AI Transparency

Understanding the AI systems that power your school assessments

Our Commitment to Transparency

SchoolQualityReview.ai uses multiple AI models to analyze your documents and generate assessment reports. We believe you should understand exactly how these systems work, what models we use, and what their capabilities and limitations are.

Below you'll find detailed “model cards” for each AI process in our pipeline, following industry best practices for AI documentation.

Assessment Pipeline Overview

Your documents flow through these stages. For a user-friendly walkthrough, see How It Works.

1
Document Extraction — Extract text and visual content from uploaded files into structured markdown
2
Indicator Assessment — Evaluate document sections against quality indicators
3
Report Generation — Synthesize findings, summaries, and recommendations

AI Model Cards

Document Extraction

Multimodal document understanding with vision

Model

Claude Sonnet 4.6

Provider

Anthropic

Extract text and visual content from uploaded documents, including charts, diagrams, tables, and images. The model "sees" the document as a human would, understanding both text and visual elements.

Indicator Assessment

Quality indicator evaluation and evidence extraction

Model

Claude Haiku 4.5

Provider

Anthropic

Evaluate document sections against quality indicators. Determines relevance, extracts supporting evidence quotes, and assigns scores based on alignment with indicator criteria.

Report Generation

Findings synthesis, executive summary, and recommendations

Model

Claude Sonnet 4.6

Provider

Anthropic

Generate the final assessment report including indicator findings, executive summary, and actionable recommendations. Synthesizes all assessment data into coherent, useful narratives.

Voice Transcription

Instant web demo — spoken school name to text

Model

Cloud Speech-to-Text

Provider

Google Cloud

In our instant web demo, a visitor can speak their school's name and city instead of typing a website. This model transcribes that short audio clip to text. Audio is held in memory only for the duration of the request and is never stored.

School Discovery

Instant web demo — name + city to official website

Model

Claude Sonnet 4.6 (web search)

Provider

Anthropic

Resolve a spoken or typed school name and location to the school's official public website, so the demo can analyze the right site. Uses a single Claude call with web search.

Live Web Assessment

Instant web demo — scoring from a public website

Model

Claude Opus 4.7

Provider

Anthropic

Score a school against quality indicators using only its public website, in a single call, with evidence quotes. This powers the ~2-minute instant demo (a preview of the full upload-based assessment).

PII Redaction

Instant web demo — removing personal data before analysis

Model

Cloud Data Loss Prevention (DLP)

Provider

Google Cloud

Redact personally identifiable information (such as names, emails, and phone numbers) from crawled public web content before it is scored or shown, even though the source pages are already public.

Data Handling & Privacy

Your documents are never used to train AI models. All data is encrypted in transit and at rest. Our AI provider (Anthropic) acts as a data processor with SOC 2 Type II compliance. Google Cloud is also a data subprocessor for supporting services such as speech-to-text and PII redaction in our instant web demo.

View full Security & Privacy details

Important Disclaimer

AI-generated assessments are tools to support, not replace, professional judgment. Reports are based solely on uploaded documentation, should be reviewed by qualified professionals, and are most valuable as a starting point for improvement planning.

Have questions about our AI systems?

Contact Us