How Callie handles your data
Callie is Calendly's AI assistant for the work before and after meetings. It helps you schedule over email, prepare for meetings with briefings and summaries, track action items, and analyze your meeting history. All of this happens through natural conversation.
The core question is simple: Can you trust Callie with your work and your data? The answer is yes. Here's why.
You stay in control
You decide when to use Callie and what tasks to ask it to help with. Callie follows your instructions. Whether you're asking Callie to schedule a meeting, summarize a past conversation, or prepare for your day, Callie works only with the information you provide and the context in your Calendly account.
Whenever Callie takes action on your behalf, such as scheduling a meeting or sending a message, it asks you to confirm first. You review what's about to happen and make the final decision. This keeps you informed and in full control of what gets done.
Your data is not sold or used to train AI models
This is important: We do not sell your data. We do not use it to train AI models. We do not allow third-party AI providers to train on it. Your information stays yours.
Callie accesses only what you need it to
Callie works with information in your Calendly account to help with what you ask. Depending on your request, this may include:
- Your scheduling preferences, availability settings, and event types
- Calendar busy times and invitee information (names, emails, replies)
- Email conversations and meeting summaries
- Your meeting history and action items
Sensitive data is protected
Callie uses Calendly's existing security infrastructure to keep your data safe. Here's what that means in practice:
- Your credentials stay protected. System credentials are never exposed to the AI model. Calendly's services handle authentication behind the scenes.
- Data is encrypted when stored. This includes email addresses, email content, meeting summaries, action items, and scheduling details.
- Risky actions require your approval. Scheduling conflicts, message sending, and other higher-risk actions all need your review and confirmation before Callie acts.
- Guardrails prevent abuse. Callie includes protections against unsafe scheduling behavior, malicious links, and sensitive data mishandling.
- Security testing validates protection. New features undergo threat modeling and security testing throughout the product development lifecycle.
AI can make mistakes
AI isn't perfect. Always review AI-generated responses, especially when accuracy matters. Before using Callie's scheduling suggestions or sending messages on your behalf, check that they're correct and match what you intended.