Carpe Diem game on the Syntra platform. A collection of tasks for students and AI tools

Abstract

This work is being developed as a collection of exercises accompanying the learning material [1]. The exercises are intended for both students and AI tools (LLMs), with the text of each exercise also serving as a prompt for an LLM. In this way, an LLM can assist the student and suggest possible solutions, while the student can critically evaluate how successfully the LLM solves the given task. The exercises also make it possible to compare and assess cases in which LLMs, based on their own interpretation of a task, propose substantially different solutions.
A second and central objective is to offer students the opportunity to learn languages of their own choice while also actively engaging with tasks involving languages that are only slightly familiar or entirely unknown to them. In such situations, students are encouraged to draw on their general linguistic knowledge and linguistic intuition. They learn to orient themselves in a situation in which they do not know the particular language but can nevertheless formulate and test hypotheses by analysing its forms and structures, comparing them with other languages, and examining solutions proposed by AI. The aim of these exercises is therefore not only the acquisition of a particular language, but also the development of general linguistic reasoning: the ability to address linguistic questions and find one’s way through unfamiliar linguistic situations even when the language itself has not yet been learned.
A third objective is to explore and develop, in practical terms, forms of human–machine collaboration in solving linguistic problems. The student is not merely a recipient of answers supplied by AI: the student formulates the task, evaluates the proposed solution, detects errors and ambiguities, corrects them, and, where necessary, prompts the AI to seek an alternative solution. AI, in turn, functions not only as a source of information but as a partner in a joint process of linguistic inquiry. In this sense, the exercises also become experiments in developing forms of collaboration between humans and machines.