Security
Your work stays yours.
How Lenz guards your workspace, your Gmail and the people in your pipeline.
Nothing sends without you. Lenz writes drafts. It has no way to send on its own: every email waits for your approval, a draft that changes waits for you again, and each approval sends once, even if you click twice.
Gmail, narrowly scoped. Connecting Gmail is optional and asks for two permissions: read your email (gmail.readonly) and manage drafts (gmail.compose). Revoke it any time from your Google Account.
Encrypted, twice where it matters. Data is encrypted in transit (HTTPS) and at rest by our database provider. Gmail access tokens and your own model key are also encrypted by the app, with AES-256-GCM.
Private workspaces. Only people you invite can see your workspace. Anyone else gets a not found page, not even a hint that it exists.
No passwords to steal. You sign in with a one-time code sent to your email. Codes expire after 10 minutes and are stored hashed, and attempts are rate-limited by one-way hashes, not raw addresses.
AI that doesn't learn from you. Research and drafting use Anthropic's models, which process your content on our behalf and don't train on it. Gmail data is never used to train general-purpose AI models.
Responsible disclosure
Found a vulnerability?
Tell us privately and we'll work with you on it. Please don't test Lenz's security without our written permission; see the Terms. If a breach affects your data, we'll tell you without undue delay.
The details, in full.
What we collect, who processes it and how long we keep it.