Guest Post on AI Supremacy

AI-Responsive Education: LLMs and Instructional Applications, Challenges, and Recommendations

Link to Guest Post on AI Supremacy

Hi, Everyone,

I just wanted to send out a note of appreciation to the open-minded, intelligent, and diverse community that has grown up around my interest in safe, effective, and engaging applications and integrations of generative AI in today’s classrooms.

It is your interest and feedback that fuels this collective enterprise. I cannot tell you how special it is to receive a comment or a question from one of my readers. Small gestures such as these keep Substack writers pushing on into the next newsletter.

This week, I have been working hard on revising some material I wrote over the summer—a longer article on LLMs instructional applications, challenges, and recommendations designed primarily for K-12 teachers. I spent many hours on this material in June and July, but AI moves so quickly, that when I picked up the draft again in early September, a substantial part of it needed to be rewritten in light of recent advances and developments.

At around the same time in September, Michael Spencer from the Substack AI Supremacy reached out to me about doing a guest post. AI Supremacy is one of the most popular, online AI newsletters with many thousands of subscribers, so I knew I needed to send him my best material. Michael writes 2-3 original articles per week and releases his signature A.I. Generative Brief usually on Thursday, an exhaustive collection of each week’s AI articles, academic studies, business transactions, etc. Incredible stuff. Well worth the price of admission! Thanks, Michael, for all the work you do!

I had toyed with the idea of sending out my summer piece to academic journals, but my drafting and re-writing experience let me know that by the time my work went through a peer review process it would be woefully out of date.

So I seized on Michael’s generosity as an opportunity to get out this material to educators, administrators, researchers, trainers, and parents while the information was still useful. Hopefully, you, my Educating AI readers, will find it useful too.

Link to Guest Post on AI Supremacy

In this piece, you will find me advocating for some recommendations for K-12, including the continued cultivation of un-assisted AI spaces. I realize that some folks in ed-tech and ed-training regard this position as “doomed to fail,” “reactionary,” etc. But I think we may be failing under the sway of a false binary when we resort to such language. It should be obvious to anyone after reading my piece that I definitely want to integrate AI into today’s classrooms. That said, I also want to have a thoughtful conversation about how we do so.

---

Envisioning the How: An Initial Hypothesis

One, I believe all students, even our youngest students, will benefit from task alternation—writing unassisted by AI, and then, writing assisted by AI—becoming stronger writers through quality/content/style comparison.

Two, I believe that time of use with AI applications in the classroom will likely unfold in a scalar fashion across the development spectrum. Our youngest learners will need more unassisted time in order to master the basic skills, competencies, and literacies that will one day enable them to be full participants in the AI revolution.

Three, I believe that case-use will grow in depth and complexity across the developmental spectrum as students master core skills, competencies, and literacies that they can then convert or translate into productive and critical processes when engaging with AI system.

---

Needless to say, there is much more to say here. I will be writing up a post for each Principle and Recommendation included in Section 4 of my article over the next several months. So stay posted…

And once again, thanks for being part of this journey as we work together improve educational outcomes with AI.

Let me know if there are specific issues you want me to focus on.

Nick from Educating AI