Even when AI-powered instructional instruments are knowledgeable by the science of learning, adverse penalties are nonetheless attainable. AI can cut back significant instructor–scholar interplay and encourage over-reliance on automated help whereas weakening college students’ unbiased considering and perseverance.
These instruments can amplify biases, restrict publicity to various concepts, or generate inaccurate explanations that college students will not be geared up to query. Additionally, the great amount of knowledge required to prepare these instruments raises privateness and surveillance issues, and unequal entry to high-quality AI instruments can widen present inequities. Teachers may develop into overly depending on automated suggestions, undermining their skilled judgment and the relational and human-centered features of learning that analysis persistently reveals are the most necessary.