Using AI to Support IEP Development: Insights from CIDDL’s AI Office Hours
Authors: Yerin Seung; info@ciddl.org
At CIDDL’s most recent AI Office Hours, the conversation focused on one of the most pressing questions in special education today: Can artificial intelligence (AI) meaningfully support the development of Individualized Education Programs (IEPs)? Dr. Olivia Coleman, assistant professor at the University of Central Florida, and Danielle Waterfield, doctoral candidate at the University of Virginia, led a thoughtful discussion on both the opportunities and the risks of integrating AI into IEP practices.
Why AI in IEPs Matters
IEPs are often described as the cornerstone of Free Appropriate Public Education (FAPE), but developing them is a time-intensive and high-stakes process. Many teachers report facing limited or inconsistent training in writing IEPs, heavy time burdens due to growing caseloads and paperwork demands, and inconsistent quality of goals and alignment across components. These challenges can result in inconsistencies across IEP development and implementation. Dr. Coleman and Waterfield pointed out that these realities leave many teachers overwhelmed and students underserved, raising the question of how AI might help.
Emerging Uses of AI in IEP Development
Teachers are beginning to experiment with AI to draft present levels of performance, turning informal notes into professional, educational language. AI can also generate measurable, standards-aligned IEP goals, often improving vague or incomplete drafts. Beyond goals, teachers have used AI to suggest accommodations grounded in research, to synthesize data into clearer progress reports, and to streamline routine paperwork so that they can spend more time directly with students. One example presented in the session illustrated how an incomplete goal, such as “Student will improve their reading skills,” can be transformed into a SMART goal (specific, measurable, attainable, relevant, and time-bound) by using carefully designed prompts. Teachers in the study appreciated how AI could refine their work and potentially give them valuable time back to focus on instruction and relationships.
Research on AI and IEPs
The field is still young, but early studies suggest promising outcomes. Studies conducted by Rakap & Balikci (2024) and Waterfield and colleagues (2025) both found that AI-generated goals were comparable in quality to teacher-written goals. Despite this, many teachers expressed uncertainty about how to use AI responsibly, even as they recognized its potential for efficiency and improved quality. These mixed feelings reflect both the excitement and the caution surrounding AI’s role in special education.
Risks and Ethical Considerations
With opportunity comes caution. Dr. Coleman and Waterfield emphasized several risks of using AI in federally mandated documents such as IEPs. One concern is the loss of individuality when AI outputs default to generic language. Another is the presence of biases in training data, which often underrepresent individuals with disabilities. They also highlighted data privacy concerns related to FERPA compliance, the risk of hallucinations where AI generates inaccurate or fabricated information, and the possibility of overreliance if educators stop critically evaluating outputs. To address these issues, the presenters are developing a framework and decision tree to guide practitioners and teacher educators. Their model emphasizes professional judgment, prompt fluency, bias checks, individualization, transparency with families, and ongoing reflection. They argued that AI should be treated as a partner that supports teachers rather than as a replacement for practitioners' expertise and care.
Practical Applications in Teacher Preparation
An audience member asked how AI could be introduced in an undergraduate IEP writing course. Dr. Coleman suggested creating assignments where students compare AI-generated IEP components with teacher-written ones, evaluate them using a rubric, and reflect on the differences. This kind of exercise encourages students to critically evaluate AI outputs rather than accept them at face value. Waterfield added that AI can also support progress monitoring and the development of instructional materials, pushing beyond the writing stage to focus on implementation. Together, their suggestions demonstrated the potential of AI not only as a drafting tool but also as a means of deepening professional preparation.
Looking Ahead
Although research on AI in IEP development is still emerging, one message was clear: AI is not going away. Teachers are already experimenting with tools such as MagicSchoolAI, Khanmigo, Brisk, Playground IEP, and Expert IEP. The challenge is not whether to use AI, but how to use it ethically, responsibly, and effectively to support both educators and students. As concluded in the office hour, our kids deserve high-quality IEPs. AI can be a partner in that process, but never a replacement for the expertise, judgment, and care of special educators.
Resources to Explore
For those interested in learning more or experimenting with tools, here are some of the resources and platforms mentioned during the session:
- Framework for Ethical AI Use in IEP Development
- Decision Tree for the Ethical Use of AI in IEP Development
- MagicSchoolAI – AI tools designed for educators, including lesson planning and IEP support.
- Khanmigo – Khan Academy’s AI-powered teaching assistant.
- Brisk Teaching – AI platform for educators, used in several school districts.
- Eduaide.AI – AI-powered tool for creating instructional resources.
- Playground IEP – Platform built specifically to support IEP writing tasks.
- Expert IEP – Another specialized platform for IEP development.
There’s an AI for That – A searchable directory of AI tools for various tasks.
Join the conversation in our community!
CIDDL is committed to providing high-quality resources to support the increasing knowledge, adoption, and use of a range of educational technologies that can be used for educators, related services, or leadership preparation programs. For more resources, including videos and blogs, subscribe to our newsletter and follow us on YouTube, Facebook, and LinkedIn. The most important part of our CIDDL community is YOU. Join our community and share the innovative ways you are using technology, ask a question about technology integration, or participate in our bi-weekly live AI Community Chats. We look forward to seeing you in our community!