Back in September, I had a feeling that this would be the year where everything starts to change around AI, where the inevitable would become, well, undeniable—in a way that every classroom would have to respond to.
Mine included.
In fact, this is what I wrote in June 2025: I really don’t think that my own “status quo” of avoiding AI as a tool is sustainable going forward, both in terms of the impending tidal wave of AI adoption that is already sweeping across education as well as the broader social integration of AI into so many facets of our lives outside of school.
Despite how good I felt about my classroom, I stepped into my 14th year of teaching very much ready for my own norms as a high school English teacher to be uprooted by metaphorical tidal wave that has been AI’s supposed impact on education.
Looking back, even in my syllabus explainer for students and families on expectations around AI and technology, my trepidation was clear.
“I’ll try to learn from other teachers, I’ll try to be open if anything changes,” I recorded at the time. “I also want to make sure we’re being open-minded and thinking about the world you’re moving into.”
At some point, despite my continued skepticism around AI that walked hand-in-hand with sincere concern, something had to give, right?
There were many signals that I was right. The uptick and normalization of AI usage by students and teachers alike is undeniable at this point, not to mention the institutional partnerships that have signaled a sort of formal momentum. Anecdotally, I’d add that now I had students who were much more open about their usage of AI, both with guidance from teachers and also on their own for homework.
However, as the year proceeded, that “tidal wave” that I anticipated last year in talking with Brett Vogelsinger? The one that poll after poll suggested was happening?
In my own classroom, I didn’t feel much of it. Hardly at all, really.
Aside from a few minor pivots, the ship that was our classroom maintained its trajectory—and we got to some really cool places, including with classroom community and student writing.
Despite all this good, though, an insecurity lingered.
Am I just enjoying a fool’s gold as a teacher in the eye of this AI storm? Have I become that teacher with my head thoroughly and stubbornly stuck in the sand?
Which is why, once again, I set aside time at the end of year to talk with students about AI and what role it should have in my classroom.
This post, once again, is about what they said.
How I Asked Students for Feedback This Time
Last year, I initiated this conversation around whether to integrate AI tools and opportunities into the classroom by asking students to evaluate my choice not to include them as part of our 11th grade English course.
This year, I made a subtle shift in offering a series of forward-looking questions—based on your experience, what would be the best choice for Mr. Luther to make next school year?
I also made one other adjustment: whereas last year I had groups come to consensus on these choices and submit their votes collectively, with each group getting a single vote, this year they discussed collaboratively but then voted individually. This allowed students who may have been “outvoted” in their group to still share their own viewpoint anonymously.
To be very clear, I was also quite intentional about presenting both “sides” of this choice with equal emphasis and sincerity. And as I explain in the video above, this was the third potential choice presented to them, with the first two resulting in a near 50/50 split as far as what to do going forward.
Our process for each of these questions in this activity:
My presentation of the options and my own potential reasons for each path
Collaborative discussion in small groups about how they felt
Group voting whole-class, with each group getting a single vote—before I offered some follow-up questions to better understand their positions.
I presented the question regarding a potential integration of AI tools and opportunities, discussion sparked across the room, and then eventually I brought the classroom back to share their group’s vote one at a time.
And almost immediately, it became clear that this would not be a 50/50 split.
🗣️ “I feel like AI defeats the purpose of this class.”
“Definitely stay away.”
This was the first group’s emphatic vote, and the trend continued as I moved through the classroom.
Definitely stay away. Probably stay away. Definitely stay away. Definitely stay away. Probably stay away. Definitely stay away.
It wasn’t just the results, however. The vibe, as they say, was clear. Decisively so.
“I feel like AI defeats the purpose of this class,” one student shared, with many classmates nodding in agreement.
Look, I’ll get to the numbers later on in this post. But there are a lot of people making decisions about AI who are not in classroom spaces and are definitely not having conversations with students about their decisions—so I just want you to imagine for a second what it feels like to have group after group after group echo the same advice to you as a teacher.
After you had just told them how much you valued their feedback.
Over three separate classroom conversations, there were a couple outlier groups that “probably would bring in AI, at least a bit”—but despite another year of increased AI awareness and usage more broadly by students, the overall results very much echoed last year’s conversations.
The message to me was very clear: Definitely stay away.
Still, I reminded them that the reason we had an individual, anonymous Google Form was to give them a space to share their independent perspective—no matter the conversation or vibe of the classroom.
And while I felt confident about what that feedback would look like, I remained patient to see what students would say on their own, names removed, about the best direction for our course going forward regarding AI…
🗣️ “I think we should all tread lightly with AI.”
Unsurprisingly, the individual results very much paralleled what I had already seen and heard and felt across our conversations.
“Definitely stay away” outperformed all the other options combined, and overall 81% of students were against the idea of integrating AI opportunities into my English courses going forward—with only 6% open to such a shift.
Here are some of the notable things students said in their explanations:
“ I don’t think looks good even though it’s for educational purposes.”
“It’s important to make students do things themselves because it brings them to acknowledge their own ability.”
“While there are some benefits, they are outweighed heavily by the cons.”
“It takes the creativity and critical thinking skills out of the matter.”
Another thing I noticed in reading through the feedback: while the numbers were similar to last year, the rationales and explanations had shifted rather substantially: [1] students were much more likely to acknowledge different ways it could be used, such as brainstorming or revising, even if they ultimately didn’t think it belonged in our course; [2] there was also a much-more-vocal opposition to AI, with the word hate being used several different times; and [3] overall there was a clear both/and of students recognizing the potential utility as a tool to get things completed and the potential, greater downside of the learning that would be lost as a result.
I believe this tracks with nationwide trends from students and young adults: exposure to and usage of AI is increasing, including in education spaces, but—as Angela Duckworth shared in this interview—”a lot of them have deep ambivalence about it. Many think it’s not good for them, and many worry it’s making them lazy and stupid.”
I repeat, for the folks not in the classroom: students do have more exposure to it and understand it more. They’re using it more, too! But many don’t feel good about what it’s doing—especially to their learning experience.
“I think we should all tread lightly with AI,” one student wrote in their feedback.
I think that might be an understatement.
🗣️ “…I am annoyed by its use.”
To extend the scope of this post a bit further, we also had an end-of-course seminar in my 10th grade classes that included the reflection question in the image above: what should the role of AI and technology be in English classes and school overall going forward?
For context, this group of students had already moved through their own research project earlier in the semester on the topic and participated in one of my favorite stretches of the school year with science fiction and AI conversations.
To lead off our “open round” discussing their end-of-semester views on technology and AI, I asked them a simple question: in your education going forward into your second half of high school, do you think there should be more, the same amount, or less technology within your classroom experiences?
Quick, silent reflection to record their answers and then, one at a time, each student in the circle shared their response.
“Less,” the first student said.
“Less,” the second student said, as did the third. And fourth and fifth.
Less. Less. Less.
I think you get the point: I did this activity once again across three different class periods of English 10, and in all cases “less” dominated the classroom, with only a handful of answers varying from it.
The reasoning they offered was intriguing, as the rationales behind saying “yes” ranged from student to student, class to class. There were some worthwhile exchanges, too, including one interesting back-and-forth between two students: one who pointed out how they found AI really helpful to create study guides for other classes; whereas another pushed back on this, suggesting that creating their own study guide would be even more beneficial.
Still, I do not think you can fully dislocate the “AI conversation” from the broader conversation around technology’s impact on students—many of whom understand that they’ve been over-inundated and that, if they’re being honest, the most meaningful learning experiences frequently happen when their personal devices are closed and away.
To be clear: these sophomores were not saying that all devices should be banned—you should hear them go off about Yondr pouches!—but instead that the pendulum has swung gone too far, with AI and otherwise.
As one 10th grader remarked in their own independent-and-anonymous survey, “I am annoyed by its use.”
It was not that they were morally opposed to AI or wanted to abandon technology completely. They just recognize it for what it is right now: something that far more often than not gets in the way of learning rather than elevating it.
🗣️ “…this class isn’t the space to learn it.”
In reviewing all the student feedback once more in preparing this post over my first week of summer break, I tried to nuance what I was seeing compared to the results from last year—even if the top-line numbers at first glance seem similar.
What makes this time different?
These are my four main takeaways at this point after these generous, important conversations in our classroom:
Students understand AI more. That doesn’t make them like it more. I had wrongly assumed going into this year that more AI exposure and experiences would increase its popularity with students, but everything about these conversations and surveys suggests the opposite. “If I used AI on my work,” one student shared, “I would feel way less accomplished.” As another student who was actually quite positive about AI advised, “Don’t use shortcuts until you actually know how to confidently do it yourself.” Overall, students tended to describe AI as a shortcut around learning far more often than something that actually helps them learn.
The “AI is inevitable” argument doesn’t mean all classes need it. I was quite open with them about the idea of preparation for college and the workforce as an argument for AI integration going into these conversations, and that was largely rejected: “If someone wants to go into a job that requires AI knowledge, this class isn’t the space to learn it.” While students remain open to targeted AI literacy opportunities, that is very different from the expectation that AI is integrated universally and used across all content areas and learning experiences. Based on everything I heard, that would be a mistake.
There is a growing group of students strongly opposed to AI. As I noted above in the video, one of the biggest surprises for me was the vocal opposition from a group of students around any AI usage at all—something that was almost entirely absent last year. I think this is important for anyone pushing for school-wide integration, as the contingent of staunchly-opposed students—and also, I presume, student families—is only growing more solidified and vocal. (And to be clear: I believe their positions are based on quite-defensible convictions.)
My final takeaway? The solution to skepticism about AI is not more AI.
Real talk: there are a lot of people talking about AI in education who have, in some way or another, aligned their own personal incentives with AI’s continued expansion into classrooms and schools. Success, then, is not measured by what is best for student well-being and learning. It is measured by how many students are using this technology, by how many school districts have implemented it across all classrooms.
No matter the consequences for students and their learning.
It is within this dynamic that people will look past the staggering lack of evidence of AI’s efficacy for learning, within this dynamic that there are inevitably folks reading this piece now who will deny what students are actually saying and think to themselves, “once they know more about AI, then they will appreciate it!”
If that’s you and you’re reading this, I need to write this as clearly and emphatically as I can.
You’re wrong.
While I recognize the incentive structures at play right now, at the end of the day, we owe it to students to actually make choices based around what is best for their learning and well-being.
We do not have evidence that across-the-board AI implementation will lead to that, and, at least for me, this feedback from students reiterates the skepticism we ought to have about the impact further AI integration will have on student learning.
No matter how incentivized some might be to look past that skepticism.
🗣️ “I personally think it would tear down the value of the class.”
This summer, I’ve been trying to get back into my running routine after a way-too-busy stretch for the final spring of the school year.
On a recent run with legs beginning to resist and the late-morning heat beginning to rise, I reached the classic runner impasse: ignore my aching legs and keep going a bit further or call it a day and turn back towards home? Conveniently, this impasse happened to be right where the sign pictured above was posted.
Life does, it seems, quite-literally offer sign posts every once in awhile.
One year ago I was open about another impasse: how I felt in the classroom as an English teacher around AI, believing so much in what was already-good but also, as much as I resisted the term, feeling the pragmatic weight of inevitability.
Thankfully, I listened to what my students said last year and, with absolutely zero integration of AI as a tool or skill, had another phenomenal year in the classroom—both in terms of learning and meaning.
A few weeks back, I asked students yet again about what to do around AI integration as a teacher with our course; and yet again they reaffirmed my own choice as the right path forward for future students in our classroom. They wanted to pay forward the learning and meaning they experienced over the school year, and they believed that would only happen if I continued to keep AI at arm’s length as much as possible.
“I personally think it would tear down the value of the class,” one student included in their feedback, “Being able to do things authentically is very possible.”
Students can be sign posts, too.
Especially when choosing a direction going forward.
(The question remains, though: are we paying attention?)
Once again, I want to end by sharing out other writings that I found meaningful in the past week—and this week, I selected four pieces related to AI since, despite my deepened convictions at this point, I also continue to believe in the value of learning from the voices of those embarking in different directions:
Writer Come Home by Aadya Gattu and Jennifer Tannous – If you haven’t read this piece on The Important Work, drop everything you’re doing and go read this piece. (And it’s definitely related to today’s post here, too!)
The Battle Over A.I. in the Classroom from The Daily – This episode was almost entirely focused on Mike Taubman and Scott Kern’s work charting an AI Literacy course at their school—including conversations with students!—and, in my eyes, it offers a much more productive and positive path forward with AI in education. (Also: you can follow Mike’s work around this with AI Waypoints!)
How ISTE turned EdSurge into slop by Stephen Noonoo – “AI doing the work formerly done by resourceful journalists, and coasting off the reputation they built, is anything but nuanced.” Stories like this are becoming all too common, and there are consequences for this direction in education journalism.
Sharpen AI Skills While Kids Learn New Words by Brett Vogelsinger – Brett continues to lead the way with tangible examples of what integrating AI as a tool into his ELA classroom looks like. That may seem like an odd thing to include given today’s post, but you never know when the landscape changes and I want to make sure I’m ready in that scenario, too.
Great post, Marcus - been looking forward to this one!
My experience has been very similar. I think many people underestimate how ambivalent students are about tech, how much many of them dislike both AI and ed tech, and how much many of them dislike AI as ed tech specifically. Thinking about things that I've seen/heard from my students (for context, I teach at an urban technical community college with a majority BIPOC student body):
- My students generally see the big-picture impact of AI in negative terms: job losses, environmental damage, slop, deepfakes, surveillance tech, etc. I've been struck by the degree to which my students are aware of the problems with e.g. facial recognition technology.
- A lot of this is very strongly felt. For many students of color especially, the use of AI as surveillance tech is very personal. For many female students especially, the use of AI to generate deepfake porn is very personal. For many students who are invested in art and creativity, or even in different fandoms, the use of AI to create slop art (etc.) is very personal. Etc.
- My students mostly don't have a lot of love for ed tech in general. It seems to be something that they experience more as a hurdle than a help, and the more something is mediated through layers of technology, the less likely they are to engage/the more likely they are to view it as busy work. (And the more they view it as busy work, the more likely they are to hand it off to AI.)
- My sense is that this is especially true of students who experienced online school during COVID. Students have told me how they felt disconnected, even abandoned, when school was online and how they essentially checked out. There's a pretty deep skepticism that more tech > more learning.
- This is *not* to say that my students don't see problems with schools as they have experienced them! To be frank, I've heard some pretty grim stuff from students about their past educational experiences. Not just lack of resources (though that, too), but a culture driven by compliance and surveillance, with a lack of trust, respect, and meaningful learning.
- The thing is, my students tend to see tech as a part of that, rather than a solution to that. To give just one example, I remember talking with a young Black man about his experience in high school and how college was going for him. He said, "You know one thing I like about this place?" and pointed up to the corner above the door, where something, from his perspective, was notably absent. "No cameras."
- Against this background, all of the energy that's going into AI-oriented surveillance is having a really negative effect. AI checkers, browser lockdowns, keyboard trackers, requirements that students be on camera when they do online assignments. Students hate this stuff, and understandably so. They recognize that it's happening in response to AI, and my sense is that it's part of what drives their negative sentiment toward AI, but it also corrodes their relationship with their teachers/professors.
- Overall, my sense is that students' views on AI reflect an experience of alienation. They connect AI in the world with a larger experience of dispossession and disempowerment. They connect AI in education with a larger experience of disconnection and distrust. And when they use AI for school, the fact that they're using it and the way that they're using it largely reflects their alienation from their education. That's not to say that I don't have students who are AI enthusiasts, because I do, typically maybe one a class. But for the great majority, as you say, the fact that they may use it in certain circumstances doesn't necessarily mean that they like what it's doing to their education.
- In spite of all this, most of my students really care about their education. They hope it will make a meaningful difference in their lives, but they also want it to be a meaningful experience in itself. In teaching classes where students have a lot of latitude to decide what they want to write about, I've been struck by (a) how many students make decisions that are driven more by personal interest than career preparation and (b) how many students choose to write about education. They care, and they want to see that we care. And your students clearly do!
Thanks for sharing these insights. Interestingly, in my class that is all about AI, I'm also finding that the more students know about AI, the more questions/concerns they have.