Two days ago, Anthropic's office-suite announcement ended with one line: Claude Meetings is coming soon. On November 12, 2026, it arrived — in beta, on Team and Enterprise plans, on Google Meet first. Claude can now join your video call as a visible participant, follow the conversation, and start on the work while people are still talking.
I haven't used it. Nobody outside Anthropic has had it long — the beta only started rolling out yesterday, and your account may not show it yet. This article is built from Anthropic's announcement, its Help Center guide and release notes, plus Axios's reporting, checked 13 November 2026. Where something is unconfirmed, I'll say so plainly.
This is the part that separates Claude Meetings from the note-taker bots most teams already have. A transcript bot sits quietly, records, and sends you a summary after everyone hangs up. Claude Meetings works during the meeting:
You add Claude to a call from the Meetings page in Claude, which pulls from your Google Calendar, or you turn on auto-join and it shows up on its own. Anthropic's docs say it can stay in a call for up to eight hours, with a warning before it leaves.
Anthropic's product lead for knowledge work, Nicholas Lin, told Axios the company tested this internally for about two months — and that Claude now joins nearly all of Anthropic's own meetings. That's the kind of claim that's easy to make and hard to verify from outside, but it's also the kind of claim you only make if the tool actually works in your own building.
The beta is narrow, and the narrowness is deliberate:
Anthropic hasn't published a separate price for Claude Meetings. The beta terms, as of November 12:
What happens after February 10 — whether artifacts start counting against usage, or get their own price — Anthropic hasn't said. If your team builds its meeting workflow around live artifact creation, put a reminder in for early February. A free window that quietly ends is how tools get expensive.
A visible participant is the right design, and it's the one detail I'd insist on: everyone on the call can see Claude is there. It doesn't keep the audio — it saves a transcript, not the recording. That's a real distinction, though it only matters if you trust the distinction; a transcript of everything said is still a complete record of the meeting.
The admin controls are the part to evaluate before the features. On Enterprise, Meetings is off until an admin enables it, and admins can restrict what it does. If your security team is the reason you can't use AI tools at work, this is the paragraph to send them — but they'll want to verify it themselves, and they should.
What changes: the meeting stops being a conversation you later reconstruct. The doc, the whiteboard, the action items with names attached — they exist while the decision is still being made, when someone can still say "no, that's not what I meant." Anyone who has ever watched a meeting's decisions dissolve into three different people's notes knows how much that matters.
What doesn't change: the meeting still needs a purpose, an agenda, and someone willing to make a decision. An AI that takes perfect notes of a pointless meeting gives you a perfect record of a pointless meeting. And the artifact it builds is a draft — the same rule as every other Claude output this week: check it before it becomes the plan of record.
Of the three Anthropic launches this week — the office suite on Monday, Manufacturing on Wednesday, Meetings on Thursday — Meetings is the one that changes how a team actually spends its day. Sheets and Whiteboards are tools you open; Manufacturing is paperwork you delegate. Meetings is the first one that sits in the room with you.
The test is simple, and it's the same test I'd apply to any meeting tool: after one real meeting, is the doc Claude built the doc your team actually uses — or did someone rebuild it properly afterwards anyway? If it's the first, Claude just bought every team back the hour after every meeting. If it's the second, it's a fancier transcript bot.
If the beta has reached your account, try it on a planning meeting, not a status update. Planning meetings are where the live artifact matters — the plan is the output, and building it while everyone is still talking is the whole point. And check the artifact before it becomes the plan. Always check the artifact. Sources: Anthropic — "Introducing Claude Meetings, now in beta" (https://claude.com/resources/articles/claude-meetings), published 12 November 2026; Anthropic Help Center — "Use Claude Meetings"; Anthropic release notes, November 2026; Axios — "Anthropic wants Claude to join your meetings", 10 November 2026 (product lead Nicholas Lin interview). All capability, rollout and pricing claims are Anthropic's, stated as such. No hands-on testing was performed for this article. This breaking article is not the daily buyer-intent post, and it contains no affiliate links. Plain link for readers: claude.com.