Using AI for Learning Coherence, not "Another Thing"

One thing I hear during Professional Development with teachers is the feeling they are always being swung into to the next new thing. This sounds something like, "We used to do blended learning, but then we did social emotional learning, then culturally responsive learning, and now we're doing X." This initiative overload, as we refer to it, leads to feelings of frustration, fatigue, and confusion. What it doesn't lead to is effective change. In this situation, pockets of innovation emerge at best, but more often, teachers wait it out until soon enough, well...they are on to the next thing anyway.

It's not that leaders are doing the wrong thing by introducing critical practices that improve student learning, but sometimes there is a lack of coherence in both the messaging and methodology. For PD and implementation of new tools or practices to be effective, we must connect the dots. At LINC, we specialize in connecting these dots so that teachers can easily see and maximize the intersectionality such as, when you are doing blended learning, you are creating more time for 1-1 and small group instruction with students, thereby naturally integrating SEL moments into the academic flow.

In the past year, we've all seen the wave (tsunami really) of the next new thing arrive with generative AI. In our work, that sometimes means we've heard less of, "LINC, we need you to help us implement blended learning effectively" and more of, "LINC, can you help us implement AI?" I would like to say, these things are not different. In fact, they are extremely connected because the highest and best use of AI can remove what was previously the heaviest lift - building learning coherence in the system.

So how can AI build learning coherence? First, we start with the "why" of what we are trying to accomplish. Most often, when we ask leaders what impact they are driving towards, they say some variation of student-centered (or "personalized") learning that meets the different needs of students within their culturally and linguistically diverse classrooms.

Once we understand this "why", we can design a system of supports that drives towards it, and those can be made wildly more efficient by generative AI. For example, we know that to plan for high learner variability, teachers need to leverage blended learning. We also know that one district's portrait of a learner is different from another's, or that one district uses this set of standards while another uses a different set. Each school or district uses different curriculum.

Therefore, AI cannot be one-size-fits-all. The use of generic, conventional AI teaching tools will simply scale conventional, teacher-centered learning which is what most systems are striving to move away from. Efficiency is not pedagogy.

And it is certainly, it is not pedagogy built to drive a specific vision of teaching and learning. We need to build custom AI tools that ingest each school or district's unique standards, values, or resources, and combine them with student-centered methodologies like blended learning in order to have all the "things" synthesized. This is the opportunity in front of us - to move beyond pockets of innovation and drive system coherence for student-centered, highly differentiated learning. In our work customizing and implementing Yourwai driven AI, we're already seeing the impact of the combination illustrated below:

As we approach AI capacity building in this moment of high demand, we need to do so with this learning coherence in mind. Otherwise, what we will see is AI as another learning fad which goes out of style when the next thing comes along, and along the way, many frustrated educators experiencing little positive impact in their classrooms.