In 2024, researchers published more than 15,000 papers on the topic of AI writing assistants, and a recent survey of teachers by The Learning Agency found that only about half of them felt that they had time to provide students with effective feedback on their writing. These are clear signs that researchers and schools alike are hungry for insights and guidance on how to use emerging AI tools for writing support.
As AI continues to touch nearly every sector, its potential to support teachers and learners is especially exciting, and literacy instruction has surfaced as one of the fastest-growing areas for AI applications.
The Evolution of AI in Writing Instruction
AI tools for teaching writing have evolved significantly in recent years. Rather than simply redlining typos or grammatical errors, today’s AI-powered assistants are capable of offering personalized feedback through conversational interactions with users. These tools can provide essential scaffolding directly to students, enabling them to practice foundational skills.
With AI now able to provide immediate feedback, real-time data, and curriculum-aligned resources, it has also become an excellent tool for educators to feel more empowered and equipped in their work. For example, with the development of more precise datasets that account for distinct organizational elements, such as a paper’s thesis or supporting claims, AI algorithms can better understand and break down each step of the writing process. This, in turn, helps educators better assess student progress and tailor their feedback and instruction to each student’s needs.
This article is based on a workshop from the Rethinking Reading: AI for Literacy Achievement workshop series, a set of webinars on education AI applications organized in collaboration with InnovateUS and the Burnes Center for Social Change at Northeastern University.
Interested in Learning More?
- Dive into this topic some more and experience the tool demo by viewing the full session.
- Catch up on other AI tools for literacy education through other webinars featured in the Rethinking Reading series.
AI Tools For Teaching Writing
The development of AI writing systems can be traced back to basic, rules-based spell-checking systems that emerged between the 1950s and 1970s. With the shift to today’s language-generating models, like ChatGPT and Gemini, these tools have become both more widely accessible and increasingly sophisticated. These platforms have now become intelligent systems able to understand large amounts of information on human language and offer real-time suggestions on writing elements beyond spelling and grammar, such as style and argumentation. This has made it so that these tools are better able to support evidence-based writing instruction through the provision of timely, actionable, and specific feedback.
Khanmigo and Quill are two examples of leading AI tools for literacy education. Khanmigo offers a writing coach for real-time student feedback and has recently made this tool available to teachers, providing them with detailed insights. Developed in collaboration with educators, the tool was first launched as a pilot in 2023, and tested across 53 United States districts. With the support of Microsoft, the tool is now available to teachers for free.
These platforms have now become intelligent systems able to understand large amounts of information on human language and offer real-time suggestions on writing elements beyond spelling and grammar, such as style and argumentation.
Quill provides free, standards-aligned writing instruction with AI-driven feedback across six different learning tools: Quill Connect, Quill Lessons, Quill Diagnostic, Quill Proofreader, Quill Grammar, and Quill Reading for Evidence. These tools enable teachers to improve their instruction and support students with writing practice, honing in on elements such as sentence structure, clarity, and reading comprehension. Just last year, a study by College Board found that students using Quill Connect made meaningful gains in sentence-combining skills, a key component of writing development.
Other major players innovating in this space include Grammarly, OpenAI, and CommonLit.
What’s The Future For AI Writing Tools?
As developers continue to refine their innovations, they are navigating a number of critical questions on how to improve performance and best integrate AI writing tools into classrooms.
In particular, developers need to be attentive to the meaningful, research-based integration of interventions into educational contexts. As tools advance, this means that it is essential that AI be applied thoughtfully, mirroring proven instructional strategies, such as providing specific, personalized feedback and embedding writing skills across the curriculum.
Innovators are also exploring how to best design experiences that are both student- and teacher-centered, ensuring that tools are created to reduce burdens on teachers. The results to date are promising, with one study finding that a writing feedback tool can reduce the amount of time teachers spend on grading by half.
AI can be a powerful ally in writing instruction, not replacing educators, but supporting them with timely feedback, data-driven insights, and personalized learning opportunities for students. It empowers teachers to focus on higher-level instruction while offloading routine tasks.
Perhaps most importantly, the reach and impact of these tools hinge largely on how accessible they are and how fairly they perform across different communities of learners. With free options like Khanmigo, ChatGPT, and Quill.org, writing support is certainly becoming more widely available. However, as the field advances, researchers are looking closely at how to reduce bias in the ways algorithms respond to different student populations, such as multilingual learners.
At The Learning Agency, our team has taken an innovative approach to advancing AI writing tools through data science competitions like the Feedback Prize and the Automated Student Assessment Prize (ASAP) 2.0. These efforts have produced high-performing algorithms for writing feedback and assessment, with many now rivaling human accuracy. These have also drawn interest from collaborators and other vendors, such as Google, looking to apply datasets and algorithms for their own work.
Beyond this work, the Tools Competition, a Renaissance Philanthropy program operated by The Learning Agency, also continues to surface and support promising tools for improved writing support, such as Smart Paper, ThinkCERCA, and Short Answer.
AI Tools for Teaching Writing Are Here to Help, Not Take Over
AI can be a powerful ally in writing instruction, not replacing educators, but supporting them with timely feedback, data-driven insights, and personalized learning opportunities for students. It empowers teachers to focus on higher-level instruction while offloading routine tasks.
Ultimately, these tools – especially when linked to proven educational practices – create opportunities for more meaningful educator-student interactions, removing friction from the feedback loop and amplifying each student’s potential to succeed as a writer.
