AI
AI is arriving in classrooms faster than most school leaders have had time to think about what it means for students with additional needs. This is the guide I wish existed six months ago.
Every school leader I speak to right now is somewhere on the same spectrum of engagement with AI: from enthusiastic early adoption to cautious watching-and-waiting to genuine anxiety about the implications. Most are trying to form a coherent position with insufficient time, rapidly changing information, and the weight of genuinely competing obligations: to keep pace with technology, to protect students, to support teachers who are already stretched, and to not get it wrong in ways that will be visible and criticised.
This guide is aimed specifically at the intersection of AI and inclusive education, an area that I think has more promise, and more specific pitfalls, than the general AI-in-education conversation usually acknowledges. The students who stand to benefit most from well-implemented AI tools are often the same students who have been most poorly served by the systems we currently have. Getting this right matters enormously.
Where the genuine opportunity is
AI tools have the potential to reduce three of the most persistent equity barriers in inclusive education: the time cost on specialist staff, the access barrier created by written-language-dependent assessment, and the variability in quality of differentiated resources.
Specialist staff in inclusion, learning support teachers, special education coordinators, student wellbeing leads, spend a disproportionate amount of their time on documentation: ILPs, support plans, review reports, meeting summaries. AI tools can reduce this significantly. A support coordinator who currently spends four hours writing three ILPs can use AI-assisted drafting to complete the same work in ninety minutes and spend the reclaimed time with students and families. This is not a minor efficiency gain. It is a shift in how specialist time is allocated.
“The students with the most complex needs require the most professional expertise. Anything that returns specialist time from administration to direct support is a genuine equity intervention.”
Text-to-speech and speech-to-text: the access tools
The most immediately impactful AI applications for students with additional needs are the access tools: text-to-speech (TTS) and speech-to-text (STT). These are not new technologies, but AI has made them dramatically better, more natural, more accurate, more accessible, and the cost has dropped to zero for many implementations.
Schools that have embedded TTS and STT as standard tools, available to everyone, not just students with documented needs, report significant benefits for students with dyslexia, processing difficulties, and physical access needs, as well as for English language learners and students with reading fluency challenges. The universal design approach removes stigma and ensures the tools are there when they're needed without requiring a separate process to access them.
AI-assisted differentiation
The resource differentiation challenge in inclusive education is significant and chronic: teachers need multiple versions of materials at different reading levels, in different formats, with different levels of scaffolding, and they rarely have the time to create them. AI tools have made this genuinely tractable in ways that were not possible three years ago.
- Tools like Diffit and MagicSchool AI can generate levelled versions of any text or topic in minutes rather than hours
- AI can produce visual summaries, simplified vocabulary lists, and step-by-step task breakdowns from standard curriculum materials
- Modified assessment versions, reduced language complexity, chunked instructions, visual supports flagged, can be drafted in a fraction of the previous time
- Communication support materials, social stories, visual schedules, choice boards, can be created with AI assistance and then refined by the teacher or specialist who knows the student
The risks worth taking seriously
AI in education is not risk-free, and the risks are not uniformly distributed. The students who are already most vulnerable to poor data practices, algorithmic bias, and reduced human connection are often the same students with additional needs. School leaders should think carefully about the following.
- Privacy and data, student data, particularly data about disability and learning differences, requires careful handling; AI tools should be vetted against your school's privacy policy and the relevant state and national frameworks before use with student information
- Algorithmic bias, AI tools trained on mainstream datasets may not serve neurodivergent students well; watch for outputs that pathologise difference or default to neurotypical assumptions
- Reduction in human contact, the students who most benefit from AI efficiency gains are also the students who most need human relationship; the goal of reclaiming time is more time with students, not less
- Over-reliance on AI assessment, AI-generated profiles of students are starting points, not conclusions; professional judgment remains essential and cannot be substituted
- The illusion of inclusion, generating a differentiated resource does not constitute inclusive practice; the resource needs to be implemented by a teacher who understands the student, and the quality of that understanding is not something AI provides
What a school AI policy for inclusion should address
Most school AI policies are currently focused on student integrity, plagiarism, academic honesty, the role of AI in student work. These are important concerns. But for inclusive education specifically, there are additional dimensions worth building into policy.
- Which AI tools are approved for use with student data, and what de-identification protocols apply
- How AI-generated ILP drafts and support plans should be reviewed and signed off, the professional accountability remains with the human
- How AI access tools are made available, ideally universally, as a design principle, rather than as individual accommodations
- What training support is available for learning support staff to use AI tools effectively
- How the school will evaluate whether AI tools are improving outcomes for students with additional needs, not just reducing staff time
The leadership question
The schools that are getting AI in inclusive education right are not the ones that have the best technology. They are the ones where leadership has made an explicit commitment that AI will be used to improve the quality of support for students who need it most, not simply to reduce costs or generate compliance documents more efficiently.
That commitment requires naming it as a value, building it into how success is measured, and creating the conditions for specialist staff to use the time AI returns them in ways that actually reach students. The technology is available. The question, as always in inclusive education, is whether the will to use it well is present.
It needs to be.
AI
A note on accuracy:While every effort has been made to ensure the information in this article is accurate at the time of writing, facts, policies and research can change. We're human, and sometimes we get things wrong. If you spot something that needs updating, we'd genuinely love to hear from you.
Newsletter
Worth reading. Not often.
Practical guides on neurodiversity, NDIS navigation, and Australian schools. Sent when there's something worth saying, not on a schedule for the sake of it.
No spam. Unsubscribe any time.
Dave Harrison
ESW · Neurodiversity Advocate · Podcast Host
Dave Harrison is currently working in Australian schools as an Education Support Worker. He's the founder of THRVHUB, host of the Different Is Normal podcast, and a parent of a neurodivergent teenager, writing from both sides of the classroom.
More about DaveMore to read
Comments
Sign in with GitHub to leave a comment. All comments are moderated through GitHub Discussions: respectful and on-topic only.