Trust and Privacy
We don't just claim to protect your content—we measure it
Measured. Tested. Proven to Protect Your Course Materials.
Most AI tools claim to protect your content. We actually measure it.
In internal testing, we track content leakage and exposure risk, which are both scored regularly and targeted at <0.1%.
Similarity Matching
Between generated text and source materials
Structural Copying Score
Tracks verbatim instructor material in responses
<0.1% Target
Content leakage and exposure risk
Security Features
How We Protect Your Content
Zero-Log Data Handling
Your course materials are processed in a temporary, zero-retention environment. Once processing is complete, original files are deleted. No long-term storage, no training on your content.
Instructor-Aligned Reasoning
The AI follows your teaching method and terminology, ensuring students learn your approach rather than shortcuts from generic AI tools.
Original Files Deleted After Processing
Files are processed into a temporary index and then permanently deleted. The system only retains the processed index for active courses, which is also deleted when you remove the course.
No Training of External AI Systems
Your content never trains external AI models. Course Companion operates in isolation, ensuring your intellectual property stays protected and never leaks into public AI systems.
Built for College Courses, Not Generic Chatbots
Designed specifically for higher education with FERPA-aligned practices and academic integrity built-in from the ground up.
Compliance & Standards
Built for Higher Education Standards
FERPA-Aligned Practices
We follow Family Educational Rights and Privacy Act (FERPA) guidelines for handling student data. Student workspace data is never used to train external models, and we maintain strict data handling protocols.
SOC 2 Type II Roadmap
We are designed to grow into SOC 2 Type II compliance as the platform scales. Our architecture and processes are built with enterprise security standards in mind.
Content Leakage Protection
Advanced monitoring and testing ensure your content never leaks into external systems. We track content leakage and exposure risk, targeting <0.1% in internal testing.
Regular Leak-Penetration Tests
We regularly run leak-penetration tests on all major data formats to ensure your content stays protected.
PDFs
DOCX
LaTeX
Spreadsheets
Annotated documents
Problem-set images
Structured and unstructured text uploads
We prompt leakage tests to protect your particular phrasings, reasoning steps, and exam questions. We actively defend against your content leaking into external AI systems.
Questions About Security?
Contact our security team for detailed documentation and compliance information