Privacy Policy

Faceport Inc. ("Faceport," "we," "us," or "our") respects your privacy. This Privacy Policy explains how we collect, use, disclose, and protect information when you use our websites, the Faceport iOS app, video meetings, AI experts, personal agents, and connected services. The information processed depends on the features you and your meeting host or organization use.

Faceport is launching for customers in the United States. Account holders must be at least 18. A US launch does not mean all information is stored or processed only in the United States.

We may collect the following categories of information:

• Account information: such as your name, email address, and profile details provided through email, Sign in with Apple, Google sign-in, or other authentication methods. We also process account identifiers, organization membership, and authentication sessions.
• Meeting data: including audio, video, screen shares, chat messages and files, captions, transcripts, recordings, meeting titles and schedules, participant names, join and leave times, and session logs. Recording and saved transcripts depend on meeting settings, plan availability, and applicable consent. Participant activity can include microphone and camera state, speaking activity, reactions, and raised hands.
• Contacts and communications: Faceport contacts, call participants and call history, invitations, guest names and contact details, RSVP answers, messages, delivery status, and support requests. Connected address books and calendars are accessed when you authorize an integration. Invitation and mailing reports can record email opens and content views.
• Photos and other content: profile images, uploaded photos, videos and documents, floor plans, whiteboards, poll responses, and content shared from Faceport Glasses. Files can contain metadata, such as their date or location. Separate website Media Library features can detect faces and group photos of the same person, as explained below.
• Personal-agent information: messages, responses, saved memories, goals, schedules, task results, action and permission records, and usage records. A private call can add a saved transcript to the conversation when you consent.
• Connected sources: account and authorization identifiers and, when you grant access, selected email content, calendar events, contacts, and website content needed for your request. Optional connected-browser features can retain website cookies and local storage in a remote browser context so an authorized session can continue.
• Training materials: documents, URLs, and other content you upload to train and configure your AI expert agents.
• Usage data: such as feature interactions, device and browser information, IP address, and diagnostic logs used to operate and improve the service. iOS calling uses an app installation identifier and Apple push notification tokens to deliver incoming calls to your registered devices.
• Payment information: billing contact details, subscription status, and transaction records when you use paid services or donations. Stripe processes payment details for supported website and meeting payment flows.

We use collected information to:

• Provide, operate, and maintain the Faceport platform
• Configure and run AI expert agents in your video meetings
• Process and index training materials so agents can respond with organization-specific context
• Authenticate users and enforce access controls
• Connect calls, deliver notifications, and synchronize meetings and contacts
• Monitor performance, troubleshoot issues, and improve security
• Communicate with you about the service
• Maintain personal-agent conversations and memories and carry out goals and scheduled work you direct

Personal agents use your messages, relevant conversation history, memories, instructions, and authorized source results as context. Relevant context is sent to OpenAI to provide the requested AI response or work. This can include email text, calendar details, and contacts retrieved from sources you enable. Public web research uses a search query sent to a search provider; it does not send the entire private conversation or mailbox to that provider.

Signing in with Google or Microsoft does not by itself authorize a personal agent to read those accounts. Agent source grants are separate from sign-in and other integrations. Supported personal-agent mailbox access reads selected messages; it does not send or delete mail, mark messages as read, or download attachments. Calendar and contact access is limited to the sources and permissions you enable.

Where connected-browser features are enabled, Browserbase hosts the remote browser and its saved session context. The agent can process website content available through the connection you authorize. Enter website passwords only in the private sign-in flow, not in agent chat. That flow keeps password entry outside the model, but authenticated website content can still become task context.

AI experts can receive meeting audio, conversation text, relevant knowledge materials, and shared-screen images when visual context is enabled. OpenAI processes speech for supported live captions and text for caption translation, summaries, and other AI features. An AI participant is a separate recipient of meeting content, even when a meeting is not being recorded. Consider what is visible on your screen before sharing it.

Uploading a floor plan for a walkthrough can send its image to OpenAI to identify walls and room labels. The uploaded plan and resulting drawing data are saved for your walkthrough. Device motion supports positioning within the plan when you enable the AR feature.

Customer-provided training materials are used to configure and inform your organization's AI agents. We process this content to enable natural collaboration in meetings, not to build generalized AI models for unrelated purposes beyond what is necessary to operate the service for your account.

AI responses are generated using third-party model providers (such as OpenAI). Information sent to these providers is limited to what is required to deliver the requested functionality and is subject to applicable provider terms and data handling practices.

Separate website creative features send prompts and selected reference images to OpenAI or Google for image and video generation. Media Library photo processing can send resized photos to Google Cloud Vision for face detection and face crops to Google Gemini to group the same person across photos. A reference image containing a person can therefore be shared with a provider. This photo-library processing is distinct from ordinary live video calling and local camera effects.

We use trusted third-party services to deliver core functionality, including:

• Apple and Google: supported sign-in methods; Apple also delivers iOS incoming-call notifications
• LiveKit: browser and native-app video conferencing and cloud recording
• OpenAI: AI language processing, voice agents, and meeting summaries
• Browserbase: optional personal-agent remote browsers and saved website sessions
• Google: Gemini AI, Cloud Vision photo processing, and Veo video generation when the corresponding feature is used
• Supabase: database, authentication, and file storage
• Cloudinary: media storage, processing, and delivery
• Resend: transactional email (invitations, reminders, summaries)
• Twilio: text-message delivery and supported telephony features
• Tavily: optional web search tools for AI agents (disabled in HIPAA mode)
• SerpApi: optional LinkedIn profile lookup for persona creation (disabled in HIPAA mode)
• Firecrawl: website content retrieval for supported knowledge features
• Stripe: subscription billing (when enabled)
• Vercel and Railway: application and worker hosting
• Sentry and Upstash: diagnostics and rate limiting when configured
• Google, Microsoft, Zoom, HubSpot, and Salesforce: the calendar, address-book, meeting, email, or CRM integrations you connect
• Snap, LemonSlice, LiveAvatar, and Tripo: selected camera, avatar, and 3D creative features

These providers process data on our behalf according to their own privacy policies and our agreements with them. Organizations that enable HIPAA mode may have additional restrictions on which subprocessors receive meeting or conversation data.

Personal-agent conversations, memories, and connected browser sessions belong to the individual account's private workspace. An organization does not gain access to this content merely by paying for that account. Service operation, authorized provider processing, support, security, and legally required disclosures still apply as described in this policy.

Meeting participants receive the content you publish to their meeting. Hosts and authorized organization administrators can access records available to their role. Links you share can give recipients access to their linked content. Some general Media Library files use public delivery URLs, so a person with that URL can retrieve the file. We may also disclose information when required by law or to protect people, accounts, and the security of the service. Providers may process information outside your country.

Personal-agent conversation history and saved memories persist across sessions. Forgetting a memory removes it from saved memory and excludes its associated source messages from future agent recall; the original messages remain visible in conversation history and exports. Similar information independently present elsewhere may still be available. Deleting an agent initiates cleanup of its conversation and related records. Contact us if you need help removing particular information.

Stop reading revokes a personal agent's source grant. Disconnecting a source removes the authorization held for that connection, but does not erase content already saved in the conversation or revoke every permission you may have granted at the provider. You can also review permissions in your Google, Microsoft, or other provider account. Disconnecting a browser connection queues deletion of its stored browser context. This does not necessarily sign out the session at the connected website or remove copies that website retains.

We retain information for as long as needed to provide the service, comply with legal obligations, resolve disputes, and enforce our agreements. Retention differs by record type and by the features and organization settings in use. A meeting retention setting does not automatically apply to every uploaded file, derived image, summary, integration, or provider copy. Disconnecting an integration does not erase copies already delivered to its provider or other participants.

You can initiate account deletion in account settings. The website control is in Settings, Security. Contact us for help with access, correction, or deletion of specific content, including files shared with an organization. Records that must be retained for legal, security, or accounting obligations, and copies in backups, may have different retention periods. Deleting the app from a device does not delete your Faceport account.

We implement reasonable administrative, technical, and organizational measures designed to protect your information. However, no method of transmission or storage is completely secure, and we cannot guarantee absolute security.

Depending on your location, you may have rights to access, correct, delete, or restrict processing of your personal information, or to object to certain processing. To exercise these rights, contact us at the address below. We may need to verify your identity before responding.

You control camera and microphone access in iOS Settings and in the meeting. Photo selection, Bluetooth, and local-network access support the features that request them. Faceport Glasses pairing and its local photo test do not by themselves post a photo to a meeting. When you enable glasses sharing in a meeting, new captured photos are uploaded to that meeting's group chat. Turn sharing off or leave the meeting to stop further sharing. Other participants may retain content already shared with them.

You can mute your microphone, stop video or screen sharing, decline an optional permission, disconnect an integration, or leave a meeting. Revoking a permission stops future access through that permission; it does not remove previously shared or saved content. If a school or organization provides your access, contact it about records it manages. Parents or guardians can contact us about a child's personal information or concerns about consent.

You can inspect or forget agent memories, manage goals and schedules, and revoke source access using the personal-agent controls. Stopping future work does not undo actions already completed or remove saved results. Contact support for access or deletion assistance.

We use cookies and similar technologies for authentication, session management, and preferences (such as theme settings). You can control cookies through your browser settings, though some features may not function properly if cookies are disabled.

The iOS app stores authentication sessions, preferences, an installation identifier, and cached catalog artwork on your device. Its calling and meeting features do not use the device's advertising identifier. The optional website measurement described below is separate from incoming-call delivery.

On the website, when you allow optional measurement, we use Google Analytics and share completed signup and first paid subscription events with Meta and LinkedIn to measure advertising. This can include campaign identifiers, a hashed email address for matching, and the subscription payment value and currency. A hashed email is a matching identifier, not anonymous data. Meeting content is not included in these advertising events.

You can choose Essential only or change your saved choice using the button below. We honor your browser's Global Privacy Control signal for this optional measurement. Declining does not prevent signup or payment.

We may update this Privacy Policy from time to time. We will post the revised version with an updated effective date. Material changes may be communicated through the service or by other appropriate means.

Faceport Inc., 169 Madison Ave #11637, New York, NY 10016, U.S.A..

Privacy-related questions may be directed to Faceport Inc. at support@faceport.com.

For help using the app, visit our support page. Our Terms of Service explain the conditions for using Faceport.