A Discussion with Assistant Superintendent Dr. Jeffrey Moore
What is the first thing you think about when you hear “AI use in schools”? Cheating, plagiarism, deep fakes, kids outsourcing their thinking to machines?
Those concerns are real. But if we stop there, we miss a more important question: Why are kids turning to AI in the first place, and what does that say about us?
One of the most striking data points Jeff and I discussed came from Harvard’s annual look at AI use. The top two uses people report?
- Using AI as a friend
- Using AI as a therapist
Think about that for a moment. Major commercial AI systems—designed to maximize engagement and “please” the user—are being leaned on for emotional support and meaning-making. That should worry us, but not only because of what AI might say.
The bigger question is: What unmet needs are driving people, including kids, to seek friends and therapists in algorithms? As Jeff put it:
“The fact that we are using it to have friends and to get therapy… that speaks to something that I don’t think schools can solve by themselves… The need for friends and the need for therapy should be striking to us more than just ‘kids are using this for friendship and therapy.’ Like, well, why?”
Jeff sees AI as a kind of mirror held up to our society. If students are turning to AI for companionship and counseling, it signals a larger breakdown in connection, community, and mental health support.
Jeff begins with the human:
My vision of this is that technology, and particularly AI, becomes an extension of one’s goals and becomes a way for someone to sort of augment their strengths and mitigate their weaknesses.
He’s interested in AI as a tool that helps students know themselves better: their strengths, their gaps, and how to navigate both.
That’s a challenge schools cannot solve alone; and one they also can’t ignore.
AI, Cheating, and What We Choose to Measure
In our conversation, I raised the tension between AI and how we currently grade and test:
We have this emphasis on test scores or grades, and we test and we grade based on output. AI at this point can produce better output than really any kid up through eighth grade, and most kids through high school or even adulthood. So how do we change from measuring just the output to actually measuring learning?
Jeff’s point was that kids using AI inappropriately isn’t just a discipline problem; it’s a design problem.
If we know students will live and work in a world where using AI is expected, then punishing them for using it on assignments that look exactly like future workplace tasks is backwards. We should instead be asking: What do we want kids to be able to do with these tools? And how do we assess the parts of learning—curiosity, ethical judgment, self-awareness, the capacity for relationship—that the tools can’t automate?
Whole Child, Whole System: The OCRA Project
This is where Mamaroneck’s approach stands out. The district isn’t just tweaking assessments; it’s investing in the whole ecosystem around kids, particularly their mental health.
Last spring, I interviewed Mamaroneck seniors Zakir Strugatsky and Gray Hackett about their OCRA Project (Original Civics Research & Action). As Mamaroneck students, they researched mental health challenges and stigma among their peers. Their work became a catalyst for district improvement.
Jeff and the district used the project’s findings to launch mental health first aid training for staff, equipping adults to recognize distress, respond constructively, and then, over time, train students in the same skills. The goal is a culture where:
- Students are seen and supported, not just managed.
- Peers know how to offer a “warm handoff” when someone is struggling.
- Mental health is treated as a shared responsibility, not an individual weakness.
The OCRA Project is a powerful example of whole child education in practice: students doing authentic civic research (both with and beyond technology), being taken seriously, influencing community decisions, and then watching their ideas scale into real training and change.
Balancing Risk and Power
AI raises legitimate concerns: skill atrophy, dependency, misinformation, emotional manipulation.
Mamaroneck’s emerging approach suggests a different path:
- Partner with parents to set the groundwork on kids’ learning and social needs as part of setting policy on AI.
- Equip students and adults with mental health skills and relational tools so that fewer people feel the need to rely on AI as their primary friend or therapist.
- Redefine success beyond test scores to include self-awareness, purpose, and the ability to harness powerful tools ethically and thoughtfully.
Jeff’s hope for this year could have been lifted from the mission of MindShifting:
This year, I just want to be able to count more and more the number of students we empower to understand what their strengths are, understand how to mitigate their weaknesses through those strengths, and really identify and begin to work toward their passions… Every day, I want to be able to say one more kid, one more kid, one more kid.
AI will reshape education one way or another. The bigger question for the Mamaroneck district is how to move from isolated great programs to innovation at scale, and in the process, finally build the kind of whole child, human-centered system our students have needed all along.
If you’d like to see the full conversation with Dr. Jeff Moore, you can watch it here:
Leave a Reply