AI and Dyslexia Instruction: Why Skilled Reading Teachers Still Matter
Artificial intelligence is definitely becoming part of everyday life in schools. We are seeing teachers use AI to brainstorm lesson plans, generate word lists, draft reading passages, create printables, write parent emails, and adapt classroom activities.
I can certainly understand why! Teachers are overwhelmed. And planning thoughtful instruction takes time. Finding the right words, sentences, and activities for a particular lesson can be exhausting, especially when you are already managing a full schedule and a wide range of student needs. So when a tool can create something in seconds, it can feel like a lifeline.
I personally use technology in my own work, and I really believe it can help teachers save valuable time. I am not interested in telling teachers that every AI tool is bad or that we should ignore technology altogether. But I do think we need to be honest about its limitations, especially when we’re talking about AI and dyslexia instruction.
AI can certainly generate a reading lesson. That does not mean it can teach a child with dyslexia.
In this post, I want to share what I have found AI can be great for and, on the flip side, why we still need skilled human teachers. When it comes to providing responsive, diagnostic reading instruction, teachers are irreplaceable.
AI Can Create Materials, But It Cannot Know Your Student
We know that a child with dyslexia does not simply need more phonics worksheets, longer word lists, or additional opportunities to practice reading. That child needs carefully sequenced, explicit, systematic, and cumulative instruction. The child also needs a teacher who understands the structure of language and can use the student’s responses to determine what should happen next.
The International Dyslexia Association explains that Structured Literacy instruction includes deliberate teaching and continuous interaction between the teacher and student. The teacher observes patterns in the student’s responses and adjusts the pacing, presentation, and amount of practice accordingly.
AI cannot not truly know that your student hesitated before reading a word. It does not hear the difference between a student who guessed quickly and a student who attempted to blend the sounds but became confused. It does not notice that a student can read a pattern accurately in a word list but loses that same pattern when reading a passage. And it will not see a child tense up when spelling begins. Nor does it know that a particular mistake has appeared in three consecutive lessons.
However, a highly trained teacher does notice these things. More importantly, the teacher knows what to do with that information.
A Polished Printable Is Not Necessarily a Good Printable
We’ve probably all tried it.. AI can generate a word list, worksheet, game, or reading passage that looks impressive. The directions may sound professional. The page may include the phonics pattern you requested. The passage may even be labeled as decodable. But polished does not always mean instructionally appropriate.
An AI generated word list often will match the general topic while completely missing the student’s actual instructional needs. This is one of my biggest concerns when teachers rely too heavily on AI generated literacy materials. Unless you have the knowledge to evaluate what was created, it can be difficult to recognize the subtle problems hiding inside a professional looking page.
Before selecting materials, a teacher needs to consider questions such as these:
- What has already been explicitly taught?
- Which skills are secure?
- Which errors are appearing consistently?
- Is there a prerequisite skill that may be missing?
- How much new information can this student manage during one lesson?
- Which words provide focused practice without introducing unnecessary confusion?
- Is the student ready to apply this concept in sentences and connected text?
- What needs to be reviewed before moving forward?
Guidance from the Institute of Education Sciences also emphasizes explicit instruction, modeling, guided practice, connected text practice, and differentiated teaching based on assessment and student needs. So while AI might be able to help you fill a lesson template, it cannot help you with these instructional practices.
Having high-quality teacher created materials to draw from can make that planning SO much easier. My Orton-Gillingham Student Notebook with Interactive Notebook Templates is one resource that I have my students create and use all the time. It is a great visual resource that I promise will make your job easier; without using AI!
Teaching Students With Dyslexia Is Diagnostic
One of the most important features of effective Orton Gillingham and Structured Literacy instruction is that it is diagnostic. This does not just mean we administer an assessment, identify a starting point, and then follow a predetermined sequence without making adjustments.
Diagnostic teaching happens throughout every lesson.
Every student response gives us information. When a student misspells a word, a knowledgeable teacher looks beyond whether the answer is right or wrong. Was the error phonological? Did the student represent every sound? Did the student forget an orthographic convention? Was the student confused about the base word or suffix? Does the same error appear in both reading and spelling? The next instructional step depends on the answers to these questions. And this is the work of a knowledgeable teacher! It is not simply the delivery of a prepared lesson plan, and it’s one of the most important distinctions to remember when thinking about AI and dyslexia instruction.
AI Does Not See the Entire Language Triangle
When we talk about reading instruction, it is easy to focus almost entirely on phonics. Phonics is essential, but teaching a student to read involves much more than practicing letter and sound correspondences.
Students must connect sounds, symbols, and meaning.
We know that teachers of students with dyslexia need to understand phonology, sound and symbol relationships, syllable structure, orthography, morphology, syntax, semantics, and written discourse. All this to say, Structured Literacy is comprehensive. It includes explicit instruction in decoding and spelling as well as vocabulary, comprehension, oral language, and writing.
A trained teacher can recognize that what appears to be a phonics problem may actually be connected to weak phonemic awareness. The teacher may notice that a student’s difficulty reading longer words is related to limited knowledge of prefixes and suffixes. The teacher will see that a student can decode a sentence accurately but cannot explain it because vocabulary or syntax is interfering with comprehension. These areas of language are connected. And a knowledgeable teacher does not simply teach them as isolated pieces. The teacher makes connections while remaining responsive to the student. AI cannot fully interpret how all of these important elements are interacting in the mind of the child sitting in front of you.
If this conversation around AI and dyslexia instruction has you thinking about how technology fits into evidence-aligned literacy teaching, I also recommend listening to this episode of the Together in Literacy podcast for more on this topic.
The Relationship Between the Teacher and Student Matters
We know that children with dyslexia often arrive at intervention carrying more than a reading difficulty. Most of these children have experienced years of frustration. They may believe they are not smart. They may avoid unfamiliar words, rush through difficult tasks, rely on guessing, or shut down when something feels too hard.
This is another important consideration when we think about AI and dyslexia instruction. An effective teacher does not simply mark an answer as incorrect and move to the next item. The teacher builds trust and knows when a student needs another model, another turn, or a reduced cognitive load.
AI can produce encouraging words, but it does not share a relationship with the child. It cannot really understand the history behind a student’s hesitation or know why one correctly read word may represent an enormous breakthrough. For our most vulnerable readers, the relationship between the teacher and student is such a key part of effective instruction.
AI Can Still Be a Helpful Teaching Tool
None of this means that teachers should avoid AI completely. AI can be useful for beginning a task, organizing ideas, or reducing the time spent on routine work.
Here are some ways that I have found AI can be a HUGE help:
- Brainstorming possible activity formats
- Draft sentences that contain a target pattern
- Generate a preliminary list of words
- Create comprehension questions for a teacher selected passage
- Reformat teacher created content
- Draft parent communication
- Suggest ways to vary familiar practice activities
But the key to using any AI created work is; the teacher must still review the result. Are the words appropriate? Are the sentences natural? Does the passage follow the student’s instructional sequence? Are there hidden concepts that have not been taught? Does the activity provide meaningful practice? Is the information accurate? Does this lesson make sense for this student at this particular point in instruction?
UNESCO’s guidance for AI in education emphasizes a human centered approach that protects human agency and strengthens teacher capacity rather than replacing professional judgment. AI can offer possibilities, but teacher must decide which possibilities are instructionally sound.
Instead of using AI to build your Word Lists, check out my Word List Builder. You can also check out this blog post to learn more.
AI Is Only as Helpful as the Knowledge Behind the Prompt
One of the greatest risks of AI generated reading materials is that they can create the illusion of expertise. Upon first glance, an AI response may be neatly organized. It may use professional terminology. It may describe a lesson as explicit aligned with the Science of Reading. But we know that using those words does not guarantee that the lesson reflects those principles.
A teacher with deep knowledge can review an AI generated lesson and immediately recognize what needs to be changed. AI does not make teacher expertise less important. If anything, it makes teacher expertise even more important. And as it is becoming easier to create instructional content, the more important it becomes to know how to evaluate that content.
Our Most Vulnerable Readers Need More Than Shortcuts
Schools are facing real challenges. Teachers have limited planning time. Intervention needs are growing. Budgets are tight. Many schools are searching for ways to provide more support without adding more demands to teachers’ already full plates.
AI will continue to be part of that conversation. But we need to be cautious about treating technology as a substitute for investing in teacher knowledge. I guarantee that students with dyslexia do not need an endless supply of automatically generated activities. What they really need is teachers who understand. Understand things like how reading develops, structure of language, when to slow down, review, change activities, and move forward.
So while using AI for dyslexia instruction may help create shortcuts around some of our planning tasks, it cannot provide a shortcut to becoming a skilled reading teacher.
If you are a teacher facing overwhelm (I know so many of us are!) I would love for you to check out my Building Readers for Life Academy. Building Readers for Life Academy was created to support educators in developing the knowledge and skills they need to provide effective Structured Literacy instruction. Rather than you needing to stay up to date on educational research, we bring the latest in literacy research to you. The library and community inside this membership is truly amazing. The best part is, you can try out the Building Readers for Life Academy membership for just $1!
Click here to check out the Building Readers for Life Academy.
Use THIS LINK to get your first 30 days for just $1!
The Teacher Must Remain the Expert
In regards go AI and dyslexia instruction, the conversation does not need to become teachers against technology. AI CAN be useful, save us time, help you get started. But it must remain a tool in the hands of an informed professional.
When we are teaching children with dyslexia and other significant reading difficulties, a polished lesson plan is not enough. The quality of the instruction still depends on the knowledge of the person delivering it. Our most vulnerable readers deserve more than content created quickly. They deserve teachers who know what they are looking for, why they are teaching each skill, and what to do when the student’s response changes the plan. That kind of expertise has always mattered, and in the age of AI, it matters more than ever.
If you’re looking for quality teacher created resources to make your life easier have a look at all of my offerings here. If you’re looking for a place to start the Orton-Gillingham Student Notebook is one of my favorites! You can also grab this 15% off coupon to save on resources from The Literacy Nest.
A Final Reflection for Teachers
You do not need to compete with AI by creating more materials or producing lessons faster. Your value is not measured by the number of printables you can make. Your value comes from the knowledge, judgment, and connection you bring to your students every day.
As AI and dyslexia instruction continue to evolve, we can absolutely use technology to save time and support our planning. But it should never replace the teacher knowledge that allows us to understand why a student is struggling and decide what they need next.
If you want to continue building that knowledge, you can learn more about the Building Readers for Life Academy or listen to this Together in Literacy podcast episode. And if you’re looking for resources to support your literacy instruction, check out my shop! You can also grab a 15% off coupon here.
AI can generate the page, but you are the one who teaches the child.
