“There are jobs and exercises that I see that help me learn something. I do those. But others that seem unnecessary to me… I tell the AI: do this job for me, I copy, paste and deliver it.” Lucía, an intermediate degree student in the field of health who prefers to remain anonymous, tells it bluntly. Last year he started use AI for their jobs. Since then, some she has made, but many others she has not.
It is not an isolated case. In many educational centers, generative artificial intelligence has become an everyday tool. ChatGPT, Gemini and other assistants have become normalized among students to do homework, summaries or papers, just as before they did it Wikipedia or search engines. The difference is that now they not only find information: they also write it.
From primary to university
“They clearly use it,” says Nerea Eguiguren, a professor of Philosophy and History at a high school in Madrid, referring to the use of these tools among her students. At its core it is something widespread, but he describes this use as “superficial”: “They don’t even open ChatGPT. They put the question on Google and the first answer is from Gemini. They copy it and tell you whatever.”
At the university the phenomenon does not go unnoticed either. “The use of AI is widespread,” explains Silvia Eva Agosto Riera, associate professor at the Faculty of Education at the Complutense University of Madrid. Students use it to search for information, write texts or correct work. Some responsibly; others, simply transferring what the tool gives them without contrasting anything. Sergio Cuevas del Valle, a doctoral student in Aerospace Engineering at the Rey Juan Carlos University and a teaching collaborator there, agrees, who is sure that his students use it: “The question is, why don’t they use it?”
Meanwhile, in other areas of teaching the impact seems more limited. Marta Benegas, a secondary school Fine Arts teacher, notices it every day. “They don’t use AI as much because they basically draw. To draw you need the notebook and the pencil.” When this use appears, it is usually for the same thing as in other subjects: preparing written work.
And the impact of AI is not the same in all subjects. In more theoretical subjects – such as language, philosophy or history – its impact is more noticeable, because many of its traditional exercises, like essays or text comments, are precisely the type of tasks that AI can solve with ease. On the other hand, in more practical subjects the margin for “copying” is smaller: drawing, solving problems step by step or practicing procedures requires demonstrating the process.
(Unsplash)
Lucía has verified this in her most practical subjects and evaluations: “In many cases, if you don’t have a basis, no matter how much you ask the AI, you won’t be able to understand it. You can ask for steps or instructions, but if you’ve never done it yourself, you won’t know.”
In primary school the debate is still in a different phase. The age still slightly limits autonomous use of these tools. Belén Álvarez, a teacher at a school in the Canary Islands, admits that she did not even want to mention AI in her classroom until recently. “I didn’t want them to know her because of me,” he says. Their youngest students are eight years old, but half already have mobile phones with internet access. “Honestly, AI doesn’t seem like the most dangerous thing they have access to.”
Given this presence almost omnipotent artificial intelligence In the educational field, teachers find new challenges when it comes to assigning homework and assignments. Teaching tries to adapt to the new scenario, which has led to rethinking the way of evaluating what students really know.
Has the end of homework and jobs come?
In many cases, the reaction has been immediate and direct. Faced with the reality of being able to solve assignments—which were previously assigned as homework—in a matter of seconds with the help of AI, one of the quickest solutions has been to bring those tasks back to the classroom.
Nerea Eguiguren did it after detecting it several times. “Before, I sent text comments home in the second year of high school. The third time I saw that they used AI, I changed.” Now, although he continues to send those exercises, they do them in class: “This way I know they can’t use it.” More face-to-face exercises, more oral activities, less homework or more practical evaluations, all these adjustments are repeated at different educational levels.
The detection tools of AI too They have become allies of teachers, who use them above all to supervise more theoretical work – although most of the time as a simple support, since they are aware that their reliability is also limited. And when its presence is evident, the consequence can also be direct. “Of course I have suspended jobs due to improper use of AI (…) You don’t have time to suspend the evaluation, but I have suspended many jobs,” says Eguiguren.
It also affects the University. (Unsplash)
Sergio Cuevas del Valle has also had to “pose everything differently”: “Almost any problem that I may pose as a challenge will have already been posed, and almost certainly, solved. It is very likely that the students could find it even without AI.” For this reason, it is proposed how “AI comes to question even the figure of the teacher, and even that of the students, to the extent that it allows human beings to have no need to accumulate internal knowledge, nor do we need someone to teach it to us.”
All of this “underlines the need to rethink teaching at all levels,” trying to ensure that students “work on skills such as the development of intuition, logical thinking and capacity for effort, which were already inherent to ‘homework’.” AI can solve mechanically almost any problem, “but you still need someone to ask the right questions.”
To these new approaches, Silvia Eva Agosto adds the so-called “AI-resistant” tasks—considered in the indications for good use of the UCMa sample of the protocols that educational institutions are putting in place. These are exercises in which “(current) weaknesses of generative AIs are exploited in such a way that repeated human intervention becomes necessary if the objectives of the task are to be achieved.”
Meanwhile, from the students’ side, some of these measures have a limited effect. Lucía, as an undergraduate student, says that some teachers have begun to ask for jobs handwritten. “It doesn’t matter,” he admits, “I can ask ChatGPT and then write it.” In this sense, Eguiguren confesses how “many times one gets tired of the role of AI detector police. I say why it is important that they not use it, I say what they can use it for, and if they use it and I don’t realize it, then that’s it.”
In earlier stages, the impact, at least for now, appears minor. Belén Álvarez has not yet changed the type of duties she commands. His students still do not master these tools well. In fact, when one day he tried to generate an image with ChatGPT in one of his classes, the reaction was surprise. “They freaked out,” he remembers.
(Unsplash)
The problem, many teachers agree, is that all these changes are being made on the fly. The syllabus remains the same, the pace of the course is not reduced and the time to rethink the evaluation is limited. “I think we are missing those times of reflection and considering how to address the issue,” laments Eguiguren. The time required to “change the structure and evaluation, as well as rethink the way of teaching” is “too much for the speed at which we are going.”
Adapting to technology that evolves so quickly has become, for many teachers, just another task on an already full calendar. “If 30 students have to present a work to me, I need five classes, and those are five classes that I don’t teach. What do I do with that syllabus that I’m not teaching?” Eguiguren asks.
Between help and shortcut
Beyond how homework or the way of evaluating changes, many teachers point out a deeper problem: what happens when students begin to delegate part of their mental processes to these tools. Summarizing a text, selecting information or contrasting sources are tasks that require reading, thinking and deciding. When the process is reduced to copying an AI-generated response, that journey disappears. “There is no residue left behind,” Eguiguren summarizes.
In primary and secondary school the brain is still developing, especially in areas related to critical thinking and decision making. Marta Benegas remembers that during adolescence the prefrontal cortex continues to mature. In this context, suddenly facing large volumes of information—and seemingly perfect answers—can make it difficult for students to learn to discriminate sources or construct their own criteria.
At the same time, almost no teachers propose eliminating the tool. They themselves often use it daily; to prepare rubrics, analyze results or save time on bureaucratic tasks. “Sometimes we are a bit hypocritical,” admits Eguiguren. “We teachers also use it a lot.” In that sense, the debate does not revolve so much around banning artificial intelligence as learning to use it judiciously.
(Unsplash)
Because, as Benegas summarizes, AI is increasingly similar to other technologies that are already part of everyday life. It can save time, facilitate tasks and speed up processes. The problem appears when it completely replaces effort. Or, as Sergio Cuevas says, when learning stops being studying or understanding something: “Learning is no longer even studying to pass; just typing to ask for the answer that, with a very high percentage, will be correct.”
More than the end of homework, what many teachers see is a deepest crisis of the model of evaluation. The Internet had already changed the way we access information, but artificial intelligence has given another twist. “The evaluation system has been in crisis for a long time,” says Silvia Eva Agosto.
The question that begins to appear is: what does it mean to learn when a tool can generate correct answers in seconds. For Cuevas, the debate affects the very nature and essence of academia: “In the coming years, the university faces the task of redefining why it is necessary to go to university exactly,” he says, and “AI is exposing that question in a more intense light.”
For their part, secondary school teachers, given that “AI is here to stay,” see it necessary to seek “a cross-school strategy” that is aimed at finding the best procedures and consensus in evaluation models, but also in education on these issues.
Image | Jessica Lewis
GIPHY App Key not set. Please check settings