Looking Beyond the Output: How Preservice Teachers Assess the Instructional Value of AI-Generated Formative Tasks

• In Event: Virtual Poster Session
  In Virtual iPresentation Only: SIG-Technology as an Agent of Change in Teaching and Learning Virtual Poster Session

Abstract

As generative AI becomes a part of education, preservice teachers (PSTs) must learn to critically evaluate its outputs. This study examined how 85 undergraduate PSTs assessed AI-generated formative assessments. Grounded in an updated TPACK framework for generative AI and Vygotskian theories of scaffolding, we analyzed student reflections and prompt revisions from an undergraduate education course on differentiated instruction. Findings revealed strong pedagogical reasoning but limited technological awareness: most PSTs accepted ChatGPT outputs at face value, rarely questioning the tool’s instructional logic. These results highlight the need for teacher education programs to embed structured, critical engagements with AI as a co-design partner—advancing both AI literacy and purposeful technology integration.

Authors

• Molly Li, University of Delaware

• Leigh Hibbard, University of Delaware

• Rachel A. Karchmer-Klein, University of Delaware