AI is rapidly changing education and research needs to keep up
The article argues that the rapid adoption of generative AI in classrooms, with nearly two-thirds of teachers using AI, has outpaced the evidence base for its effectiveness. Traditional randomized controlled trials (RCTs) are ill-suited for evaluating AI tools because these tools are unstable, rapidly evolving, and used in varied ways. The author advocates for implementation research and development (implementation R&D) as a more agile approach, focusing on iterative testing, real-time data capture, and design-focused experiments before large-scale impact evaluations. Examples include IES-funded generative AI centers and partnerships like Leanlab Education and Boston University's EVAL initiative. The article provides guidelines for building evidence for AI tools, emphasizing the need for actionable evidence now rather than perfect evidence later.
Editorial responsibility
- No named human review is recorded for this page.
- Reports are grouped by semantic similarity and deterministic rules. Language models may assist titles, summaries, translation and cross-source analysis; the page itself is projected from evidence records.
- Current automated evidence projection