Writing Instruction in the Age of AI

Lately, I’ve received several questions about the role of generative AI in writing instruction and student writing in school. This post highlights key points from a longer article I recently wrote with the same title.

Since the launch of ChatGPT in late 2022, the use of generative artificial intelligence (AI) for writing in schools has captured the attention of educators across the country. Although much remains to be learned about using AI wisely, the technology is here—and students are already using it to write their papers. As a result, educators must consider how AI will impact writing instruction and the role of writing in their classrooms. Additionally, because AI is becoming commonplace in professional work environments­, students must learn how to use AI effectively and ethically to prepare for the workforce.

How does the availability of AI influence writing instruction? How can students use AI to support their writing development? Will using AI diminish the cognitive and academic benefits students gain from writing as a tool for learning? My response to these questions is clear: just because AI can write for students doesn’t mean they no longer need instruction to become proficient writers. The challenge—and opportunity—is to determine how AI can supplement, not replace, best practice for teaching writing.

What is AI Text Generation?

Generative artificial intelligence chatbots use large-scale language models (LLMs) to produce text based on the statistical distribution of words, phrases, and punctuation based on massive amounts of internet-based language. It’s important for students and teachers to understand that AI does not think, reason, or believe in the content it generates. It cannot determine whether the information is accurate because it operates by predicting language patterns, not verifying truth. Sometimes AI models will generate inaccurate or nonsensical responses by identifying false patterns or “hallucinations”—fabricated facts, citations, or quotations. As a result, students need to remember that AI is not a consistently reliable source of information. Additional concerns about using AI are explained in the full article, including biased content, academic integrity and plagiarism, equity and access, data privacy and security, and dependency on technology.

Why Students Must Learn to Write in the Age of AI

As generative AI continues to advance, a critical distinction has emerged: students who use AI to support their writing versus those who rely on AI to write for them; this distinction is important. The availability of AI does not eliminate the need for students to learn how to write, nor does it replace the importance of explicit writing instruction. Instead, educators must guide students to use AI as a supportive tool—not a substitute—for learning to write, think critically, and communicate effectively. Reasons why writing instruction is needed:

• Writing Supports Comprehension and Content Learning: Writing is not merely a final product—it is a process and a powerful tool for critical thinking that enhances students’ understanding of content across all subject areas (Graham & Perin, 2007; Graham, Harris, & Santangelo, 2015; Graham et al., 2020). Research supports writing about texts and other sources to build both comprehension and writing skills, and learning in different subject areas (Biancarosa & Snow, 2006; Graham & Perin, 2007; Graham & Hebert, 2010). Writing personal reactions, analyzing and interpreting the text in writing, writing summaries and notes, and answering and creating questions about text in writing are particularly helpful (Graham & Hebert, 2010; Graham et al., 2020).
• Writing Supports Critical Thinking: When students write about what they are reading and learning they are thinking on paper (Sedita, 2020). Writing helps them organize, clarify, and better understand information by extending their thinking and connecting new information to prior knowledge. As they write, students become more engaged, making judgments about what is most important, analyzing and evaluating information, and synthesizing information from multiple sources. This holds true for creative writing as well—poetry, plays, personal narratives, fantasy, and realistic fiction. Relying on AI to generate these forms of writing deprives students of the opportunity to express themselves and develop their creative voice.
• Writing Improves Reading Skills: As noted in the Writing to Read research guide, teaching students how to write strengthens their reading comprehension, fluency, and word reading skills (Graham & Hebert, 2010). The act of composing text helps students gain insights that support text reading, causing them to be more thoughtful readers, more attuned to how meaning is constructed. Research shows that instruction in the writing process (planning, drafting, revising), sentence construction (including sentence combining), paragraph development, and text structures improves reading comprehension. Moreover, teaching spelling and sentence construction supports reading fluency, and spelling instruction specifically improves word recognition.
• Writing Demonstrates Learning: Writing is a form of formative assessment that allows teachers to evaluate students’ depth of understanding. It can serve both formal grading and instructional planning purposes. Writing tasks such as quick writes, summaries, journal entries, note-taking, responses to teacher-generated prompts, or essay questions on tests allow students to demonstrate their grasp of content. To complete these tasks independently, without relying on AI, students need foundational writing skills. Exceptions may include adaptive technologies such as speech-to-text, spelling checkers, or grammar checkers for students with disabilities like dyslexia or dysgraphia.
• Writing Serves Multiple Purposes in Life: Beyond academic assignments, writing serves many purposes in everyday life (Hebert, 2025; Graham, 2024b). People write to communicate with family, friends, and colleagues (emails, texts, and letters), to share personal stories (memoirs, biographies), to inform (articles, blog posts, slide presentations), to persuade, and to entertain. Writing is also needed to complete forms, such as job and college applications. Writing also provides a space for emotional expression and self-discovery through journals or diaries. These deeply human acts—expressing emotion, developing a personal voice, and conveying lived experience—cannot be authentically replicated by AI. While AI may refine writing, it cannot replace the writer’s unique perspective, creativity, and voice.

The Writing Rope: AI and Multiple Writing Components

The Writing Rope framework (Sedita, 2019; Sedita, 2023) provides an instructional framework that supports writing instruction across all grade levels. It organizes the essential skills, strategies, and techniques students must master to become proficient writers into five interwoven components, represented as strands in a rope: critical thinking, syntax, text structure, writing craft, transcription. Students develop proficiency by activating and integrating the skills within each strand. To learn more about this framework, read this 2020 blog post, read this article, or watch this recorded webinar.

Although AI tools can support the use of these skills and strategies, they cannot replace the need for students to learn how to independently integrate and apply them when writing.  The five strands are summarized below, along with commentary on how AI may support—but not substitute for—student learning.

Critical Thinking: Students engage in critical thinking as they decide what they want to communicate and build background knowledge on a writing topic. This strand also includes understanding the writing process: thinking, planning, writing, and revising. During the thinking stage, students benefit from explicit instruction in brainstorming strategies and in gathering information from written and multimedia sources, including note-taking and summarizing. Instruction in planning strategies, such as the use of graphic organizers, helps students structure their ideas. They should also develop metacognitive awareness and work intentionally through the recursive stages of the writing process, supported by direct instruction in revising and editing.

AI can assist at each stage of the writing process, but it must be used judiciously. Over-reliance on AI may deprive students of the benefits that come from using writing as a tool for learning. Students must develop the ability to make decisions about the ideas to include in their writing, synthesize information from multiple sources, organize ideas before drafting, and express those ideas in their own words. Suggestions for using AI to support each stage of the writing process are provided later in this article.

Syntax: Sentences are the building blocks of written expression. A strong understanding of sentence structure supports both comprehension and effective communication when speaking or writing. Explicit instruction in sentence-level skills—such as sentence combining and elaboration—lays the foundation for fluency and clarity in writing.

AI can provide feedback to help students improve their sentences, but they must still learn to craft clear, concise, and well-developed sentences on their own. Writing strong sentences requires students to think critically about the ideas they want to convey, how those ideas relate to one another, and how to express them using precise language. Mastering sentence construction enhances students’ ability to articulate complex thinking.

Text Structure: Text structure is a defining feature of written language. Understanding its multiple layers strengthens both reading comprehension and writing development. Students benefit from direct instruction in the following areas: types of writing structures (narrative, informational, opinion/argument); paragraph structure; patterns of organization; and transitions.

AI can support students by suggesting organizational improvements, headings, transition words, and paragraph groupings. However, students must first develop a foundational understanding of writing structures and patterns of organization. Only then should AI tools be used to refine the organization of their writing.

Writing Craft: This strand focuses on the stylistic and rhetorical elements that make writing engaging and effective. Often referred to as “writer’s craft” or “writer’s moves,” these include: word choice, writer’s voice, and literary devices. This strand also emphasizes consideration of task, audience, and purpose, which shape decisions about tone, length, structure, and style.

When students draft their own writing, their unique voice and style emerge through word choice and literary techniques. If AI generates a draft or significantly revises their work, this personal voice may be diluted or lost, replaced by generic language drawn from AI training data. Students must learn to make intentional stylistic choices and maintain ownership of their writing.

Transcription: This strand includes foundational skills such as spelling, handwriting, and keyboarding, all of which are necessary for translating thoughts into written words. When students develop fluency and automaticity in transcription, they can devote more cognitive resources to higher-order tasks like composing and revising.

AI can be especially helpful in supporting spelling and basic editing—particularly for students with dyslexia, other learning disabilities, or those still learning English orthography. Tools like spell-checkers and speech-to-text can reduce the cognitive load and allow these students to focus on content and organization.

Using AI at Each Stage of the Writing Process

The stages of the writing process represent the steps that effective writers follow (Hayes, 1996; Hayes, 2004; Graham et al., 2012). This process is one of the instructional topics in the Critical Thinking strand of the Writing Rope framework. Keys to Literacy developed a model to help students remember these stages: Think, Plan, Write, Revise (Sedita, 2020). You can learn more about these stages in a 2023 blog post about the writing process.

Suggestions for using AI are provided below for each stage of the writing process. These suggestions are intended as starting points. Additional details, including sample AI prompts and important points to keep in mind for each stage are included in the full article. It is important to remember that students need explicit instruction, modeling, and guided practice to learn how to use AI appropriately, without crossing into having AI compose the text for them. They need support to avoid over-relying on AI to generate their writing. Instruction should focus on teaching students how to think critically about their writing so they can use AI to enhance the writing they produce independently.

Think Stage: At the Think stage, writers identify the audience and purpose for a writing task. They determine the topic they will write about and what they want to say. This includes brainstorming ideas and gathering information into notes. AI can be used to narrow the scope of a writing topic, build knowledge about a topic, and brainstorm ideas.

Plan Stage: At the plan stage, writers organize the ideas and information they have gathered and decide how to structure their writing. This includes creating a pre-writing outline or graphic organizer. Students may also need to gather additional information as needed. AI can be used to confirm the logical progression of ideas, improve on an outline, and identify missing information or topics.

Write Stage: During the Write stage, students generate a first draft based on their plan. They expand ideas into sentences and paragraphs, typically starting with an introduction, ending with a conclusion, and structuring the body in a logical way. Using AI to generate drafts is one of the most controversial aspects of generative AI. It’s important to keep in mind the principle: “The one who does the work does the learning.” Ideally, students should write their own first draft, as this is where critical thinking and idea development take place. AI is best used after the student has written a draft, as a tool for feedback and revision. Students can ask AI to suggest improvements to their writing. They can then compare the AI-generated suggestions to their original draft and analyze whether the recommendations are appropriate. This process helps students learn how to strengthen their writing. In this way, AI can serve as a writing coach, offering feedback that guides students in refining their work.

Revise Stage: In the Revise stage, writers improve the content, organization, and language of their draft, followed by proofreading for spelling, punctuation, and grammar. Using AI for revision is a good example of how AI can support writing without replacing essential writing skills. AI can be used to identify spelling and punctuation errors, correct grammar errors, and assist with formatting references.

Instructional Implications

Given schools’ relatively recent access to generative artificial intelligence (AI), it is not surprising that many educators have concerns about how students will use AI, how it affects the skills students need to develop as proficient writers, and what this means for writing instruction.

The most effective way for teachers to address these concerns is by experimenting with platforms like ChatGPT. Hands-on experience provides an opportunity to better understand how generative AI works, its capabilities, and its potential for supporting student writing. For example, teachers can use ChatGPT to generate writing prompts, brainstorm lesson ideas, and provide feedback on samples of student writing. By becoming comfortable with AI as a professional tool, teachers will be better prepared to guide students in using it effectively and responsibly.

Additional suggestions:

Teach students about AI, how it works, and the ethical implications of its use.

• Explain what AI is, how it works, and its limitations.
• Emphasize that AI does not think. It identifies patterns in existing language and generates responses based on those patterns.
• Discuss concerns related to AI use, including bias, academic integrity, data privacy, and the risk of students over-relying on AI for generating writing.
• Teach the value of originality and the importance of contributing their own ideas when writing.

 Reinforce writing as a thinking and learning tool.

• Explain how critical thinking, reading, and revision during the writing process lead to deeper learning.
• Clarify the purpose of writing assignments, emphasizing when AI should not replace the student’s own thinking. Encourage students to generate examples of when using AI makes sense and when it hinders learning.
• Focus on all stages of the writing process, not just the final product.
• Require students to document how they use AI during the Thinking, Planning, Writing, and Revising stages of the writing process. This could include sharing brainstorming notes from the Thinking stage, submitting outlines and graphic organizers from the Planning stage, providing drafts before and after AI use, or submitting a brief explanation describing how AI supported their writing when turning in final drafts.

Equip students with practical strategies for using AI meaningfully, ethically, and selectively.

• Teach students how to use AI tools effectively and ethically.
• Explicitly teach students how to evaluate AI-generated output. Analyze examples together to identify where AI supports the writing process (e.g., brainstorming, organizing, revising) and where it falls short.
• Model and think aloud about the effective and thoughtful use of AI during writing tasks.

 Set clear expectations for AI use and diversify writing tasks to promote originality.

• Establish a clear classroom policy regarding acceptable AI use. Be explicit about guidelines, boundaries, and potential consequences.
• Include specific directions in writing assignments about when and how AI can be used.
• Assign alternative writing tasks that combine writing with digital, visual, or auditory literacies (e.g., creating infographics, websites, or multimedia presentations). These forms of expression are less easily replicated by AI and require students to think critically and communicate original ideas.

Writing Instruction Remains Essential

As schools navigate the rapidly evolving landscape of generative AI, it is important to remember that writing instruction is more essential than ever. While AI can be a powerful tool to support the writing process—from brainstorming and planning to revising and editing—it cannot replace the deep learning that occurs when students do the work of writing themselves. Students still need to learn how to generate and express their own ideas, craft clear and coherent sentences and paragraphs, and structure their writing effectively. These skills are critical for academic success.

Educators play an important role in helping students understand how to use AI to support their writing without allowing it to replace their own thinking and creativity. Providing clear guidelines, designing purposeful assignments, and offering explicit instruction in the ethical and effective use of AI ensures that students remain active participants in the writing process. Ultimately, the goal is to develop skilled, reflective writers who view AI as a tool for learning and communication—not a shortcut around it.

To learn more, read the full article: Writing in the Age of AI.

References:

Biancarosa, C., & Snow, C. E. (2006). Reading next—A vision for action and research in middle and high school literacy: A report to Carnegie Corporation of New York (2nd ed.). Washington, DC: Alliance for Excellent Education.

Graham, S., & Perin, D. (2007). Writing next: Effective strategies to improve the writing of adolescents in middle and high schools – A report to Carnegie Corporation of New York. Washington, DC: Alliance for Excellent Education.

Graham, S. and Hebert, M.A. (2010). Writing to read: Evidence for how writing can improve reading. A Carnegie Corporation Time to Act Report. Washington, DC: Alliance for Excellent Education.

Graham, S., Bollinger, A., Booth Olson, C., D’Aoust, C., MacArthur, C., McCutchen, D., & Olinghouse, N. (2012). Teaching elementary school students to be effective writers: A practice guide (NCEE 2012- 4058). Washington, DC: National Center for Education Evaluation and Regional Assistance, Institute of Education Sciences, U.S. Department of Education.

Graham, S., Harris, K.R., & Santangelo, T. (2015). Research-based writing practices and the Common Core: Meta-analysis and meta-synthesis. The Elementary School Journal 115 (4). (pp. 498-522

Graham, S., Kiuhara, S. A., & MacKay, M. (2020). The effects of writing on learning in Science, Social Studies, and Mathematics: A meta-analysis. Review of Educational Research. 90, (2).

Graham, S. (2024b). Why should we teach writing in an age of artificial intelligence? Inspired Ideas, McGraw Hill. https://medium.com/inspired-ideas-prek-12/why-should-we-teach-writing-in-the-age-of-artificial-intelligence-c49c50300584

Hayes J.R. (1996). A new framework for understanding cognition and affect in writing. In C.M. Levy & S. Ransdell (Eds.) The science of writing: Theories, methods, individual differences and applications (pp. 1-27). Mahwah, NJ: Erlbaum.

Hayes, J.R. (2004). What triggers revision? In L. Allal, L. Chanquoy, & P. Largy (Eds.), Studies in writing: Vol. 13. Revision: Cognitive and instructional processes (pp. 9-20). Norwell, MA Kluwer.

Hebert, M. (2025). Using writing as a tool for improving reading and learning. EdWeb webinar February 8, 2025. https://media.edweb.net/edWebinar/?view=20250206edwebnet15

Sedita, J. (2019). The strands that are woven into skilled reading. Retrieved from: https://284ivp1abr6435y6t219n54e wpengine.netdna-ssl.com/wp-content/uploads/2020/02/The-Strands-That-Are-Woven-Into-Skilled-WritingV2.pdf

Sedita, J. (2020). Keys to content writing. Rowley, MA: Keys to Literacy.

Sedita, J. (2023). The writing rope: A framework for explicit writing instruction in all subjects. Baltimore, MD: Paul H. Brookes.

Sedita, J. (2025). Writing in the age of AI. Keys to Literacy.