Artificial intelligence is reshaping how educators assess students, according to Temple Lovelace, executive director of the Assessment for Good program at the Advanced Education Research and Development Fund. The program focuses on formative assessments for learners aged eight to thirteen and has been exploring AI’s potential since the mid‑2020s.
Lovelace highlights several AI use cases for assessment. These include generating test questions, measuring durable skills such as critical thinking, and speeding the scoring process. She also notes that AI can create greater coherence across assessment systems, compiling a more complete, integrated portrait of a learner over time to inform teachers and school leaders when planning subsequent lessons.
One notable application is the creation of inventive question formats that move beyond traditional Scantron‑style tests. Over the past two and a half to three years, the team has used AI to design items that involve Socratic conversations, where the learner speaks and the system evaluates their thought process. This approach, coupled with rapid summative scoring, disrupts the typical delay in delivering results, which can often take weeks or months.
In addition to faster scoring, AI provides richer information. Lovelace explains that large language models can supply initial insights and new scoring models, delivering unique inferences to learners and educators in real time. The program’s recent “Future of Assessment” forecast envisions assessments that go beyond raw reading or math scores, capturing how students apply critical thinking and collaboration. Students interact with chatbots, upload work samples, and the system records evidence of cognitive flexibility. Finally, AI aggregates data from disparate subject systems, enhancing coherence and building a comprehensive learner portrait.
