Beyond Detection: Designing AI-Resilient Assessments with Automated Feedback Tool to Foster Critical Thinking

Abstract

The growing use of generative AI tools like ChatGPT has raised urgent concerns about their impact on student learning, particularly the potential erosion of critical thinking and creativity. As students increasingly turn to these tools to complete assessments, foundational cognitive skills are at risk of being bypassed, challenging the integrity of higher education and the authenticity of student work. Existing AI-generated text detection tools are inadequate; they produce unreliable outputs and are prone to both false positives and false negatives, especially when students apply paraphrasing, translation, or rewording. These systems rely on shallow statistical patterns rather than true contextual or semantic understanding, making them unsuitable as definitive indicators of AI misuse. In response, this research proposes a proactive, AI-resilient solution based on assessment design rather than detection. It introduces a web-based Python tool that integrates Bloom's Taxonomy with advanced natural language processing techniques including GPT-3.5 Turbo, BERT-based semantic similarity, and TF-IDF metrics to evaluate the AI-solvability of assessment tasks. By analyzing surface-level and semantic features, the tool helps educators determine whether a task targets lower-order thinking such as recall and summarization or higher-order skills such as analysis, evaluation, and creation, which are more resistant to AI automation. This framework empowers educators to design cognitively demanding, AI-resistant assessments that promote originality, critical thinking, and fairness. It offers a sustainable, pedagogically sound strategy to foster authentic learning and uphold academic standards in the age of AI.