Camille Faivre stands at the forefront of modern education management, navigating the complex intersection of digital innovation and traditional pedagogy. As institutions grapple with the post-pandemic reality of e-learning, her expertise has become essential for schools seeking to balance the convenience of new technologies with the foundational goal of developing sharp, independent minds. In this conversation, we explore the shifting landscape of student behavior, the ethical gray areas of automated assistance, and the urgent need for cohesive institutional policies in a world where the classroom is increasingly defined by algorithms.
Since usage rates for AI tools among middle and high school students have surged recently while college usage remains steady, what factors are driving this younger demographic to adopt these tools so rapidly? How do their motivations for using AI differ from those of older students?
The surge from 48 percent to 62 percent in student usage over just a few months is a clear indicator that younger learners are increasingly treating these platforms as a secondary support system. Middle and high school students often face a high volume of diverse subjects and may feel more pressure to quickly decode complex instructions or find immediate explanations that their textbooks might not provide. For them, the motivation is often about immediate clarity and overcoming the initial “blank page” paralysis that comes with adolescence. While college students have more established, specialized study habits, these younger students are leaning on tools to bridge gaps in their foundational understanding, often reporting that they use it to get better explanations of their assignments. It is a sensory shift in how they experience homework, moving from a solitary struggle to a more interactive, albeit digital, dialogue.
A growing number of students believe that AI tools are actively eroding their critical thinking skills, yet they continue to use them for schoolwork. How can educators help students navigate this contradiction, and what specific exercises can be used to ensure AI enhances rather than replaces deep analysis?
It is a striking paradox when 67 percent of students admit that these tools might be harming their ability to think critically, yet they cannot seem to put them down. To break this cycle, we have to move away from assessments that only prize the final answer and instead focus on the “logic trail” a student takes. Educators can introduce exercises where students are required to “fact-check” an AI-generated summary, specifically looking for the subtle hallucinations or oversimplifications that 78 percent of non-users are so worried about. By making the critique of the machine part of the grade, we turn the tool into a laboratory for analytical thought rather than a shortcut to completion. This approach helps students feel the weight of their own intellectual agency again, proving that while a machine can draft, only a human can truly discern and validate truth.
While chatbots are the primary choice for most students, many still rely on dedicated writing helpers and homework assistance platforms. What are the distinct academic risks associated with each type of tool, and how can students determine which platform is appropriate for a specific assignment?
The landscape is highly fragmented, with 60 percent of students favoring chatbots like ChatGPT and Gemini, while smaller segments utilize writing helpers or general homework platforms like Chegg or Brainly. The primary risk with chatbots is the “illusion of competence,” where a student receives a polished explanation that feels right but lacks the nuance of the actual curriculum. Writing helpers like Grammarly, used by 21 percent of students, can sometimes sanitize a student’s unique voice, making their work feel sterile or overly formulaid. For homework platforms, the risk is a direct hit to academic integrity, as 15 percent of students might be tempted to pull pre-existing answers without engaging with the problem-solving process. Students need to learn that if an assignment requires creative synthesis, a chatbot is a better brainstormer; if it requires technical precision, a writing helper is a better editor; but if it requires original thought, no platform should be the primary driver.
Most students distinguish between using AI to brainstorm and using it to obtain direct answers, which they view as cheating. How should institutions define the boundary of academic integrity in this new landscape, and what practical steps can teachers take to create AI-resistant assessments?
The data shows a clear moral divide: nearly 80 percent of students believe using AI for understanding is fair game, while 45 percent admit that getting direct answers is cheating. To protect academic integrity, institutions must codify these distinctions into clear, living documents that evolve alongside the software. Teachers can create “AI-resistant” assessments by grounding questions in very specific, hyper-local contexts or personal experiences that a general language model simply won’t have in its training data. We should also lean back into oral exams, in-class essays, and collaborative projects where the process is visible and the student’s real-time thinking is the primary metric of success. This shifts the focus back to the human element, ensuring that the 35 percent of students using AI to brainstorm aren’t accidentally crossing the line into total substitution.
Many schools currently lack formal policies, leaving rules to vary by classroom, and research shows that female students often express higher levels of concern about AI’s impact. Why might these gender-based perspectives differ so significantly, and what essential elements must a schoolwide AI policy include?
The discrepancy where 75 percent of female students express concern over critical thinking erosion—compared to 59 percent of their male peers—is a significant data point that suggests young women may be more attuned to the long-term cognitive costs of technology. This heightened concern often translates into a greater worry about the ethics of cheating, which makes the current lack of schoolwide policies even more stressful for them. A unified policy is essential because when only one-third of students have a clear set of rules, it creates an atmosphere of academic anxiety and perceived unfairness. Every policy must include three pillars: a clear definition of “permitted assistance,” a mandatory disclosure requirement for AI-generated content, and a commitment to teaching AI literacy so students understand the “why” behind the restrictions. Without this, we are leaving both students and teachers to navigate a digital wilderness without a compass.
What is your forecast for student AI use?
I forecast that we are entering an era where AI use will move from being a deliberate “event” to a seamless, invisible background feature of all educational software. We will see the 28 percent usage rate of tools like Google Gemini continue to climb as these features are baked into the very documents students use to type their essays. However, this normalization will lead to a “literacy gap” between students who use AI as a collaborator to deepen their inquiry and those who use it as a crutch to bypass the struggle of learning. Success in the next five years won’t be measured by who can block AI the best, but by which institutions can most effectively teach students to audit, challenge, and ultimately master the machines they are already using.
