Get Full Government Meeting Transcripts, Videos, & Alerts Forever!
Get email alerts on the Education Ai Policy topic
No spam. Unsubscribe anytime.
Board reviews draft AI policy that would bar default student use of AI for submitted work
Summary
The Orange County Public Schools work session reviewed updates to Policy IJNDC to limit AI in high‑stakes decisions, require district‑vetted systems that do not train on OCPS data, prohibit using AI‑detection tools as the sole basis for discipline, and set default prohibitions on AI for submitted student work unless an educator documents approval.
Get email alerts on the Education Ai Policy topic
No spam. Unsubscribe anytime.
Orange County Public Schools officials reviewed an updated draft of Policy IJNDC on July 21, a sweeping draft that would set clear guardrails for staff and student use of artificial intelligence in classrooms.
The board heard a presentation from Maurice Dragon, identified in the meeting as the district’s chief information officer, who summarized technical changes in consumer AI — including “thinking” models that internally reason and multimodal systems that process text, images, audio and video — and the policy implications for schools. Dragon told the board that such models are becoming more powerful and costly, likely reducing free daily access and risking new inequities between students who can pay for advanced tools and those who cannot.
Board members were briefed on how the district’s chosen educational AI tool, Conmigo, uses retrieval‑augmented generation and scaffolding checks to constrain answers to vetted curricular resources and preserve ‘‘productive struggle’’ rather than deliver final solutions. Dragon said the district is also exploring a tool to assemble initial drafts of individualized education program (IEP) documents from district systems (Skyward, Performance Matters and others) to improve staff efficiency; he emphasized that any IEP decision would remain with human IEP teams.
The draft policy would require staff to use only district‑vetted enterprise AI systems that do not use OCPS data to train public models and would explicitly prohibit use of unvetted open‑source or commercial models. In addition, the draft bars reliance on AI‑detection tools as the sole basis for student discipline, requiring secondary verification such as document edit history, baseline writing samples or a direct conversation with the student before disciplinary action.
Several board members pressed for clearer language in multiple sections. Member Banos sought clarification about the word “final” in the description of high‑stakes decisions, and whether parents must consent to AI being used in IEP drafting; Dragon and staff replied that the tool under consideration is intended for draft assembly and that staff—not AI—would make the final determinations in IEP meetings. Members also asked for a documented disclosure process when district‑approved instructional platforms that incorporate AI are used in a classroom; staff said educators would be expected to document expectations in a syllabus, lesson plan or initial parent communication.
Members discussed equity concerns at length. Several board members, including Member Salamanca and Member Douglas, warned that gating of higher‑quality models behind paywalls could widen disparities and urged that AI literacy and curricular guidance be part of implementation. Staff said Conmigo is supported by state grant funding, that Google and Microsoft tools are providing certain AI features under existing licensing, and that the district will offer training and curricular guidance during preplanning and the first semester.
Dragon recommended ongoing review of the policy as technology evolves, and the superintendent told the board the draft is slated for the regular July board meeting next week so the rules are in place before the school year begins. No formal vote or adoption occurred at the work session.
The board will consider clarified wording on the student default prohibition, documentation requirements for teacher authorization of AI in assignments, and the district’s process for vetting enterprise AI tools before the policy returns as an action item.

