New York City AI Leadership
AI Misuse Across New York City Schools: A Five-Borough Integrity Plan
A distinct five-borough strategy for multilingual AI use, scale, translation, deepfakes, student voice, and transparent schoolwork.
2026-08-26 · 13 min read
Easy-read summary
The big idea for New York City, New York schools
New York City cannot treat student AI misuse as one uniform behavior. A seventh grader translating directions, a newcomer rehearsing English, a senior polishing a college essay, and a student generating an offensive image are using similar technology for fundamentally different purposes. Across the five boroughs, school leaders need a framework flexible enough for multilingual classrooms, specialized programs, career pathways, and different device policies. The constant should be authorship: students must not use a machine to create a false picture of what they know or who created the work. That standard leaves room for accessibility and tutoring while drawing a firm line around deception. It also shifts staff conversations away from guessing whether prose ‘sounds like AI’ and toward evidence of learning, disclosure, consent, and harm.
Teach students to disclose AI help when a teacher or assignment requires it.
Use AI as a coach for questions, practice, feedback, and revision—not as a ghostwriter.
Verify important AI claims with reliable sources before sharing or submitting them.
Protect student privacy by avoiding personal, family, medical, or school-identifying details in prompts.
Section 2
One city, many classroom contexts
Translation is one of the city's most important gray areas. AI can help a student understand directions or compare a phrase in two languages, but full translation of a graded response may hide the language skill being assessed. Teachers should state whether the goal is content knowledge, English development, or both. When translation is permitted, the learner can save the original draft, identify translated sections, and explain important word choices. When it is not permitted, schools should provide approved human or accessibility support rather than simply removing assistance. This approach respects multilingualism without allowing an uncredited system to become the author. It also prevents inconsistent discipline when two teachers interpret the same translated paragraph in opposite ways.
Teens reported using generative AI for school assignments in Common Sense Media research.
RAND reported student AI homework use among middle school grades and up by December 2025.
College Board reported high school students using generative AI for schoolwork.
Section 3
Translation support versus hidden authorship
A blanket ban is unlikely to work at New York City's scale. AI features are embedded in common search, document, phone, and tutoring products, so a blocked domain does not equal a protected assessment. More durable rules focus on behaviors. Students may not submit unacknowledged generated work, fabricate evidence, expose private data, or imitate another person's identity. Departments can then publish examples relevant to their disciplines. Art classes may allow image ideation but require process sketches; computer science may allow debugging but require line-by-line explanation; social studies may allow counterargument generation but require verified primary sources. A short core policy plus subject examples gives educators consistency without pretending every classroom task is identical.
Section 4
Why a single ban will not scale
Student voice is essential if the rules are expected to survive real use. A representative group from different grades, boroughs, language backgrounds, and programs can review sample scenarios and identify confusing language. Students often know where tools are embedded and which shortcuts circulate among peers. Inviting that knowledge is not surrendering authority; it helps adults close gaps before enforcement begins. A student advisory team can also create disclosure examples, peer-facing videos, and an anonymous question channel. Their message should emphasize that disclosure is not automatic guilt. Honest use might read, ‘I asked an approved tool for three counterarguments, checked them against our sources, and wrote the response myself.’ Normalizing that sentence makes hidden use less attractive.
Section 5
Use student voice to write workable rules
Synthetic media requires a different response from copied homework. In a densely connected city school, a fake voice clip or humiliating image can spread across group chats before the first bell. Students need to understand consent: another person's face, voice, name, or private story is not raw material. Schools should publish a rapid reporting route, preserve evidence, support the targeted person, and discourage students from forwarding the content ‘to warn others.’ Administrators should connect AI fabrication to existing harassment, bullying, and safety policies. Classroom integrity matters, but an incident that attacks a person's dignity is not merely cheating. Treating these categories separately helps schools respond proportionately and communicates that artificial media can create immediate, real-world harm.
Student prompt examples
- “Coach me, but do not complete the assignment for me.”
- “Ask me three questions before giving suggestions.”
- “Show what might be wrong or missing in this answer.”
Section 6
Respond quickly to synthetic media
Assessment redesign is more reliable than automated detection. Detectors can be wrong, especially with short, formulaic, or multilingual writing. New York City teachers can gather a richer authorship record through quick writes, planning sheets, conferences, version history, oral defenses, and in-class application. A student who used AI appropriately should be able to describe the prompt, identify weak output, and defend the final choices. A student who cannot explain a central claim needs more learning regardless of how the text was produced. Schools can reserve controlled, no-tool conditions for essential independent skills and use open-tool tasks to assess evaluation and judgment. Clearly separating those purposes reduces accusations and gives students a fair chance to demonstrate mastery.
Section 7
Assess learning through explanation
Coordination matters in a system where students may encounter many adults and programs. Principals can provide a one-page core standard, department leaders can add examples, librarians can lead verification lessons, counselors can address personal-data risks, and families can receive translations of the guidance. Family sessions should avoid alarmist demonstrations and focus on practical questions: Who created this? What information went into the prompt? Which claims were checked? Students also need to know where approved accessibility support ends and concealed completion begins. Reviewing the guidance each semester allows schools to respond to new embedded features without rewriting their values. Honest authorship, verified information, privacy, consent, and explainable thinking remain stable even as products change.
Section 8
Coordinate families, teachers, and leaders
Jim Jordan can give this large, varied audience a common vocabulary. His school assembly does not reduce AI to cheating or celebrate every new tool. It shows middle and high school students how convenience can quietly replace capability, why digital choices affect reputation, and how thoughtful AI use can prepare them for work without hollowing out education. Jim brings two decades of student speaking together with hands-on JimmyAI experience in training, chatbots, applications, and AI phone services. For New York City schools, the keynote can launch a student advisory process, department-specific rules, and family outreach. The most successful outcome is not fear of being caught; it is a student who can explain when AI helped, when it should stay closed, and why the difference matters.
Action checklist
Use this after the assembly
- ●Create green, yellow, and red examples of acceptable AI help.
- ●Give students simple prompt language such as: coach me, do not complete it.
- ●Ask students to cite or explain how AI was used when appropriate.
- ●Revisit AI expectations during major writing, research, coding, and media projects.
Sources and citations