Otter.ai Review 2026: Free Plan, 90-Minute Caps, an MCP Server, and a Privacy Lawsuit You Should Read Before You Invite the Bot
Otter is one of the best-known AI notetakers and, since April 28, it calls itself a "Conversational Knowledge Engine". This review checks what is true today against Otter's own pricing page: Free is 300 minutes a month, Pro is $8.33 a month billed annually, Business is $19.99, and AI Chat is metered at 20, 50 and 200 queries. Then it adds what real users say, and the August 13 ruling that let core privacy claims against Otter proceed.
| Ownership | Independent private company |
| Free tier | Real -- 300 min / month |
| Cheapest paid plan | $8.33/mo (Pro, annual) |
| Users (Otter-stated) | 35 million |
| Trustpilot record | 4.1/5, 711 reviews |
One of the most established AI notetakers, with a free tier that is honest about its limits. Also a product whose consent model is now being tested in federal court.
"Otter's pitch has moved from 'we transcribe your meeting' to 'we turn every conversation into a searchable knowledge base your AI tools can query'. The transcript part is mature. The new part, chat across meetings, connectors and MCP, is where the plan limits start to bite: 20 AI-chat questions a month on Free, 50 on Pro. And the part that should slow you down is not a feature at all. It is that a bot joining a call is a decision about other people's voices."
My honest read: if you are an individual or small team that lives in Zoom, Teams or Google Meet and you want searchable notes and action items, Otter is a safe first thing to try, and the Free plan is a fair test. If you run long workshops, interview-heavy work, or you operate in two-party-consent or biometric-privacy jurisdictions, read the Security section before you turn on auto-join. This is a VERIFIED EDITORIAL REVIEW: Otter's own pricing page, product pages, YouTube channel and press release checked on Oct 3, 2026, plus court-reporting and published user sentiment. I did not run a controlled hands-on test.
One product, three jobs: capture the meeting, answer questions about it, and push the result into other tools.
Strip away the branding and Otter does three things. It captures a conversation, either live in its app or by sending a notetaker into a Zoom, Microsoft Teams or Google Meet call. It distils that into a transcript, a summary and action items. And since the April 2026 relaunch it wants you to ask questions across everything it has heard, from Otter itself or from ChatGPT and Claude through its MCP server. Otter now calls this a "Conversational Knowledge Engine"; I read that as the same notetaker with a bigger ambition and a more connected chat window.
Ownership: Otter.ai, Inc. is a private company led by co-founder and CEO Sam Liang. I found no acquisition or ownership change in the sources I read, which matters because a neighbouring notetaker, Fathom, was acquired by Superhuman on Sept 14, 2026.
A scaled incumbent in a category that is consolidating, with a legal overhang its rivals share.
Otter press release, Apr 28, 2026. A vendor count; Otter does not define whether this means registered accounts or active users.
Same release. Volume of meetings, not customers or revenue.
Otter-stated milestone in a March 2025 press release. No newer figure was found.
Third-party summaries (Sacra and press): last round a $50M Series B in Feb 2021. Not stated on otter.ai.
| Date | Event | Source |
|---|---|---|
| 2025-03 | Otter says it passed $100M ARR and launches its AI Meeting Agent suite | Otter press release (BusinessWire) |
| 2025-08 to 09 | Four separate privacy class actions filed against Otter over its notetaker | Court reporting (UC Today, law-firm write-ups) |
| 2025-10-22 | Cases consolidated as In re Otter.AI Privacy Litigation, N.D. Cal., Judge Eumi K. Lee | UC Today |
| 2026-04-28 | Conversational Knowledge Engine launch: AI Chat Connectors (MCP client), rebuilt AI Chat, Otter for Desktop (Mac, Windows), expanded MCP server | Otter blog / press release |
| 2026-08-13 | Judge rules core wiretap, CIPA and BIPA claims can proceed; other claims dismissed with leave to amend | RecordingLaw.com, court-ruling summary |
| 2026-09-14 | Superhuman announces it has acquired rival notetaker Fathom | TechCrunch and others |
The market signal I weigh most is the Fathom deal. When a notetaker with a generous free plan sells itself to an email and productivity suite, it tells you the stand-alone notetaker is turning into a feature. Otter's answer, visible in the April launch, is to be the memory layer other AI tools plug into rather than just a transcript archive.
People like the product and the people who answer the support queue. The complaints are accuracy on messy audio, and the billing and consent experience.
4.1 out of 5. 57% five-star, 30% one-star: a polarised split. Praise centres on support staff who resolve things quickly (reviews dated Sept 29 to Oct 2, 2026). Complaints: transcripts that "jumble different speakers together", unexpected charges, trouble cancelling, and support that felt like "repeatedly being redirected through standard responses".
Reports "speaker misattributions, language mix-ups, and transcription errors are still common" and inconsistent speaker naming across identical call setups; likes live transcription and real-time AI chat. Its "3 supported languages" claim is out of date: Otter's own pricing page lists six.
One commenter said Otter, by default, emails every contact on a call inviting them to view the notes, calling it spam. This is one person's account that I could not verify; Otter's pricing matrix does list a "recording disclaimer email" admin control. HN also has a July 2026 thread titled "I'm Begging You to Leave Your AI Note-Taker at Home".
Free, Pro, Business, Enterprise. The price is the easy part; the caps are the plan.
Basic
300 transcription minutes a month; 30 minutes per conversation.
- 3 lifetime file imports
- AI Chat: 20 queries a month (3 per conversation)
- 25 most recent conversations kept
- 1 concurrent meeting
Pro
$16.99 billed monthly. 1,200 minutes; 90 minutes per meeting.
- 10 file imports a month
- AI Chat: 50 queries a month
- Unlimited history, advanced search and export
- Zapier, Salesforce, HubSpot integrations
Business
$30 billed monthly. Unlimited meetings; 4 hours per meeting.
- AI Chat: 200 queries a month
- 3 concurrent meetings
- Activity logs, usage analytics
- Prioritised support
Enterprise
Sales-quoted.
- SSO and SCIM (100-user minimum)
- Otter API and webhooks
- HIPAA compliance as an add-on
- Sales Notetaker, custom integrations
| Plan | Annual billing | Monthly billing | Transcription cap | Per-conversation | AI Chat queries / mo |
|---|---|---|---|---|---|
| Basic | $0 | $0 | 300 min / mo | 30 min | 20 (3 per conversation) |
| Pro | $8.33 / user / mo | $16.99 / user / mo | 1,200 min / mo (in-app); imports 10 / mo | 90 min | 50 |
| Business | $19.99 / user / mo | $30 / user / mo | Unlimited meetings; imports 6,000 min / user / mo | 4 hours | 200 |
| Enterprise | Custom | Custom | Unlimited | 4 hours | 200 |
Arithmetic you can check: a 10-person team on Business annual is 10 x $19.99 = $199.90 a month; on monthly billing it is $300. Pro's 1,200 minutes is 20 hours of meetings a month, and a meeting longer than 90 minutes is capped on that plan.
Pick by meeting length, volume and how much you need AI chat.
Start on Basic: 300 minutes is five hours of meetings. Check speaker labels on your own audio.
Pro at 1,200 minutes, as long as no single meeting goes past 90 minutes.
Business (4 hours per conversation) or look at Fireflies, which lists 2 to 4 hour recording limits by plan.
Use Otter's MCP server; mind the AI-chat query caps (50 on Pro).
Read the Security section first; announce the bot and get consent.
Enterprise, with a 100-user minimum for SSO and SCIM.
Capture is the solved part. Asking questions of your archive is the product, and it is metered.
The core is the Meeting Agent, Otter's evolved notetaker. It joins Zoom, Microsoft Teams or Google Meet, writes a transcript in real time, labels speakers, and produces a summary, outline and action items. Otter's March 2025 blog describes it as voice-activated: you can speak to it during the meeting and ask questions based on your meeting history. Beyond meetings, the Otter app records in-person conversations with live transcription, and since April there is a Mac and Windows desktop app.
What makes the 2026 version different is AI Chat across meetings. Instead of opening one transcript at a time you ask, "what did the client say about timeline across the last four calls?" and get an answer from the whole archive. The April launch added AI Chat Connectors, which let Otter pull from other apps, and an expanded MCP server so other assistants can query Otter. The pricing page also lists "AI Chat with Gmail, Google Drive, Notion, Jira" as coming to Enterprise.
| Capability | Basic | Pro | Business | Enterprise |
|---|---|---|---|---|
| Transcription minutes / month | 300 | 1,200 | Unlimited | Unlimited |
| Max per conversation | 30 min | 90 min | 4 hours | 4 hours |
| Concurrent meetings | 1 | 2 | 3 | 3 |
| File imports | 3 lifetime | 10 / month | Unlimited (6,000 min / user / mo) | Unlimited |
| Conversation history | 25 most recent | Unlimited | Unlimited | Unlimited |
| Export formats | mp3, txt | mp3, txt, pdf, docx, srt | + bulk export | + bulk export |
| Custom vocabulary | 5 terms | Team 100 names + 100 terms; user 100 + 100 | Team 800 + 800; user 200 + 200 | Team 800 + 800; user 200 + 200 |
| Transcription languages | English, Spanish, French, German, Japanese, Chinese | Same | Same | Same |
- A real Free plan with live transcription, speaker identification and AI Chat.
- Six transcription languages on the official page, which beats the "three" some rival reviews still repeat.
- Slide capture, real-time highlights and comments, and a Takeaways panel that turns annotations into action items.
- Accuracy on accents, crosstalk and phone-quality audio: independent reviewers describe speaker mix-ups.
- A 90-minute Pro cap means a two-hour session is cut off.
- AI Chat is metered by query: 3 per conversation on Free.
Invite it, review what it wrote, then let the notes flow to the tools you already use.
The human-review step is the one people skip. Otter lets you edit text and speaker tags, add custom vocabulary, and annotate in real time, and the paid plans add advanced search and export. Treat the first transcript of any new meeting type as a draft. If you are going to forward a summary, read it once: independent reviews report summaries that attribute a statement to the wrong person.
Otter's own introduction to the agent that joins meetings and answers questions. It is the primary-source demo for the Core Feature and Workflow sections above.
Published by the official "Otter_ai" YouTube channel (youtube.com/@Otterai) (channel UCc-hdvsHPLPL3YzM3a9nY5Q). Vendor demo, so it shows the best case.
Otter sells role-specific agents on top of the notetaker. Treat the SDR Agent as a different product.
Beyond the notetaker, Otter has a family of "use case agents". The sales, education and media ones are mostly the same engine with different templates and integrations. The SDR Agent is genuinely different: it is a customer-facing agent that talks to your website visitors on live video. That puts an AI in front of prospects, which is a brand and consent decision, not a note-taking one. There is no public price for it on the pricing page; it is "Schedule a Demo".
| Agent / surface | What Otter says it does | Where it shows up |
|---|---|---|
| Sales | Automatic follow-ups, notes capture and CRM sync | otter.ai Sales page; Enterprise lists an Otter Sales Notetaker |
| SDR Agent (new) | Engages website visitors with real-time, personalised video calls and tailored demos; qualifies leads and books meetings | Pricing page, marked NEW; contact sales |
| Recruiting Agent | Candidate insights, automatic follow-ups, notes sync with Greenhouse | Homepage and pricing page |
| Education | Lecture transcripts, note organisation, insight extraction, summaries | Education page; 20% .edu discount on Pro |
| Media | Interview transcription, thought organisation, content creation | Media page |
- Sales teams that already live in Salesforce or HubSpot and want notes synced without retyping.
- Recruiting teams on Greenhouse.
- Salesforce and HubSpot syncs are "subject to user limits" on the pricing page (1, 5 or unlimited users by tier).
- Agents that talk to customers need legal and brand sign-off before you launch.
Otter wants to be the meeting memory your other AI tools can read.
Otter's homepage lists 15+ integrations including Zoom, Microsoft Teams, Google Meet, Google Calendar, Google Docs, Slack, Salesforce, HubSpot, Jira, Notion, Asana, Dropbox and Glean, and shows Claude and ChatGPT as MCP clients. In April Otter described two directions: its MCP server (other assistants query Otter) and AI Chat Connectors, an MCP client (Otter queries other apps). The pricing page lists the Otter MCP server on the Basic plan and Zapier on Pro and above.
| Integration | What it does | Plan / limit |
|---|---|---|
| Otter MCP server | ChatGPT, Claude and other MCP clients query your transcripts and summaries | Listed on the Basic plan page; check current availability in your workspace |
| AI Chat Connectors (MCP client) | Otter AI Chat reaches other connected apps | Launched Apr 28, 2026; Gmail, Drive, Notion, Jira connectors listed as Enterprise |
| Zapier | Trigger workflows from meeting notes | Pro and above |
| Salesforce, HubSpot | Sync notes to CRM records | Subject to user limits |
| Otter API and webhooks | Custom integrations | Enterprise |
| Zoom, Teams, Google Meet | Auto-join and live notes | All plans, with concurrency caps (1 / 2 / 3) |
If you want to understand what an MCP connector is before turning this on, JAVIS's Claude MCP and Connectors guide walks through the idea, and the Zapier review covers the automation layer Otter plugs into.
A short official walkthrough of connecting an AI assistant to Otter through MCP, for the section above.
Published by the official "Otter_ai" YouTube channel.
Think in minutes per meeting, meeting length and AI questions per month.
| What you are doing | What you pay / hit |
|---|---|
| Trying it on a couple of meetings | $0 on Basic: 300 minutes, 30 per conversation |
| Two 45-minute calls a day, 20 working days | 2 x 45 x 20 = 1,800 minutes a month: over Pro's 1,200 (JAVIS arithmetic), so Business or fewer recorded calls |
| One 2-hour workshop on Pro | Capped at 90 minutes per conversation; Business allows 4 hours |
| Asking AI Chat 10 questions a week | About 43 a month: fits Pro's 50, not Basic's 20 (JAVIS arithmetic) |
| 10-person sales team on Business, annual | 10 x $19.99 = $199.90 a month; $300 on monthly billing |
| Need API, webhooks, SSO or SCIM | Enterprise, quoted; SSO and SCIM need a 100-user licence minimum |
| Pay-as-you-go overage | Not listed on the pricing page: when a cap is hit, you wait, upgrade or stop recording |
Solid enterprise controls on paper, and a consent question a court is now looking at.
| Control | What Otter says |
|---|---|
| Compliance | SOC 2 Type 2 report; HIPAA available as an Enterprise add-on; GDPR and CCPA practices (otter.ai trust page) |
| Encryption | AES-256 for stored data and real-time content; TLS in transit |
| Model training | Otter says it does not use user data for training in raw form and uses an automatic process to de-identify data before training, with no manual human review of that step |
| Access | Two-factor authentication; sharing controls; export restrictions and meeting access controls shown in its security-settings panel |
| Admin | Activity logs and usage analytics (Business); SSO, SCIM, domain capture, custom retention (Enterprise) |
| Consent tools | Pricing matrix lists a recording disclaimer email, Teams recording-permission capture, and custom notetaker name and image |
| Fact | Detail | Status as I could verify |
|---|---|---|
| Case | In re Otter.AI Privacy Litigation, No. 5:25-cv-06911, N.D. Cal., Judge Eumi K. Lee | Four suits filed Aug to Sept 2025, consolidated Oct 22, 2025 |
| What is alleged | Otter's notetaker joins meetings and records people who never agreed, and Otter keeps data (including voiceprints) for its own purposes | Allegations only; Otter denies wrongdoing |
| Aug 13, 2026 ruling | Survived: federal Wiretap Act, CIPA section 631, both Illinois BIPA voiceprint claims, unjust enrichment, unfair competition. Dismissed with leave to amend: both CFAA counts, CDAFA, Washington Privacy Act, most intrusion claims | Pleading-stage ruling, not a finding of liability |
| Otter's argument | It is "merely the meeting host's tool"; the court said plaintiffs plausibly allege Otter independently collects and uses communications | Case proceeds toward discovery; no trial date; no certified class |
I am not a lawyer and this is not legal advice. The practical read: the lawsuit is about the product design of a bot that records everyone on a call when one person turns it on, which is true of every meeting notetaker, including Fireflies. Whatever you use, tell people, get consent where your jurisdiction requires it, and keep the bot off sensitive conversations such as medical, legal and HR calls.
Otter's own security overview for enterprise buyers. It is a vendor pitch, so read it alongside the legal table above.
Published by the official "Otter_ai" YouTube channel.
Best for individuals and small teams who live in Zoom, Teams or Meet and want searchable notes without effort.
| Situation | Why Otter fits | Likely plan |
|---|---|---|
| Solo professional with a few meetings a day | Live transcription, summaries, action items, a real Free tier | Basic, then Pro |
| Students and teachers | Lecture transcripts; 20% .edu discount on Pro | Pro with discount |
| Sales and recruiting teams | CRM and Greenhouse sync, agents, shared notes | Business, then Enterprise |
| Anyone who wants ChatGPT or Claude to read meeting history | MCP server and AI Chat across meetings | Pro or Business (query caps) |
| Long workshops, interviews and legal or medical calls | Not its best fit: 90-minute Pro cap, consent and privacy risk | Business with consent controls, or a different tool |
| Multilingual teams beyond six languages | Look elsewhere: Fireflies lists 100+ languages on its pricing page | Compare Fireflies |
Five real gaps, not manufactured ones.
The federal case over how its notetaker records and stores other people's voices is moving toward discovery. Allegations, not findings, but the product design question is real.
300 minutes and 30 per conversation on Free; 1,200 and 90 on Pro; AI Chat is 20, 50 and 200 queries a month. No overage option is listed.
Independent testers and Trustpilot reviewers report mixed-up speakers and errors on accents and crosstalk. The vendor has not published an accuracy benchmark that I could verify.
Trustpilot has a recurring pattern of unexpected charges and cancellation trouble, set against praise for responsive support.
Fifth gap, smaller: the best new features are unevenly available. Gmail, Drive, Notion and Jira chat connectors are listed for Enterprise, and video replay and the Sales Notetaker are Enterprise too.
Otter is the broad, familiar option. Others win on price, languages or specialty.
| If you mostly need... | Compare Otter with... |
|---|---|
| Cheaper paid seats, 100+ languages, CRM-heavy sales workflows | Fireflies.ai (Pro $10 annual, Business $19 annual on its pricing page, Oct 3, 2026) |
| Editing the recording itself as text, then publishing | Descript (a different category: an editor, not a meeting notetaker) |
| A general assistant that can take notes and also write, search and reason | ChatGPT review, Claude review |
| Piping meeting notes into other apps automatically | Zapier, n8n |
| Speech to text you control yourself, via API | ElevenLabs review (speech models and voice tools) |
| A free tier from an independent notetaker, or notetakers built into your meeting app | Fathom (now owned by Superhuman, per TechCrunch, Sept 14, 2026), Zoom, Teams and Google Meet built-ins: JAVIS has not researched these for this review |
| Entry point (verified Oct 3, 2026) | Otter | Fireflies.ai | Descript (editor) |
|---|---|---|---|
| Free plan | 300 min / mo, 30 min per conversation | Unlimited transcription (asterisked), limited AI summaries, 20 AI credits | 60 media minutes / mo, 100 AI credits one-time |
| Cheapest paid, annual billing | $8.33 / user / mo (Pro) | $10 / seat / mo (Pro) | $16 / mo (Hobbyist) |
| Cheapest paid, monthly billing | $16.99 / user / mo | $18 / seat / mo | $24 / mo |
| Unit that runs out | Transcription minutes, AI Chat queries | AI credits, storage minutes | Media minutes, AI credits |
No affiliate relationship shapes this review.
JAVIS has not established a publisher affiliate program with Otter.ai. Every CTA in this article points to the official product. If that changes, the review's facts, limitations and alternatives will not.
Go to the official page.
Open Otter.aiRead the alternative that matches your real question.
Choose the next articleOtter sits between the notetaker and the AI-assistant memory layer.
Every card below links to a published JAVIS review.
Card images are official assets from each product's own site or the target JAVIS page.
Questions people are actually searching right now
How much does Otter.ai cost in October 2026?
Basic is free with 300 transcription minutes a month. Pro is $8.33 per user per month billed annually or $16.99 monthly. Business is $19.99 annually or $30 monthly. Enterprise is custom. Verified on otter.ai/pricing, Oct 3, 2026.
Is Otter.ai free?
Yes. Basic gives 300 minutes a month, 30 minutes per conversation, 3 lifetime file imports and 20 AI Chat queries a month, with live transcription and summaries.
What are the Otter.ai limits that matter?
Per-conversation caps (30, 90, then 240 minutes), monthly minutes on Basic and Pro, and AI Chat queries (20, 50, 200 a month). Minutes do not roll over.
Does Otter work with Claude and ChatGPT?
Otter lists an MCP server that Claude, ChatGPT and other clients can use to query your meeting knowledge, and an MCP client in AI Chat for connecting other apps. Query caps still apply.
Is there a lawsuit against Otter.ai?
Yes. In re Otter.AI Privacy Litigation is a consolidated federal class action in N.D. California. On Aug 13, 2026 the judge let the wiretap, CIPA and Illinois BIPA claims proceed. These are allegations; Otter denies wrongdoing and no class has been certified.
Which languages does Otter transcribe?
Otter's pricing page lists English, Spanish, French, German, Japanese and Chinese. You choose the language per recording; it transcribes but does not translate.
Is Otter safe for company data?
Otter states SOC 2 Type 2, AES-256 encryption, 2FA and, on Enterprise, SSO, SCIM and a HIPAA add-on. Suitability depends on your consent and retention rules; read the Security section.
What are the best Otter alternatives?
Fireflies.ai for price and languages, Descript if you need to edit the recording, and assistants like ChatGPT or Claude. See Alternatives above.
Where this came from, and when it was checked.
Pricing and plans: otter.ai/pricing, checked Oct 3, 2026, including the feature matrix and FAQ.
Product and launches: otter.ai homepage, otter.ai/blog (April 28, 2026 "Conversational Knowledge Engine" announcement; Meeting Agent and AI Chat demo posts), otter.ai/privacy-security.
Market: Otter's own press release for 35M users, 1B+ meetings; March 2025 press release for $100M ARR; funding totals from Sacra and press summaries (third-party). Fathom acquisition: TechCrunch, Sept 14, 2026.
Legal: RecordingLaw.com, UC Today, Jackson Lewis and Sheppard Mullin write-ups of In re Otter.AI Privacy Litigation, No. 5:25-cv-06911 (N.D. Cal.). I did not read the docket.
Real users: Trustpilot (4.1/5, 711 reviews) read directly; tl;dv review (competitor, labelled); Hacker News comment. G2 (~4.3 to 4.4) and Capterra (~4.4) are search-aggregated because they could not be read directly. Reddit could not be read.
Official videos: youtube.com/watch?v=va9kUilGiiQ, AOJuW-NhL0E and 6nmz-FJ7dtM; author "Otter_ai" on YouTube.
JAVIS Fit Score (74/100): editorial composite of capture quality and breadth (strong), pricing transparency and free tier (good), AI chat and MCP capability (good), plan-cap friction, user sentiment on accuracy and billing, and consent and legal risk (weak). Not a benchmark; no automated test was run.
