Security by Design
Privacy isn't a feature we bolted on. It's the reason Beaver AI exists.
No Audio. Ever.
Zero microphone access. Zero audio files. Zero audio data of any kind.
Not Used for AI Training
Your data will never be used to train AI models — by us or anyone else.
Yours, Always
Your meeting data belongs to you. Delete it any time — it's gone for good.
How Beaver Works — The Privacy-First Architecture
Most meeting tools work by recording your audio, uploading it to a server, and running speech-to-text. That creates an audio file of your conversation that lives on someone else's servers. Beaver doesn't do that.
Beaver reads captions, not audio
Beaver joins your meeting as a participant and reads the live text captions generated by the meeting platform (Google Meet, Microsoft Teams). The caption stream is text-only — no audio signal is ever accessed.
Captions are stored securely, linked only to your account
The text captions are stored in an encrypted database, accessible only to you (and team members you invite). No one else can see them — not even us in normal operation.
AI summarisation happens under strict data controls
When the meeting ends, the caption text is sent to an AI model (Anthropic Claude) via API to generate a summary. Anthropic's API data processing agreement prohibits using this data for model training. The prompt and transcript text are not retained by Anthropic.
Our Security Commitments
All data is encrypted in transit using TLS 1.2+
All data is encrypted at rest
Passwords are hashed using industry-standard algorithms — we never store plain-text passwords
OAuth tokens (for Google Calendar) are encrypted at rest and can be revoked at any time from your settings
We do not have access to your meeting platform credentials
We conduct regular dependency and security reviews
We follow the principle of least privilege — internal access to production data is minimised and logged
Found a Vulnerability?
We take security reports seriously. If you discover a potential security issue, please disclose it responsibly by emailing us at [email protected] with the subject line "Security Disclosure". We'll respond within 48 hours and work with you to address the issue promptly.
Please don't publicly disclose security issues before giving us a reasonable opportunity to fix them.
Questions?
If you have questions about our security practices, email us at [email protected] or use our contact form.