Verified · Nov 15, 2026
Independently verifiedElevenLabs previews ElevenAI — an AI teammate that listens to meetings, takes notes with everyone's consent, and hands back next steps; free early access, still in preview
2 sourcesElevenLabs' summit post (November 11, 2026) introduced ElevenAI in what the post itself calls a "sneak preview": "It functions as an AI teammate for enterprise teams; it listens to spoken conversations, including in-person meetings, and turns those into notes and next steps." Per the post: "ElevenAI offers dictation and meeting notes powered by Scribe v3, ElevenLabs’ newest and most accurate Speech to Text model. In meetings, it takes notes with everyone’s consent, then delivers a summary and action items for each attendee when the call ends." The post states: "Free early access to ElevenAI is opening today." — dated to the post's own November 11 publication. TechCrunch (November 13) supplies the current state: "ElevenAI is still in preview, and you have to sign up for early access instead of just downloading an app" — plus the category frame ("Fireflies, Wispr Flow, and Granola have all released competing capabilities in the dictation and meeting note-taking space" in the last few months), the dictation figure "support for over 90 languages" (TechCrunch's number; the official post states none), and the ElevenAgent follow-up example ("writing a draft email based on meeting notes").
Why now
The dictation-and-meeting-notes category has a new entrant with an unusual pedigree: TechCrunch names "voice AI giant ElevenLabs" "The newest entrant in that category" against Fireflies, Wispr Flow, and Granola, and the official post announced free early access opening the day of the November 11 summit. TechCrunch's Friday piece confirms the current state — still in preview, signup-based — so creators who cover productivity tools get a fresh launch with an official record to quote from, and the consent mechanics give the review an angle most notetaker coverage skips.
Why it is worth publishing
Creator-tool audiences live in this category daily, and the card's evidence graph is unusually clean: the official post carries the product mechanics and the consent design, TechCrunch carries the market frame and the current preview state, and the two documents' renderings stay separated. An early-access notetaker review is also a consent discussion the audience can act on — who agreed, what gets recorded, where the audio goes.
Evidence basis
Two opened sources, one official, both read in full this run: ElevenLabs' summit post (November 11, 2026; embedded payload datePublished 2026-11-11T12:00:00.000Z; 406,323 bytes raw HTML) and TechCrunch (November 13, 2026; 2026-11-13T14:44:48+00:00; 233,011 bytes). Heat rides the category timing — TechCrunch frames the entrant against Fireflies, Wispr Flow, and Granola — and the product's daily-use fit for creators (meeting notes; dictation with "support for over 90 languages" per TechCrunch), tempered by the preview state: neither opened source claims general availability, pricing, or documented data handling.
“ElevenLabs — the company behind those AI voices — is letting teams sign up for an AI teammate that sits in meetings, takes notes once everyone consents, and hands every attendee a summary when the call ends.”
Angle
A consent-first review of a meeting notetaker: walk the official post's own design statements (consent-based notes, built-in notice and consent, per-attendee summaries) against TechCrunch's market frame and current preview state — then show what early access means in practice, because "still in preview" is the state every claim on this card inherits.
Format
Short talking-head video
Demo idea
A consent-timeline demo: start ElevenAI on a test call you host, show the participant-notification step TechCrunch describes ("notify other participants that the conversation is being recorded"), then the end-of-call output — a summary and action items for each attendee per the official post — and close on the state line: free early access, still in preview, signup-based.
Platform notes
Consent is the non-negotiable: the official post says notes are taken "with everyone’s consent" and that "Notice and consent mechanisms are built in" — present that as the vendor's own design statement, never as an independent guarantee, and never demo the tool on a call where a participant has not agreed. The preview state travels with every claim: TechCrunch's "still in preview" and "sign up for early access" mean no general availability, no pricing, and no documented data handling is claimed by either source — the "over 90 languages" dictation figure, the Fireflies/Wispr Flow/Granola category map, the "$22 billion" valuation, and the "newest entrant" framing are TechCrunch's, attributed wherever reused. Scribe v3's "newest and most accurate" ranking is the official post's own claim, not an independent benchmark. No numbers merge across the two sources.
Usable claims
- ElevenLabs' summit post (published November 11, 2026; calendar-verified Wednesday) states: "In a sneak preview during his keynote, Staniszewski introduced ElevenAI, a new app from ElevenLabs. It functions as an AI teammate for enterprise teams; it listens to spoken conversations, including in-person meetings, and turns those into notes and next steps." Per the post: "ElevenAI offers dictation and meeting notes powered by Scribe v3, ElevenLabs’ newest and most accurate Speech to Text model. In meetings, it takes notes with everyone’s consent, then delivers a summary and action items for each attendee when the call ends." and "It can also give live assistance during a call, sharing context and enhancing collaboration between people and agents. Notice and consent mechanisms are built in, so the people on the other end feel comfortable." The post states: "Free early access to ElevenAI is opening today." — "today" being the post's own November 11 publication date. TechCrunch (November 13, 2026; calendar-verified Friday) describes the current state: "Launched during its 11/11 Summit in New York, ElevenAI is still in preview, and you have to sign up for early access instead of just downloading an app." and frames the category: "In the last few months, Fireflies, Wispr Flow, and Granola have all released competing capabilities in the dictation and meeting note-taking space." Per TechCrunch: "It can take meeting notes across your meetings and notify other participants that the conversation is being recorded."; "Plus, there’s a dictation feature, which lets users convert speech to cleaned-up text in any app with support for over 90 languages."; and "ElevenLabs also allows companies to integrate ElevenAgent and take follow-up actions such as writing a draft email based on meeting notes." TechCrunch also reports: "In the last week, the company also launched its v3 series of Scribe speech-to-text models with live transcription capabilities and the ability for companies to fine-tune models using their own data, a tool to pick the right language model based on use case, and a routine creator that automates tasks based on the transcript."
Evidence pipeline
From the news
Breakdown
Two opened sources, one primary. The official summit post (November 11) carries the product mechanics — the "sneak preview" framing, the AI-teammate positioning, Scribe v3 as the powering model, consent-based note-taking, per-attendee summaries and action items, live assistance, and the free-early-access announcement dated to the post's own publication day. TechCrunch (November 13) carries the current state ("still in preview", signup-based) and everything the official post does not state: the category map (Fireflies, Wispr Flow, Granola), the "newest entrant" and "voice AI giant" characterizations, the "over 90 languages" dictation figure, the ElevenAgent email-draft example, cross-meeting querying, and the $22 billion valuation. What stays open: no opened source documents where meeting audio is processed or stored, no pricing exists for a preview product, and "newest and most accurate" is ElevenLabs' own claim about Scribe v3 — not an independent benchmark. The consent mechanism is the vendor's design statement; the risk card treats it as the load-bearing surface, not a guarantee.
Sources
Risks
- Before you demo or use ElevenAI in any real meeting, get explicit consent from every participant on the record, verify your company's meeting-recording policy and your local recording law, and run your first tests on meetings you host yourself; treat the preview's data handling as unverified until ElevenLabs documents where audio is processed and stored, and say the consent mechanism is the vendor's own design statement — not an independent guarantee — whenever you present it.
Demo ideas
- A category-map card: TechCrunch's frame ("In the last few months, Fireflies, Wispr Flow, and Granola have all released competing capabilities in the dictation and meeting note-taking space.") with each rival row labeled by what TechCrunch says they released, and ElevenLabs' row carrying the official post's own mechanics (Scribe v3-powered dictation and meeting notes, consented recording, per-attendee summaries) — footer: two sources, two layers, no blended claims.
- A consent-flow explainer board: the official post's consent surfaces ("takes notes with everyone’s consent", live assistance "sharing context and enhancing collaboration between people and agents", "Notice and consent mechanisms are built in") beside TechCrunch's participant-notification line, with the open question stated plainly: neither opened source documents where meeting audio is processed or stored.
- A what-early-access-means card: the official post's "Free early access to ElevenAI is opening today." (dated November 11, the post's own publication day) beside TechCrunch's November 13 state ("ElevenAI is still in preview, and you have to sign up for early access instead of just downloading an app") — one line on why preview status changes the weight of every claim.