The Anne Arundel County Board of Education conducted the first reading of a new responsible-use-of-artificial-intelligence policy, designated Policy DM, at its meeting on October 21, 2026. The draft policy is now open for 30 days of public comment, with a second reading scheduled for December 2 and final adoption set for December 16, according to the district. The policy committee had met on October 6 to shape the proposal before sending it to the full board as a first-reader item.
The draft policy establishes a central principle that AI should support people, not replace their judgment. It addresses human oversight, privacy, equity, appropriate student use, employee training, and vetting AI tools before adoption, according to a district news release and local news reports. Under the plan, administrative regulations will be developed after the policy is adopted.
State mandate and missed deadline
Maryland's SB 720 (Chapter 634), approved by the governor on May 26, 2026 and effective June 1, 2026, requires the Maryland State Department of Education to provide AI guidance to local school systems and requires each local school system to adopt an AI policy aligned with that guidance within 120 days of its release, according to the state law. The law also requires districts to designate an AI coordinator, use a state-developed rubric to evaluate AI tools, and prohibits AI from making final determinations on awards, funding, eligibility, discipline, or other policy decisions.
The state set an October 22, 2026 deadline for school districts to adopt their own AI policy. Anne Arundel County will miss that deadline due to its procedural requirement of three board readings before adoption, as reported by the Capital Gazette. The delay is not unique to Anne Arundel. Howard County, Baltimore County, and Baltimore City were also expected to miss the October 22 deadline, according to Hoodline.
Local precedent: the Broadneck High School case
The new policy arrives after a documented local incident involving AI-detection software. In 2024, Anne Arundel County's Broadneck High School used GPTZero AI-detection software to flag a student's assignment as AI-generated, resulting in a plagiarism violation, as reported by the Capital Gazette and WBFF Fox45. The case raised concerns under the Family Educational Rights and Privacy Act (FERPA) because the school uploaded a student's work to a third-party site without consent. This occurred before the district had any AI-specific policy in place, highlighting the gap that Policy DM seeks to fill.
Public response at the first reading
At the October 21 meeting, parents and teachers voiced concerns about AI's impact on critical thinking and creativity, according to CBS News Baltimore. Art teacher Amanda Nickelson said AI diminishes creativity, communication, and critical thinking. Parent Hillary Klemmt urged the board to let students think and problem-solve independently. Board member Dr. Joanna Tobin noted the difficulty of predicting AI's consequences. Some parents asked for a moratorium on AI use in schools.
The district scheduled seven community meetings at 6 p.m. to gather public input before the December 16 vote: North County High School (October 27), Glen Burnie High School (November 2), Northeast High School (November 5), Arundel High School (November 10), Southern High School (November 12), Annapolis High School (November 17), and Meade High School (November 19), according to Hoodline.
State guidance and research evidence
Maryland's statewide AI guidance, published in June 2026, establishes a framework covering data privacy and security, equity and access, academic integrity and assessment, professional learning, student AI literacy, operational uses, and procurement and vendor governance, according to the Maryland State Department of Education. The guidance states that AI must enhance, not replace, educators, counselors, and school leaders, and that generative AI systems require human oversight.
The state guidance cites research showing that AI amplifies existing risks related to data collection, surveillance, re-identification, and cybersecurity, particularly for minors (Crawford, 2021; Reddy et al., 2020). It also cites research that algorithmic systems often reflect and reinforce existing racial, linguistic, economic, and disability-based inequities if not carefully designed and monitored (Benjamin, 2019; Noble, 2018), and that AI-detection tools are unreliable and inequitable (Liang et al., 2023). The guidance documents research on automation bias, a documented tendency of educators to over-trust automated recommendations, and emphasizes that professional learning must address critical evaluation, verification, and reflective practice when using AI tools (Goddard et al., 2012; Holmes et al., 2021).
A separate meta-analysis published in PubMed Central found that requiring students to construct their own queries to get useful feedback from generative AI can cultivate initiative in knowledge exploration and divergent thinking, complicating the binary of banning versus freely embracing AI tools, according to that study cited by Hoodline. The finding is described as mixed in the research, as the effects depend on how the tools are implemented in instruction.