Camille Faivre is an education expert who helps institutions navigate the complexities of modern learning, particularly in the post-pandemic shift toward e-learning. With a deep focus on education management, she has become a leading voice in the conversation around integrating technology thoughtfully into the classroom. Her work explores the critical balance between leveraging powerful new tools like artificial intelligence and preserving the essential human connections that lie at the heart of teaching and learning, especially in the deeply personal act of writing. Today, she shares her insights on the promises and pitfalls of using AI in writing education, arguing that while AI can be a helpful assistant, it should never become the final judge. She discusses the risk of reducing writing to a transactional task, the importance of authentic audiences, and practical ways teachers can use AI to support, rather than supplant, the creative process.
Your article opens by questioning if AI should grade papers, even if it can. You quote a student, Jane, who notes AI can’t recognize growth or effort. Could you share an anecdote where your human understanding of a student’s journey led to a different assessment than an automated score?
I vividly remember a student who was working on a poetry assignment. He was usually quite reserved, but this topic—writing for a sick grandparent—brought out a new level of determination in him. When I saw his final draft, I could see the raw emotion poured onto the page, but from a purely technical standpoint, a rubric-based AI would have found it flawed. The meter was a little off, a metaphor didn’t quite land, but I saw beyond that. I saw the hours of struggle, the breakthrough in his willingness to be vulnerable, and the immense personal significance of the piece. An automated score would have missed all of it. It would have quantified the mechanics but completely ignored the heart of the work and the growth of the writer, which is precisely what Jane meant—that is not what actually goes into writing.
You highlight a student’s stark comment: “If you can use AI to grade me, I can use AI to write.” How does this transactional mindset harm the writing process, and what concrete steps can teachers take to re-center writing as a meaningful, human act in their classrooms?
That comment from Wyatt really gets to the core of the problem. When grading becomes automated, the entire act of writing risks becoming a mechanical exchange. The student’s goal shifts from communicating an idea or telling a story to simply satisfying an algorithm. It teaches them that writing is not a human act of connection but a task to be completed for a grade. To counteract this, teachers must re-center the process and the audience. First, we need to treat AI as an in-process guide, not a final judge. Second, and more importantly, we must create assignments where students write for someone other than the teacher or a rubric. When students are writing for their peers, their parents, or their community, the stakes change. They are no longer just trying to get a good score; they are trying to be heard.
The article suggests AI is best used in the developing stages, like brainstorming or offering alternative endings. Can you provide a step-by-step example of how a teacher could structure a lesson using AI as an in-process “guide or aide” without it defining the final product?
Absolutely. Imagine a narrative writing unit. A teacher could first have students write a complete first draft of a short story on their own. For the revision lesson, they could structure it like this: First, students identify a part of their story they feel is weak—maybe a clunky transition or an uninspired ending. Next, they use an AI tool with a very specific prompt, something like, “Provide five alternative endings for a story that concludes with this paragraph.” The crucial third step is human interaction. Students would bring these AI-generated options to a small peer-review group to discuss. They wouldn’t just copy and paste; they would analyze why the AI made those suggestions and use the discussion to brainstorm a new, original ending that is truly their own. The AI becomes a catalyst for thought, not the author.
You express concern that when students write for a bot, their audience “doesn’t matter.” What are two or three specific writing assignments you’ve found effective for connecting students with a broader, real-world audience, such as their community, peers, or even a panel of judges?
Moving beyond the teacher-as-audience is critical. One incredibly effective assignment is a community-action project. I’ve seen students research a local issue they care about—say, the need for a new crosswalk near their school—and then write persuasive letters to the city council or articles for the local newspaper. Suddenly, their audience isn’t a bot or a rubric; it’s their neighbors and local leaders. Another powerful assignment is creating a class literary journal. Students write poems and stories, then act as an editorial board to select, edit, and publish the work for their peers and parents. In both cases, the writing has a life beyond the classroom. The students see their words have the power to create change or evoke emotion in real people, which is a lesson no automated grading tool can ever teach.
What is your forecast for the future of AI in the writing classroom? As technology continues to evolve and promise efficiency, what is the most critical piece of advice you have for educators trying to maintain the balance between helpful tools and the essential human connection in teaching?
I believe AI’s presence in the classroom will only become more sophisticated and integrated, with tech companies continuing to promise efficiency, like grading hundreds of essays in an hour. The temptation to outsource the more time-consuming parts of our job will be immense. My most critical piece of advice for educators is to hold fast to the principle that AI should support the writing process, not define the product. We must be the ones who use these tools with intention, always as in-process feedback to enhance learning, never as the final word. We have to remember that our students are not data points; they are people with stories to tell and arguments to make. We must never confuse the cold efficiency of an algorithm with the warmth of human understanding and empathy. The relationship between a teacher and a student writer is where the real learning happens, and no technology can ever replace that.
