Enterprise Knowledge Search
SchoolAI needed to help teachers personalize instruction and spot struggling students earlier without letting AI simply hand answers to students or create unsafe, opaque classroom behavior.
Reported impact
SchoolAI reached 1 million classrooms in 80+ countries, embedded itself in hundreds of education partnerships, saved teachers more than 10 hours a week, and gave educators earlier intervention signals without removing them from the learning loop.
SchoolAI designed its classroom platform as a teacher-in-the-loop agent system rather than a generic chatbot. Teachers use Dot to create Spaces, where lesson goals are translated into interactive activities and apps, while students work through those activities with Sidekick, an AI tutor built on GPT-4o and GPT-4.1. Under the hood, each student interaction runs through an agent graph with dozens of specialized nodes that call models, tools, or guardrails before returning an answer, keeping the system focused on coaching instead of answer-giving. GPT-4o handles conversational lesson construction and response logic, GPT-4.1 handles deeper reasoning such as multi-step math scaffolding, image generation produces classroom visuals, and text-to-speech delivers spoken feedback in more than 60 languages. Logs and reports remain observable to teachers and administrators, so instructional data, safety controls, and intervention signals flow back to the humans responsible for student outcomes.
Teachers prompt Dot to create Spaces and lessons, then students interact with Sidekick inside those environments. Each student input passes through the agent graph and the appropriate OpenAI models before a guided response is returned, while logs, reports, and intervention signals stay visible to teachers and administrators.
"If a teacher types ‘create a differentiated reading activity for students at three different levels,’ Dot will assemble a ready-to-use lesson in seconds."
"In just two years, SchoolAI has reached 1 million classrooms across 80+ countries"
"Students engage with these lessons through Sidekick, an AI tutor built on GPT-4o and GPT-4.1."
"every student input runs through an agent graph with dozens of specialized nodes"
Review how this reference was scored and trace each normalized claim back to its documented source.
Scoring methodology
Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.
Traceability
Field coverage shows which facts each source supports. Dates are shown only when documented by the source.
Related
OpenAI Sample Aug 26, 2024 Arizona State University Today, ASU is enhancing educational outcomes by integrating ChatGPT Edu into projects across teaching, research, and operations. Guided by the ASU charter, which prioritizes inclusion over exclusion, research benefiting… ChatGPT Edu OpenAI Sample Apr 1, 2024 Zelma Zelma needed to make fragmented K-12 assessment data usable for parents, educators, and policymakers because the information was scattered across portals and formats and was too hard for most people to query or… around a 77 percent proficiency rate GPT-4 · OpenAI API · vector database 
Microsoft Sample Oct 15, 2026 Bar-Ilan University Students who miss lectures struggle to reconnect new lessons with prior material when course information is spread across notes, recordings, slides, and readings, and lecturers need a reliable way to answer routine… 71 courses Azure AI Services · Microsoft Azure · Microsoft Copilot Studio 