI that all employees are expected to use. However, many of the young engineers who relied only on this AI tool to complete their work were dismissed when it was discovered that the tasks being issued to them were easily completed by the AI without the need for the middleman. Those who used it to elevate their work, and to bring added value to both what the AI could do alone, and to what an engineer could do alone, quickly advanced within the company and were given responsibility over more complex projects.

In higher education both sides of the approach to AI must exist, the resistors and the integrators. For our students to thrive we need to have instructors that resist. Students need to know the boundaries and the tasks for which AI is not the solution. In this they need to know “The Why” behind resistance. Skill erosion is a real consequence of AI misuse or overuse and just getting the grade in a class using AI devalues the experience. I worked with a math teacher when I taught in the k-12 system, who continuously preached to students “Chase the knowledge, not the grade.” Those words have never been truer than right now. However, for those who are the resistors, it is not just as easy as flipping a switch. Converting back to blue book tests is likely no better at assessing student learning than allowing them to submit electronically what AI has written. The blue book approach, as noted by Dr. Yee, places students under constraints they have not been trained to address. Thus, we will not be able to assess accurately what has been learned. The AI integrated approach, however, risks placing students in a situation where those who let AI do the work for them can earn higher grades than those who create authentic work using AI as a tool. These situations can lead to both Type I and Type II errors. The error of knowing but being assessed as not knowing, or the error of not knowing but being assessed as knowing.

So, how do we create significant learning (pun intended). If we are the resistor in the scenario, it is incumbent upon us to train our students in the types of thinking that will be used in our assessment. For example, timed writing that requires quickly organizing thoughts and reproducing them on paper in hand-written responses, uses many useful skills that our students may never have been exposed to in our digital age. In many cases these are the exact skills and thinking processes that are going to be needed to thrive and use AI more effectively. Thus, the work of the resistors is essential to ensure our students are able to think in ways that facilitate AI fluency.

On the other side of this particular binary, integrators will need to ensure that our students are able to use and interact with AI effectively. If they learn all the needed skills but have not involved AI, they will not be seen as prepared by future employers. The data on employment cited by Dr. Yee demonstrates that employers expect AI fluency in their employees, but many have difficulty in describing exactly what this means. We are expected to train students who have not accomplished a degree only through AI usage, but we are also expected to have students who can skillfully use AI.  How are we to accomplish both? In addition to creating the space for resistance and informed refusal, we also need to make space for our students to learn fluency with AI.

Situations and scenarios that create friction for the students and challenge their thinking, even when AI is being deployed, lead to the kind of fluency described by Dr. Yee. The example of having students reach out to businesses and evaluate real issues provides students with a way to use AI to create authentic solutions. Authentic assignments cannot be simply pasted in, and reflections on the work cannot be easily generated. Also, assignments such as real-world problems are interesting and students are more likely to engage and not take shortcuts with the kind of work they see themselves doing in their career.  In this case, AI may have a role to play, but students walk away with genuine learning, and AI becomes a tool to address parts of a more complex overall process.

Another way in which AI can be integrated is through the creation of what Dr. Yee calls resilient assignments. Resilient assignments can be based on lectures or other artifacts that cannot simply be pasted into AI to produce a final product. Assignments may also include components that have multiple inputs and outputs, which emulates real working situations the students will one day encounter. This is far more involved than simply prompting an AI and creates real transferable skills that can be used across many fields. The act of breaking multifaceted situations into multiple prompts and responses based on AI outputs creates new kinds of learning that are needed even if skillsets such as these have not yet been identified by employers.

The overall takeaway here, for higher education, is that although AI is often met with very binary responses – YES! and ABSOLUTELY NOT! –  our reality and the reality of our students sits right in between the two. The space between yes, and absolutely not is a hard place to be and our curriculum will need to be carefully constructed and delivered intentionally to our students with an emphasis on creating transparency around our intent. We want the learning to come first, and we want their use of AI in the workplace to reflect that learning and to be effective. Thus, we need to integrate into our curriculum not only AI Fluency but AI Resistance, because increasingly in the world of AI they are less like opposite ends of a spectrum or sides of a coin, and more like components of the same organism. (Final Note: I refuse to let the proliferation of AI, and those who suspect it’s use in everything I create to diminish my use of the em dash.)

Leave a Comment