Earlier today, I saw a linkedin post by Code.org's Pat Yongpradit summarizing a series of articles on AI in K12 classrooms in Ed Week. I couldn't get far into the articles due to paywall restrictions but Pat's summary and the ensuing comments and discussion let me to writing this.

One thing that stuck out in the comments, and this has been a recurring theme in AI in K12 classroom discussions is that a teacher who allows students to use AI, teaches AI, etc. will be more effective than one who doesn't.

While I'm against an AI ban and if I were still in the classroom I don't doubt that I'd be exploring the tools, it's not that simple.

Let's look at elementary math education. When I was in grade school in the 70s, we learned our math facts, long division, and all sorts of things that were positively uncool when my kids were in school in the early 2000s. They didn't learn there math facts and teachers had very different ways of approaching most of the topics. This struck me as, well, not good, so I started to talking to my students who had gone through this "new math" education as well as to parents.

While this wasn't an official study, it certainly appeared that by the time kids who were educated in this newfangled way using all the new tools and tech were by and large more poorly prepared than students who either had an old school teacher who bucked the trend and did things the old way or had parents who supplemented on their own.

How does this relate to AI in the classroom? Well, a teacher only has so much time with their students and way more to teach than can realistically be done in the time allotted. So, teacher A focuses on the new hotness. Goes all in on AI - well, that's great but something's got to give. At the same time, teacher B doesn't do the AI but goes old school - pen and paper and really focuses on teaching the kids to write - that is - communicate effectively.

Fast forward a few years and teacher A's kids are proficient in what was modern AI a few years ago. Maybe this will be a tremendous boon but maybe we've moved on to the next thing. Teacher B's kids, on the other hand are stronger communicators regardless of medium.

Who's better off? Maybe A's kids, Maybe B's.

To make things more complicated, teachers have not been educated in AT and neither have administrators and just adding more and more to teachers plates is both unjust and not sustainable.

This hypothetical has played out before - with elementary math, calculators, writing across the curriculum, CS4All and more. Over and over again. There's no simple answer. I wasn't sure that calculators were the way to go back when they came in. I'm still not and even if they were the right move, why are kids and schools still forking money over to TI when there are free and better alternatives on phones and computers? Writing across the curriculum? Great idea but nobody carved out anything from my required curricula to make room for it and while I'd like to think I could do it, I've got nothing to show in my work history or education to say that I was qualified to teach effective writing.

Yes - let's work on AI policies - hopefully informed by people who know both the tech an the teaching but let's not be so fast to say that a teacher who embraces AI is good while one who does not is bad.