Practitioner Voices Summit: How Teachers Evaluate AI Tools through Deliberative Sensemaking

Teachers develop nuanced evaluation criteria when given proper support: When 61 K-12 mathematics educators were given dedicated time and space to deliberate, they generated over 200 criteria for evaluating AI tools organized around four key themes (Practical, Equitable, Flexible, and Rigorous), demonstrating that teachers can be sophisticated decision-makers about AI integration when positioned as such.

Productive tensions reveal the complexity of responsible AI use: Teachers identified inherent trade-offs in AI tool design—such as balancing personalization against fairness and adaptability against efficiency—suggesting that responsible AI integration requires acknowledging competing values rather than seeking single "best" solutions.

Teachers prefer AI as an assistant, not a replacement for professional judgment: The vast majority of educators framed AI tools as assistants rather than coaching tools for their own professional learning, indicating teachers want to maintain agency and decision-making authority in their classrooms.

Five specific conditions enable teachers to make sound AI decisions: Deliberative sensemaking was supported by: dedicated time and space for discussion, using concrete tools and artifacts to ground thinking, collaborative reflection with diverse perspectives, opportunities for knowledge-building, and psychological safety to voice concerns—conditions that could be replicated in other educational settings.

Teacher expertise and confidence reinforce each other in AI evaluation: As teachers built knowledge about AI capabilities and limitations, they made more grounded evaluative judgments, while exercising agency in evaluation strengthened their ability to learn—suggesting professional development on AI should emphasize both technical understanding and teacher decision-making authority.