Simulacre Oposicions Anglès (Lengua Extranjera) #4

Read the text and answer the questions. **Assessment in the Age of AI: Rethinking Language Learning Evidence** During the past two years, generative AI tools have entered classrooms with unusual speed. In many secondary schools, students now use AI to brainstorm essay ideas, check grammar, and generate model responses before submitting final drafts. For language teachers, this shift has triggered a central question: what kind of evidence should count as learning in an environment where polished output can be produced in seconds? Traditionally, language assessment has focused heavily on product: the final essay, the oral presentation, the reading comprehension score. However, when students can outsource part of the production process, product alone becomes a weak indicator of competence. As a result, many experts propose moving towards process-informed assessment, where teachers also evaluate planning notes, revision decisions, and students’ metalinguistic explanations. Proponents argue that this change does not lower standards; it raises them. If learners must justify lexical choices, explain discourse structure, and reflect on register decisions, they are demonstrating deeper control of language than in a one-shot exam script. In this view, AI becomes not a shortcut but a catalyst for richer pedagogical conversations about quality, audience, and communicative purpose. Critics, however, warn that process assessment can increase teacher workload and reduce reliability. Observing drafting habits and giving qualitative feedback demands time that many schools cannot easily provide. There are also equity concerns: students with better digital literacy may use AI more strategically, widening existing gaps. A balanced approach is emerging in current policy discussions. Rather than banning AI or ignoring it, schools can design hybrid tasks that combine supervised in-class production, oral defense, and reflective commentary. This triangulation gives teachers multiple windows into learner ability. Ultimately, the objective is not to protect assessment models from technological change, but to ensure that assessment continues to capture what matters most: meaningful communicative competence.