Last updated: April 2026
Saya operates inside real classrooms with real children. Safety is not a feature — it is the foundation everything else is built on.
Saya never operates autonomously. Every session is initiated, configured, and supervised by a credentialed teacher. The teacher can silence Saya, override its tone, issue private instructions mid-session, or end the session instantly at any moment.
Teachers have a dedicated private channel (invisible to students) through which Saya reports real-time risk assessments. If a student's contribution contains language that suggests distress, bullying, crisis ideation, or safeguarding concerns, Saya flags it to the teacher immediately with a suggested action — without alerting the class.
Saya's system prompt contains non-overrideable instructions that prohibit generating: violent content, sexual content, self-harm instructions, politically charged rhetoric, hate speech, or commercially biased content. These constraints cannot be removed by teachers, students, or any session configuration.
Saya adjusts vocabulary, sentence complexity, and cognitive depth based on the grade band configured by the teacher. A Grade 2 class receives simpler, warmer language than a Grade 11 class. Word and sentence ceilings are enforced at the model level — not just in the UI.
Teachers configure the class culture setting (faith-sensitive, secular, or mixed). Saya adapts accordingly — avoiding content that conflicts with the cultural context of the classroom. Saya does not promote any religion, political ideology, or cultural norm above others.
Saya does not have a public-facing interface for students. Students cannot create accounts, send messages to Saya outside class, or interact with the platform independently. All interactions happen through the teacher's dashboard, in a live classroom context.
If you observe Saya producing content that appears unsafe, inappropriate, or in violation of these commitments, please report it immediately:
All safety reports are reviewed by a human member of our team — not an automated system.
Saya routes requests through carefully selected AI inference providers including Anthropic (Claude), Google (Gemini), and others. Each provider is bound by their enterprise data processing agreements and content policies. Classroom transcripts are sent to these providers only to generate Saya's next response and are never used for model training.