Draft for review. This is a baseline template, not legal advice. Have it reviewed by counsel before launch.
Security
We know you may bring sensitive research into KinetixAI, so we design the service to protect it. This page describes how we approach security. It describes our design intent, not a certification.
Last updated: October 2026
Your research stays yours
Papers, notes, and repositories you add to your Research RAG are stored in your workspace and used only to answer your questions. We do not use your uploaded content to train models, and we do not share it with other users.
Encryption
Traffic between your browser and KinetixAI is encrypted in transit using HTTPS. We protect stored data with access controls appropriate to its sensitivity.
Authentication and access
Accounts are protected by passwords that we store only as salted hashes, and you can sign in with Google. Access to production systems is limited to the people who need it. [Add multi-factor and SSO details as they ship.]
Safe retrieval
Our retrieval pipeline checks for prompt injection and poisoned documents so that answers stay grounded in your sources rather than following hidden instructions.
Reliability
We design for durability of your workspace data. We do not currently publish a formal uptime commitment outside a written service agreement.
Reporting a vulnerability
If you find a security issue, please email hello@kinetixai.ai with the details and steps to reproduce. We will acknowledge your report and work with you on a fix. Please give us reasonable time to respond before any public disclosure.
What we are working toward
We are building toward independent security reviews and formal certifications. We will state clearly which ones we hold once they are in place, and we will not claim a certification we have not earned.
Contact
Security questions: hello@kinetixai.ai.
Questions? Email hello@kinetixai.ai.