How AI Works at Tutor Volt

Last updated: February 2026

Tutor Volt uses artificial intelligence to generate educational content, provide tutoring support, and adapt to each student's learning needs. This page explains what our AI does, how it works, and the safeguards we have in place.

1. How We Use AI

Tutor Volt uses AI to transform teacher-uploaded materials into structured lessons, quizzes, flashcards, and mind maps. When students use the AI tutor chat, the system applies the Socratic method to guide them toward understanding rather than giving answers directly. The AI also adapts content difficulty and language complexity based on the student's year group and learning profile.

AI is a tool in our platform, not a replacement for teachers. It generates content from source materials that teachers provide, and all output is designed to support — not replace — classroom teaching.

2. Our Models

We use AI models built by Anthropic (the company behind Claude):

• Claude Haiku: A fast, lightweight model used for quick tasks such as routing requests, classifying content, and generating chat suggestions.
• Claude Sonnet: A more capable model used for detailed content generation, including lessons, quizzes, flashcards, and tutoring conversations.

No student data is used to train these models. Anthropic does not use API inputs or outputs for model training. The AI processes your conversations to provide tutoring help within the session, but this data is not retained by the model provider for any other purpose.

3. Safety Measures

We take student safety seriously. Our AI systems include the following safeguards:

• Real-time content monitoring: All AI-generated content is checked for appropriateness before being shown to students.
• Safeguarding detection: The system monitors conversations for indicators of risk or harm and can escalate concerns to designated safeguarding leads.
• Age-appropriate language: The AI adjusts its vocabulary, sentence complexity, and tone based on the student's year group and key stage.
• Content filtering: The AI is instructed to stay within the boundaries of the lesson content provided by the teacher. It will not discuss topics outside the curriculum or provide inappropriate content.
• Anti-anthropomorphism: The AI does not claim to have feelings, emotions, or consciousness. It identifies itself as a learning tool, not a person.

4. Your Data and AI

When you use the AI tutor, your conversations are processed to provide tutoring help within that session. Here is what you should know:

• We do not store more data than is necessary to provide the service.
• Your school's designated safeguarding lead can review AI tutoring conversations if a safeguarding concern is flagged.
• Teachers can see summaries of student engagement with AI features, but not the full content of chat conversations (unless a safeguarding review is triggered).
• No personally identifiable information is sent to the AI model provider. Conversations are processed without names, emails, or other identifying details.

For full details on how we handle your data, see our Privacy Policy and Data Processing Agreement.

5. Content Generation

AI-generated educational content follows a clear pipeline with human oversight at every stage:

1. Teachers upload materials — lesson slides, notes, or other source content are uploaded to the platform.
2. AI generates structured content — the system transforms raw materials into structured lessons with learning objectives, worked examples, quizzes, flashcards, and mind maps.
3. Teachers review and approve — all generated content is held in draft until a teacher reviews and publishes it. Teachers can edit, reject, or regenerate any content.
4. Students access approved content — only teacher-approved content is visible to students.

6. Human Oversight

AI does not operate autonomously in Tutor Volt. Human oversight is built into every stage:

• Content approval: All AI-generated lessons, quizzes, and flashcards are reviewed by qualified teachers before publication. No AI content reaches students without teacher approval.
• Safeguarding review: Safeguarding alerts triggered by the AI monitoring system are reviewed by the school's designated safeguarding lead, not handled automatically.
• Teacher control: Teachers set preferences for how the AI tutor behaves in their classes, including hint levels, tone, and whether to use the Socratic method.
• Transparency: Teachers and school administrators have access to analytics showing how students interact with AI features, including independence metrics and hint usage.