Our Kids Shouldn't Be Guinea Pigs For Chatbots

Why it concerns me, an education reporter, when McMahon backs classroom AI use by referencing Alpha School

During the summer, I was invited by Taiwan’s Hualien Family Education Center to talk about fostering artificial intelligence literacy in K-12 students. I got invitations like this a lot, mainly because I was the U.S. Correspondent for Taiwan’s leading education magazine for twelve years, and also becuase, for a very long time, Taiwanese educators believed the U.S. education system could serve as a valuable reference.

But I’m not so sure about it anymore. Not with all the nonsense our education secretary, Linda McMahan, keeps babbling about when she goes on TV. 

McMahan’s recent appearance on CNN was a prime example. When anchor Dana Bash asked her about AI use in K-12 classrooms, McMahan answered by referencing a visit she made last year to Alpha School, a private school in Austin, Texas, that was co-founded by an AI enterpreneur and where the students spend two hours a day doing learning on screens.  

“The first couple of hours every day, students really have on headphones,” McMahan said. “They’re in front of their computers. They’re getting subjects fed to them, with all teachers monitoring what they’re doing. And the beauty of that is that a student almost has a one-on-one tutor.”

Bash asked if this was a “proper model,” and Mchan replied, “It’s working wonderfully,” even though she was “not saying that works everywhere.”

Hearing the woman who once referred to AI as A1, as in the steak sauce, glorify AI use in schools made me deeply uncomfortable. Alpha School’s claims of student success have never been independently verified, and investigative reports have even found faulty AI-generated lesson plans and questionable data practices. 

Investigations by media outlets like 404 Media revealed that the private school network’s proprietary tools and AI-driven lesson plans included illogical questions and inaccuratge materials. Not to mention that Alpha School is a private school; it gets to choose its own students. The school charges up to $75,000 in annual tuition, and journalistic investigations note that the school relies on self-selected, affluent student bodies and has failed to release raw data to independently back up its claims of rapid academic advancement. 

With the tuition the school charges, it doesn’t even employ certified teachers. Instead, it uses higher-paid adult “guides,” only a fraction of whom hold education degrees. It was reported that a handful of families have pulled their kids out of Alpha School campus, some in the middle of the school year, becuase their experience was miserable.

But McMahan was apparently not up-to-date on the latest research. Watching our education secerety be so credulous about the aggressive use of AI in the classroom is disturbing. 

As a former education reporter and a current schoolteacher, I firmly believe that education policy should be evidence-based. A recent study of over 26,000 Chinese students from grades 7–12, published this June by the Center for Economic Policy Research, found that while AI use raised homework scores among students, it lowered monthly exam scores by 20 percent within six months. High-stakes entrance-exam scores fall by 18 to 24 percent, with student performance worsening over time. The students who experienced the greatest negative effects on their exam scores appeared to be those who rushed through their homework by outsourcing it to AI.

Our educational leaders shouldn’t abdicate their responsibility to stuednts, teacher, and parents by allowing Big Tech to dictate national policy on screens in public schools. In a New York Times feature story, reporter Natasha Singer wrote:

Tech companies finance national polls and research reports to help make products like A.I. chatbots, and courses like computer science, more urgent priorities for students and school districts. They advocate for new laws to require schools to teach industry-aligned subjects. They sit on committees that help shape learning standards for public school students. They supply or sponsor teacher training and free course materials for students. They fund the National PTA, teachers’ unions and other prominent groups.

The good news is, as I write this article, two of the nation’s largest school systems are drawing a sharp line around AI: New York City has declared a one-year moratorium restricting student-facing generative AI from pre-K through eighth grade. Los Angeles is blocking student use of generative AI on district-owned devices. 

This is the right thing to do. But AI conversation doesn’t have to stop at choosing between banning it completely or embracing it without limits.

More than asking, “Should schools ban AI?” I’d like to explore, “How can schools introduce AI in ways that protect what matters most about learning: human connection, curiosity, creativity, and critical thinking?”

I look forward to see our policy makers take this opportunity to properly regulating AI use in our schools, because simply banning it for now will not stop Big Tech from filling the vacuum and use our students as guinea pigs for chatbots.