{"id":2065910,"date":"2026-07-15T12:45:16","date_gmt":"2026-07-15T10:45:16","guid":{"rendered":"https:\/\/www.luxsure.fr\/2026\/07\/15\/in-schools-ai-does-not-replace-teachers-it-reveals-what-schools-fail-to-measure-properly\/"},"modified":"2026-07-15T12:49:07","modified_gmt":"2026-07-15T10:49:07","slug":"in-schools-ai-does-not-replace-teachers-it-reveals-what-schools-fail-to-measure-properly","status":"publish","type":"post","link":"https:\/\/www.luxsure.fr\/en\/2026\/07\/15\/in-schools-ai-does-not-replace-teachers-it-reveals-what-schools-fail-to-measure-properly\/","title":{"rendered":"In schools, AI does not replace teachers. It reveals what schools fail to measure properly"},"content":{"rendered":"\n

Artificial intelligence is already making its way into classrooms through students, teachers, and after-school activities. The real question is no longer whether it will be adopted, but exactly what it claims to improve. <\/p>\n\n

In a classroom, artificial intelligence rarely makes a grand entrance. It creeps in through an assignment written too quickly, a review sheet generated at 11 p.m., a lesson plan drafted between meetings, or a video tailored to a student\u2019s interests. It doesn\u2019t always ask the school for permission. It settles in wherever the system leaves gaps: teacher burnout, varying student ability levels, the pressure of testing, and an obsession with results. <\/p>\n\n

This is precisely what makes the debate about education more complex than the debate about technology. AI does not emerge within a perfectly stable system that it would simply optimize. It enters a school system already strained by its own tensions: lack of time, teacher shortages, classes too diverse for a one-size-fits-all approach, and difficulty in assessing anything other than final performance. <\/p>\n\n

The Estonian case, cited as one of the most advanced in Europe, illustrates the scale of this shift. According to data cited during the panel discussion, 90 to 95 percent of Estonian high school students are already using generative AI every week. The government has therefore not so much \u201cintroduced\u201d AI as it has chosen to respond to a fait accompli. Perhaps this is the first lesson: a blanket ban does not eliminate usage; it merely drives it underground. <\/p>\n\n

The Illusion of Personalization<\/h2>\n\n

The word is on everyone\u2019s lips: personalization. It promises a school system that is finally tailored to each individual, free from the uniform pace of the classroom. It\u2019s a powerful argument. In a traditional classroom, some students fall behind because the teacher goes too fast; others get bored because the teacher goes too slow. AI, then, seems to offer an elegant solution: adjusting the content, varying the examples, posing questions, and supporting students outside of school hours. <\/p>\n\n

But this promise should be taken with a grain of salt. Personalization isn\u2019t just about modifying the surface of content. It\u2019s about understanding what\u2019s holding a child back, what\u2019s causing resistance, and what\u2019s at play in a child\u2019s relationship with knowledge. A machine can adapt an example; it doesn\u2019t always know how to recognize the fear of failure, hidden boredom, mimicking behavior, or the shame of not understanding in front of others. <\/p>\n\n

The risk lies in confusing personalization with optimizing the path toward the same destination. In this case, AI does not broaden learning. Instead, it streamlines, accelerates, and standardizes in a different way. It gives each person a seemingly different path to reach the same metric. <\/p>\n\n

What AI Does Well<\/h2>\n\n

However, it would be too simplistic to reject the whole idea. The most convincing examples do not involve the teacher being replaced, but rather a specific, limited, and context-specific use. <\/p>\n\n

A tool that tests a student on the material covered in class from multiple angles can be a great study aid. A teacher who records their explanations, transcribes them, and then turns certain passages into learning materials can capitalize on their own creativity instead of letting it go to waste during class. A video tailored to a student\u2019s personal interests can sometimes capture their attention in a way that a generic format could not. <\/p>\n\n

These uses have one thing in common: they build on a preexisting human framework. The teacher remains the one who shapes the learning progression. AI does not invent pedagogy; it extends, reformulates, questions, and archives. It becomes interesting when it serves a purpose that is already clear. <\/p>\n\n

The problem arises when we ask it to make up for a lack of intent. A poor AI-assisted course remains a poor course. A half-hearted instruction yields a half-hearted result. In education, as in the luxury sector, the tool does not automatically elevate the hand that wields it. <\/p>\n\n

Details<\/h2>\n\n

During the panel discussion, one of the speakers estimated that only 30 to 40 percent of high school students find a real use for tools designed to pose questions rather than provide direct answers. This figure highlights a key tension: a tool that is pedagogically sound is rarely the one students spontaneously choose when they are evaluated based on the final result. <\/p>\n\n

The Real Issue: Assessment<\/h2>\n\n

This may be where AI is most helpful to schools: not by solving their problems, but by exposing them with a newfound bluntness.<\/p>\n\n

As long as the system primarily evaluates the final product\u2014a submitted essay, a completed exercise, a correct answer\u2014the rational student will seek the shortest path to that result. Generative AI excels precisely in this area: producing a plausible, fluid, presentable form. It thus becomes less of a learning tool and more of a compliance accelerator. <\/p>\n\n

So the question isn\u2019t just, \u201cAre students using ChatGPT to cheat?\u201d It\u2019s a more uncomfortable one: \u201cWhy are our assessments so easily replicated by a machine?\u201d <\/p>\n\n

If we want AI to support learning, we\u2019ll need to shift our focus to the process: drafts, the thought process, verbalizing ideas, justifying answers, successive errors, and the ability to question an answer. In other words, we should evaluate not so much the finished product as the hand that shaped it. <\/p>\n\n

School is not a platform<\/h2>\n\n

The debate takes on even greater depth when we remember that education is not limited to the acquisition of skills. It certainly equips students with skills. But it also socializes. It teaches us to think with others, sometimes against others, and often in front of them. It also fosters a form of individuation: the ability to find one\u2019s place in the world, to form judgments, and not merely to give the correct answer but to become a thinking individual. <\/p>\n\n

That\u2019s where individual screens can become a trap. Thirty students, thirty computers, thirty personalized learning paths: it sounds modern. But it could also describe a classroom where no one looks at their neighbor anymore. The school would then lose what is most time-honored and precious about it: the art of being present. <\/p>\n\n

The issue is not to ban technology. It is to refuse to let it become the invisible framework of every educational setting. Estonia, in fact, seems to be maintaining this cautious approach: according to the data cited, more than 70% of learning time there is still spent in face-to-face settings. This ratio speaks to something essential. AI can enter the classroom, but it must not become its central feature. <\/p>\n\n

Train Before Deployment<\/h2>\n\n

The practical conclusion is almost stark: we must train teachers before training students to use these tools. Not just how to click on an interface, but how to understand what these systems actually do: they do not understand, they predict; they do not judge, they imitate; they do not guarantee the truth, they produce plausibility. <\/p>\n\n

This literacy must be technical, practical, and human. Technical, so we understand how the models work. Practical, so we can decide when to use them or reject them. And, above all, human, so we can assess their effects on attention, children\u2019s rights, the environment, the teacher-student relationship, and trust. <\/p>\n\n

Clear certification of educational tools is therefore essential. No device intended for children should be allowed in the classroom based solely on a marketing promise. There should be evidence, safeguards, transparency regarding data, and an independent assessment of the benefits and risks. <\/p>\n\n

Schools don’t need AI to do their thinking for them. They need a way to decide what is still worth thinking about carefully. <\/p>\n\n

The question that remains, then, is not whether AI will have a place in education. It already does. The real question is a more challenging one: Will schools be able to limit its role enough to preserve what only they can still impart? <\/p>\n\n

\"\"<\/figure>\n","protected":false},"excerpt":{"rendered":"

Artificial intelligence is already making its way into classrooms through students, teachers, and after-school activities. The real question is no longer whether it will be adopted, but exactly what it…<\/p>\n","protected":false},"author":2,"featured_media":2063587,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_lmt_disableupdate":"","_lmt_disable":"","footnotes":""},"categories":[77972],"tags":[79289,79286,78663,79287,79288],"class_list":["post-2065910","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-luxury-and-ai","tag-ai-literacy","tag-edtech","tag-japanese-square-knot","tag-personalized-learning","tag-teachers"],"_links":{"self":[{"href":"https:\/\/www.luxsure.fr\/en\/wp-json\/wp\/v2\/posts\/2065910","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.luxsure.fr\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.luxsure.fr\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.luxsure.fr\/en\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.luxsure.fr\/en\/wp-json\/wp\/v2\/comments?post=2065910"}],"version-history":[{"count":0,"href":"https:\/\/www.luxsure.fr\/en\/wp-json\/wp\/v2\/posts\/2065910\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.luxsure.fr\/en\/wp-json\/wp\/v2\/media\/2063587"}],"wp:attachment":[{"href":"https:\/\/www.luxsure.fr\/en\/wp-json\/wp\/v2\/media?parent=2065910"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.luxsure.fr\/en\/wp-json\/wp\/v2\/categories?post=2065910"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.luxsure.fr\/en\/wp-json\/wp\/v2\/tags?post=2065910"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}