AI for Education: Solutions in the K-12 Teaching Experience

I started teaching in 2007, right before the slide began in education to **waves hand around** whatever this mess is.  At the time, I was convinced I knew what good teaching looked like. Students would read, we’d discuss, they’d write essays, and I’d grade them. Simple. Predictable. Control.

Almost two decades later, I’ve learned that control was always an illusion…and nowhere is that clearer than in today’s conversation about AI for education.

Here’s what I know now that I wish I’d known then: AI is not the biggest threat to education. Our “from my cold, dead hands” attachment to outdated assignments is.

I’ve spent the last six years training K-12 teachers across the country on how to implement student-centered learning, and I’ve watched the panic around AI’s impact shift from “how do we block it” to “how do we teach through it.” The schools making progress aren’t the ones with the best detection software. They’re the ones asking better questions about what learning actually looks like.

When a generative AI tool can complete your essay prompt in twelve seconds, that’s not just proof that students are cheating. It’s proof that the assignment was never measuring thinking in the way we thought it was. For years, we confused compliance with cognition…five paragraphs, three sources, quote sandwich, Works Cited page, submit, done. But if a machine learning model can replicate the product instantly, the product was never really the point.

What I Actually See Happening in Classrooms

Last fall, I worked with a school district in the Midwest that had banned ChatGPT across all devices. Teachers were spending hours running student work through detection tools, chasing ghosts, writing students up, and destroying trust. The administrative tasks around policing AI had become overwhelming.

Meanwhile, I was also working with a CTE school where students were using large language models to draft project proposals and then spending their actual class time refining those proposals with community partners. The difference wasn’t the technology…it was the design.

Even back when I redesigned my own history courses, I stopped asking “Can students produce an essay?” and started asking “Can students show me how they think?” This shift is where I tell the teachers I’m working with to begin. For instance, the first step for older students who should be writing on a deeper level is requiring process logs where students document their writing choices, version histories showing how their arguments evolved, and AI use disclosures explaining where and why they used AI-powered chatbots for help.

The irony? When students know they have to show their thinking, they actually use AI more responsibly. They treat it as a thinking partner rather than a replacement for thinking.

Why Experience Matters More Than Ever

Over my classroom teaching career, working in both a nationally ranked academic school and a Title I school, I learned something crucial: students everywhere want the same things. They want to feel capable. They want work that matters. And when given the choice between traditional methods that feel like busy work and AI-generated content that completes the task instantly, they’ll choose efficiency every time.

That’s not laziness. That’s human nature.

I’ve talked to teachers who had students crying because they got caught using AI on an assignment they found meaningless. One high school teacher told me a student looked her right in the eye and told her, “I didn’t know why we were writing it, I just knew I needed the grade.” That student wasn’t cheating the system…he was responding to a system that had taught him the product mattered more than the process.

Many times, we understand as the teacher the benefit of an assignment, but to a student growing up in the world that is (slowly imploding) around us, that benefit is lost in translation.

The learning experience changes when students understand why they’re doing what they’re doing. My post-course surveys started showing something I’d never seen before: joy. Actual joy in assignments. Students talking about how they finally understood what they were learning and why.

The Strategy Gap in AI Training

I was working with another teacher who was frustrated that her principal was adamant that they needed to train all the teachers on AI, and then showed them a ninety-minute tutorial on how to write ChatGPT prompts.

Another meeting that could have been an email, am I right?

Here’s the problem: teachers don’t need more tech training. They need strategy.

The teachers I work with can figure out how to use a tool. What they can’t figure out on their own is how to fundamentally redesign their lesson plans so AI integration actually improves student outcomes. They need frameworks for evaluating whether a tool serves their unique needs, not just lists of the latest digital tools that will be obsolete in six months.

When I lead professional development sessions, I focus on practical applications that teachers can use on Monday morning. Not theory. Not fear-mongering. Not “here’s how to catch them.” Instead, we work through questions like: What skills do you actually want students to leave with? What thinking work can’t be outsourced? How do you design assessments where the process matters as much as the product?

The school administrators who get this right are the ones giving teachers permission to experiment without fear. They’re treating AI as what it is: a permanent shift in the educational systems we work within, not a temporary disruption to manage.

And, let’s be honest, the long-term effects of AI haven’t even been studied yet…we just need to roll with the punches in the meantime because it’s here, and we can’t pretend it isn’t.

What Actually Works With Students

Most AI use is more complicated than outright cheating. Students use AI-powered chatbots to brainstorm ideas when they’re stuck, to get immediate feedback on whether their thesis makes sense, and to help them understand complex concepts they’re struggling with alone at midnight. These aren’t shortcuts…they’re additional support systems that students have always sought out. It’s a tutor without the embarrassment of asking for one.

The virtual tutors and intelligent tutoring systems now available can provide extra help that some students simply don’t have access to otherwise. I’ve watched English language learners use natural language processing tools to check their writing before submitting, building confidence and fluency in ways that wouldn’t happen if they were just guessing alone.

The question isn’t whether students will use AI. The question is whether we’ll teach them how to use it ethically, critically, and strategically. That means having conversations about academic integrity that acknowledge complexity. It means teaching students when AI is appropriate and when it undermines their own learning. It means helping them understand that the process of working through difficult thinking is the skill they’ll need in the workforce, not the ability to produce something that looks finished.

And, it means remembering that our students are human, so if there’s a shortcut, they’re going to take it. We need to help them realize which shortcuts are worth it (and which ones we should elimate all together).

The Hard Questions Nobody Wants to Ask

The ethical concerns around AI in education are real, and pretending otherwise helps nobody. When AI systems collect and analyze student data, who has access to that information? What happens when federal laws about privacy conflict with the capabilities of these tools? How do we protect special needs students whose data might be particularly sensitive?

These questions don’t have easy answers. But avoiding them while students use AI anyway isn’t protecting anyone.

I’ve also watched schools invest heavily in AI tools without investing in teacher support. They buy the learning management system upgrades, the fancy AI tutor subscriptions, and the educational content generators. But the teachers who are supposed to use these tools haven’t been asked what they need. They haven’t been given time to learn. They haven’t been included in decisions about how AI will be used in their own classrooms.

The schools doing this well are the ones treating AI as a conversation, not a mandate. They’re bringing staff members together to talk about best practices. They’re creating AI usage guidelines collaboratively. They’re acknowledging that nobody has all the answers yet, and that’s okay.

What I’ve Learned From Training Thousands of Teachers

Since 2018, I’ve worked with teachers across the United States on implementing student-centered learning. Before AI was on anyone’s radar, we were talking about how to make thinking visible, how to design assessments that actually measure understanding, how to move beyond traditional methods that had stopped serving students.

When AI arrived, those conversations became urgent in ways I couldn’t have predicted.

The teachers who adapt best aren’t the ones who are most tech-savvy. They’re the ones who already understood that their job wasn’t to deliver content…it was to design learning experiences that develop critical thinking. They already knew that student engagement comes from meaningful work, not compliance. They already believed that human interactions and relationships were the core of teaching.

AI can generate course materials. It can help with grading. It can provide additional support to students who need it. What it can’t do is build the relationships that make learning possible. It can’t notice the student who’s quiet today in a way that means something’s wrong. It can’t inspire a kid who’s never believed they could do hard things. It can’t model what it looks like to struggle with an idea and work through it.

Those remain uniquely human. Those are what we need to protect and measure.

The Path Forward

If your school district is investing in AI, make sure the investment includes time for teachers to think together. Make sure professional development focuses on pedagogical shifts, not just tool training. Make sure school administrators are asking teachers what they need rather than telling them what they’ll do.

The teachers I work with want to get this right. They want to prepare students for a world saturated with advanced technologies. They want to honor academic integrity while acknowledging that the definition of integrity might need to evolve. They want to protect student privacy without locking down learning.

None of this is easy. Recovering from the initial panic around AI and moving toward thoughtful integration is a long-term process. Based on what I’ve seen in schools across the country, the ones making progress are the ones treating this as an opportunity to finally ask the questions we should have been asking all along: What do we actually want students to learn? How will we know when they’ve learned it? And what does meaningful intellectual work look like when the tools keep changing?

I don’t have all the answers. Nobody does. But I know that the teachers who will thrive in this new landscape aren’t the ones with the strictest policies or the best detection software. They’re the ones who remember why they started teaching in the first place…and who recognize that AI’s potential to transform education is only as powerful as our willingness to transform alongside it.

This article was originally published on December 26, 2022.