Fourth-Grade Math Question Shows AI Classroom Stalemate

What can – A study convened 17 teachers worldwide in fall 2024, revealing an “air of indifference” to generative AI in classrooms: many use tools for lesson planning or drafts, but keep students away from using them for research while still struggling to answer a simple

A fourth-grade teacher started with a question that sounded almost too plain for the noise around generative AI: “What can I actually use this for in math?”

It was not a rhetorical move. It was the opening line of a larger classroom problem educators are still working through: generative AI tools like ChatGPT have spread quickly. but the “use case” in everyday instruction has not caught up. Over the past several years. schools have been urged to respond to generative AI with limited information and a lot of hype and horror stories—some people calling it transformative. others insisting it will do harm. In many classrooms, adoption has been slower and more selective than the surrounding excitement suggests.

That hesitation, the teachers involved in a recent research project argue, is easy to misread as resistance. Their view is different: teachers tend to behave like professionals—testing whether a tool solves a real problem. For many, though, the answer is still unfinished.

In fall 2024, EdSurge researchers facilitated discussions with a group of 17 teachers from around the world. The group included teachers from third grade through 12th grade. Some of the participants designed and delivered their own lesson plans, teaching with generative AI or teaching about it. Across the conversations, several themes emerged. The most common sentiment was not excitement or panic. It was indifference—teachers using AI in some places, but not reorganizing their classrooms around it.

One fourth-grade math teacher tried to use generative AI in instruction. but before adoption she asked how it could help elementary students learn math. That question echoed what 2024 data from the Pew Research Center found: educators were split on whether student AI use was more harmful than helpful.

A high school computer science teacher from Georgia put the emotional cost of this mismatch into stark terms. worrying that schools and families were treating AI like magic. “One of my biggest fears is actually Arthur C. Clarke’s rule: any sufficiently advanced technology is indistinguishable from magic…we have students. parents. and teachers looking at AI as if it’s magic.” A high school library media specialist from New York described it as a different kind of pressure. saying there was “a fear about not being able to keep up with how things progress…the new tools and the impact it has on education.”.

Those worries land harder because generative AI did not enter schools through the usual channels. Schools typically adopt new technologies through deliberate cycles of experimentation, professional development and evaluation. Generative AI. by contrast. arrived through a different pathway: consumer tools became available to teachers and students simultaneously. often before schools developed policies or instructional frameworks for using them. That left educators trying to understand the implications while the tools were already being used.

In the day-to-day work of teaching, the first wins were immediate.

Multiple teachers described using generative AI for workload—especially lesson planning and administrative text. An engineering and computer science teacher in New Jersey said she had a routine “running discussion” with colleagues about using AI for lesson planning. adding. “I use it routinely to lesson plan. I don’t really use the lessons. but we have to produce all this stuff for admin that no one reads… AI will just roll it off.” Another teacher described similar experimentation across colleagues: “It’s really great that so many people have kind of scratched the surface and are using it to support their productivity and efficiency… lesson planning and newsletters and stuff like that.”.

Teachers described chatbot help as a productivity and efficiency tool. and the same picture appears in broader research and survey data. The article notes that recent research and national survey data from RAND’s American Educator Panels suggest teachers are adopting generative AI primarily as a productivity tool rather than a core instructional technology. That aligns with what these teachers said about early experimentation.

But administrative convenience is not the same thing as deciding how students learn.

When teachers thought about introducing AI during class time. they described a shift in the key question—from whether AI can save time to whether it solves a student learning problem. “What student learning problem does this tool solve?” is the calculation many educators are still trying to answer. even after several years of exposure.

Some teachers have limited AI use to specific roles. A science teacher from Guam said that students “write a first draft and then feed it into ChatGPT for a second draft… but I push them not to use it for research.” Others designed instruction where AI itself became the lesson. A high school special education teacher in New York shared how she removed the “veil” from the chatbot experience. “We purposely trained [a chatbot] wrong. so students could understand the data is only as good as how and who trains it.”.

In these designs, the technology is not presented as an authority. The participants in the discussion did not attribute AI as a source of authoritative knowledge. Instead, AI became something students analyze and critique. The approach reflects a broader idea from learning science research that students benefit most when technology supports reflection and revision rather than replacing productive struggle in critical thinking and problem solving.

That is why many educators see AI literacy as the most realistic entry point.

Teachers in the discussions said AI literacy feels more appropriate to teach because it matches what students are already encountering outside school. International guidance from UNESCO and the OECD increasingly frames AI literacy as a foundational skill. The guidance encourages schools to help young people understand how algorithmic systems generate information. rather than incorporating AI tools into everyday classroom tasks. The participants described classrooms that treat AI less like a shortcut and more like a window into how digital systems generate knowledge.

An elementary teacher from New York state said her focus starts with helping students understand how systems produce information and where they fail. “For me it starts with literacy — [teaching] students how to prompt. and then how to fact-check the information that’s generated to make sure there’s no bias in it.” A middle school teacher from New York described using a simple exercise about making the best peanut butter and jelly sandwich to teach how machine learning works: “We used an exercise about making the best peanut butter and jelly sandwich. The ingredients were the dataset, the procedure was the algorithm, and the output depended on how it was designed.”.

Across these classrooms, teachers did not ignore the stakes. They talked about hallucinations, bias, and trust.

An elementary library media specialist from New York said. “You ask ChatGPT to write a paper on something and it makes something up totally imaginary.” A high school French teacher said she had tried ChatGPT and found it useful only when students—or writers—already knew the content well: “I think it’s very useful if you know your content very well. IIf you don’t know your content, it’s hard to tell whether or not it’s accurate.”.

Another teacher connected these issues to algorithmic bias and the fear that students will rely on tools without understanding their limits. A high school computer science teacher in New Jersey said she works at a school with large populations of African American. Latino and Black newcomer families from African and Caribbean countries. When they talk about bias. she said. “we look at hiring data and incarceration data… and facial recognition systems where error rates vary depending on who the system is trying to recognize.” In these contexts. she described AI not as a quick answer engine but as a case study of how technological systems shape information.

Taken together, the conversations paint a stance that public debates often miss. Teachers are not rejecting generative AI wholesale, but they are also not redesigning classrooms around it. The posture described is pragmatic indifference: “I use it for lesson planning… but I don’t really use the lessons.” And: “I push students not to use it for research.”.

They use AI where it saves time while keeping boundaries around core learning tasks. That boundary matters because schools exist to create conditions for students to do complex cognitive work—deep reading. methodical writing. reasoning through problems. and evaluating evidence. If a tool primarily reduces the need to do that work. teachers have reason to question whether it advances or undermines learning.

And that returns to the fourth-grade question, asked at the beginning of these discussions: “What can I use this for with fourth-grade math?”

If the instructional use case for AI remains unclear, the deeper conversation becomes what students should be learning instead—skills built for judgment, evidence, and thinking that doesn’t disappear just because the tool is fast.

generative AI ChatGPT teachers AI literacy classroom instruction lesson planning education policy learning science hallucinations bias