Is Claude Meetings private? What it saves, who sees it, and outside guests
The Morse teamPublished 9 min read
As of November 2026, Claude Meetings is private from people outside your company, who can't open the meeting page, but not from your company's Claude workspace. Claude saves a transcript, a summary, and screen descriptions of each call, but not the audio (Anthropic). People from your organization on the invite can see that meeting page, and it can be shared with anyone else in the organization (Anthropic). Anthropic doesn't train on it by default (Anthropic). On Team plans, Claude can join calls with people from outside your company unless an owner turns that off (Anthropic).
Anthropic started rolling out Claude Meetings in beta on November 12, 2026, for Team and Enterprise plans in Claude and Claude Desktop, with Google Meet as the only supported platform (Anthropic release notes). Anthropic says other video conferencing platforms are coming soon (Anthropic). We read the launch post, the help center article, and Anthropic's pages on commercial retention, training, data exports, custom retention, audit logs, the Compliance API, and custom roles on November 14, 2026, plus Google's Meet help on removing a participant. Anthropic deserves credit for the defaults it picked. Claude joins as a visible participant, posts a notice in the chat, keeps no audio, and on Enterprise stays out of any call with outside guests until an owner allows it.
Claude Meetings privacy settings at a glance
Each row is from Anthropic's own pages, read November 14, 2026. "Not stated" means the pages we read don't say.
| Setting or data | Default | What's kept, and for how long | Source |
|---|---|---|---|
| Claude Meetings itself | On for everyone on Team; off on Enterprise | Not applicable | help |
| Calls with outside guests | Allowed on Team; internal meetings only on Enterprise | Guests' words go into the notes and transcript | help |
| Auto-join | Each person's own setting; owners can't set it for you | Not applicable | help |
| Audio | Not saved | Nothing | help |
| Transcript, summary, and screen descriptions | Saved to a meeting page in claude.ai | No meeting-specific period stated | help |
| Chats on Team and Enterprise | Kept "in the product" until deleted | Deleted from back-end storage within 30 days of deletion | privacy |
| Enterprise custom retention | "Retained indefinitely unless a custom retention period is set" | Minimum 30 days; covers chats, projects, and artifacts; meeting pages not stated | help |
| Model training | Off for commercial plans | Thumbs-up or thumbs-down feedback keeps the whole conversation up to 5 years | privacy |
| Organizations with ZDR, HIPAA-ready setup, or CMEK | Can't use the beta | Not applicable | help |
Does Claude Meetings record audio?
No. Anthropic says Claude "saves a transcript, summary, and screen descriptions, not audio" (Anthropic). The screen descriptions mean that what someone shares on screen during the call can end up on the meeting page as text, and Anthropic's pages don't describe how detailed those descriptions are.
To ask Claude something during the call, you say "Claude" and then the question, or type it in the Meet chat (Anthropic). Claude's tile in the call is live and shows its current state (Anthropic). To pause it for a sensitive stretch, ask it to stop listening or use the meeting page, and resume from the meeting page (Anthropic). Left alone, Claude can stay for up to eight hours and warns you before it leaves (Anthropic).
Will other people know Claude is in the meeting?
Yes. When Claude joins, it appears as a participant and posts a notice in the call's chat (Anthropic). It joins when someone adds it from their calendar in Claude, or when someone on the invite has auto-join on (Anthropic). Auto-join can cover all of a person's meetings or all of them except one-on-ones (Anthropic).
A Google Meet host can keep Claude out or remove it. If Claude asks to be let in, the host can decline, and Claude doesn't try that call again unless someone adds it (Anthropic). Once it's in, a Google Meet host or co-host can click Show everyone, then More actions next to Claude, then Remove from the call, and choose Block so it can't rejoin that meeting code (Google Workspace). People in your organization can also ask Claude to leave or select Remove Claude on the meeting page (Anthropic).
Does Claude Meetings join calls with outside guests?
On Team plans, yes, by default. Anthropic's help center says "On Team plans, Claude can join calls with outside guests by default, and an owner can turn this off" (Anthropic). On Enterprise, an owner must allow it first (Anthropic).
Anthropic's launch post says that by default Claude "is set to stay out of calls when someone from outside your organization is invited" (Anthropic). The help center's owner settings show that this is the Enterprise default only, with Meetings with external guests set to Internal meetings only on Enterprise and Allowed on Team (Anthropic). If you're a guest on a call with a Team customer, look for Claude in the participant list and its notice in the chat. Anthropic's help center tells the person who adds Claude or turns on auto-join to make sure everyone on the call is okay with being transcribed first (Anthropic).
When guests are kept out, Claude leaves if an outsider joins partway through, and another company's notetaker bot counts as an outsider (Anthropic). When guests are allowed, Claude posts its notice in the chat but doesn't reply there (Anthropic). What guests say goes into the notes and transcript (Anthropic). Guests don't get the meeting page or the recap email, so they don't see the transcript of what they said (Anthropic). Each member also has a personal Meetings with external guests setting under Settings > Capabilities > Meetings (Anthropic).
Who can see a Claude Meetings transcript?
Your colleagues on the invite, and anyone they share it with inside the company. People from your organization on the invite get a recap email with a link, can see the notes, and can edit what Claude made (Anthropic). The meeting page can be shared with others in your organization (Anthropic). After the call, anyone with access can ask Claude about the meeting in a regular chat (Anthropic).
The chat in the call is more public than the meeting page. Anything Claude posts in the Meet chat is visible to everyone on the call, guests included, while the chat on the meeting page stays inside the organization (Anthropic). Claude can also use each person's connected tools, like sending a Slack message, and a yes the first time applies going forward (Anthropic).
Your company manages all of it. Anthropic says the Primary Owner "manages your Work account and all associated data," including exports that "may contain your conversations with Claude" (Anthropic). As of November 14, 2026, the export, audit log, and Compliance API pages don't mention meeting pages either way, so we can't tell you whether they're included. We cover what a Claude Team or Enterprise admin can read in ordinary Claude chats. Anthropic also says it's working with Gong, Attention, Granola, Fireflies.ai, Otter, and Plaud so teams can soon send a meeting's transcript to those tools (Anthropic). Once that ships, the transcript also falls under that tool's rules, like Granola's or Otter's.
Does Anthropic train on Claude Meetings transcripts?
Not by default. Anthropic says it "will not use your inputs or outputs from our commercial products (e.g. Claude for Work…) to train our models" by default (Anthropic). Its Commercial Terms, effective June 17, 2025, say "Anthropic may not train models on Customer Content from Services" (Anthropic).
Feedback is the exception for Claude chats. A thumbs-up or thumbs-down stores "the entire related conversation" for up to 5 years, and Anthropic may use it to train its models (Anthropic). An owner can turn that button off for everyone with the Rate chats setting under Organization settings > Data and Privacy (Anthropic). Our page on which AI companies train on your data compares the defaults.
How long does Claude keep meeting transcripts?
Anthropic hasn't published a period for meeting pages. For Team and Enterprise chats in general, Anthropic says it retains them "in the product," and a deleted conversation is removed from back-end storage within 30 days (Anthropic). Chats flagged for breaking the Usage Policy keep their inputs and outputs for up to 2 years (Anthropic). Anthropic may also keep chats "as required by law" (Anthropic).
On Enterprise, data "is retained indefinitely unless a custom retention period is set," and the shortest period is 30 days (Anthropic). That page lists chats, projects, and artifacts as covered, names several features that aren't, and doesn't mention meetings in either list. If retention matters to your company, ask Anthropic before you rely on it. Stored transcripts are what courts can reach, which we cover in can AI chats be subpoenaed.
Can companies with zero data retention use Claude Meetings?
No, not during the beta. Anthropic says it isn't available to organizations with zero data retention (ZDR), a HIPAA-ready configuration, or customer-managed encryption keys (CMEK) (Anthropic). Our guide to what zero data retention means explains the first of them.
How Morse keeps your own questions off your employer's Claude workspace
Everything Claude Meetings makes lands on a page in your employer's Claude workspace, under your employer's owner, sharing, and retention settings. That's fine for the meeting. The questions you have about your own job afterwards, like how to answer the criticism you got, whether a reorg mentioned in passing touches your role, or how to word a request for a raise, are better asked somewhere else. Morse is an AI chat that keeps those chats in your browser, encrypted, not on Morse's servers.
- E2EE, for the question you'd never want your manager to read. Your browser checks the secure enclave's proof that it's genuine, then seals the message on your device so only that enclave can read it. Morse's servers relay text they can't read. E2EE runs open-weight models like GLM, Kimi, and DeepSeek on enclaves run by third-party providers, not by Morse, and it rests on the enclave hardware and attestation working as designed. It has no web search or image generation, and it comes with Pro, at $19 a month.
- Private, the mode every new chat starts in, for drafting the follow-up you'll send later. Morse has zero data retention agreements with its inference, search, and other providers, so they've agreed to keep nothing. Where a provider confirms zero retention on each reply, Morse checks it and refuses a reply that doesn't.
- Anonymous, to ask Claude itself without your work account behind it. Our proxy removes your account, email, and IP address and sends the question under Morse's own account. The provider sees the message without knowing it's you, and may keep it for a limited time to check for abuse, under its own policy, so leave colleagues' names out.
Morse doesn't train AI on your content, sell your data, or show ads (Morse privacy policy). Because we don't store chats, we usually have very little to give authorities. Clearing your browser or deleting your account removes your chats for good, and Morse can't recover them, so keep a copy of anything you'll need. Use your own device for this, since a work laptop has monitoring layers of its own.
What to change before Claude joins your next call
- Auto-join: in Settings > Capabilities > Meetings > Auto-join, pick all meetings except one-on-ones if you keep it on.
- Outside guests: on Team, an owner can switch Meetings with external guests to Internal meetings only in Organization settings > Products > Claude Meetings (beta), and each member can turn off their own setting.
- A sensitive stretch: ask Claude to stop listening, or select Remove Claude on the meeting page.
- Feedback: owners can turn off Rate chats, so no one can send a whole conversation to Anthropic as feedback.
- Retention: on Enterprise, set a custom period and ask Anthropic whether it covers meeting pages.
- Your own questions afterwards: take them to a Private or E2EE chat on Morse, where the history stays in your browser.
Try a chat from the Morse home page without an account, or start free with no card needed.
Common questions
Is it legal to let Claude Meetings transcribe a call without asking everyone?
It depends on where the people on the call are. Anthropic's help center says to make sure everyone on your call is okay with being transcribed before you add Claude or turn on auto-join. This isn't legal advice.
Can I use Claude Meetings on a Claude Pro or Max plan?
Not as of November 14, 2026. Anthropic's release notes list the beta for Team and Enterprise plans only, in Claude and Claude Desktop.
Does Claude Meetings work in Zoom or Microsoft Teams?
Not yet. Anthropic says Google Meet is the only supported platform for now, with other video conferencing platforms coming soon.
How long can Claude stay in a meeting?
Anthropic's help center says Claude can stay in a call for up to eight hours, and that it warns you before it leaves.