A short overview of how Cortox protects your account and your data.
All traffic to and from Cortox is encrypted over HTTPS/TLS. There is no unencrypted path into the application.
Cortox is multi-tenant software: every account's customers, products and proposals are scoped to that account only.
Sign-in uses one-time passcodes by email (or Google Sign-In) instead of long-lived passwords, and login attempts are rate-limited.
Your data is never used to train AI models — not by Cortox, and not by our AI provider, Anthropic, whose commercial terms prohibit it.
Payments are processed by Stripe — Cortox never stores your card details.
Cortox is hosted on Railway, Vercel and Supabase, which operate their own physical and network security programs.
Our dependencies are scanned for known vulnerabilities as part of our normal development process.
No. Cortox does not train any AI models on your data. To answer a request, the AI assistant sends your question and the business data needed to answer it to our AI provider, Anthropic, through its commercial API. Under Anthropic's Commercial Terms of Service, Anthropic may not train its models on customer content from its services — so your customers, products, proposals and chat messages are not used to train Anthropic's models either.
If you believe you've found a security issue, please email support@cortox.ai — we take these reports seriously and will respond promptly.