Otter.ai vs Everpaper: 2026 Private Mac Guide

If you're comparing Otter.ai vs Everpaper on an Apple Silicon Mac, the choice comes down to what happens to your meetings after you press Record. Otter sends audio to its cloud and excels at shared transcripts, CRM integrations, and cross-platform team workflows. Everpaper records and transcribes locally, without adding a bot to the call, then connects the meeting to a private memory of your whole workday.

Full disclosure before anything else: Everpaper is our product. We'll be fair about where Otter genuinely wins, because it does win in real places. But you should know who wrote this.

Quick verdict

On macOS 14.2+ with Apple Silicon, Everpaper is the stronger choice when privacy, no-bot capture, offline operation, unmetered local transcription, and personal recall matter most. Otter remains stronger for collaborative teams that need CRM integrations or mobile, web, and Windows access.

Choose Otter.ai if…

• You need a web app and apps on Windows, iOS, and Android, not just a Mac.
• Your team edits transcripts together live, with comments and highlights, in one shared workspace.
• You want Salesforce, HubSpot, and Zapier integrations wired into your meeting pipeline.
• Cloud storage of your meetings is acceptable to you and your legal team.

Choose Everpaper if…

• You don't want a bot visibly joining your calls or recording others without clear consent.
• You want capture, transcription, and search to run locally with no account and no mandatory network traffic.
• You're tired of counting transcription minutes and 30-minute conversation caps.
• You want meetings to become todos, sourced answers, and context in a private memory of your whole workday.
• You want to import your exported Otter VTT transcript history.

At a glance

Why Everpaper is the stronger choice on a supported Mac

If you use macOS 14.2 or later on Apple Silicon and do not need a shared transcript workspace, Everpaper differs from a cloud meeting bot in several practical ways:

• Your meetings can remain on your Mac. No bot joins the call, and capture, transcription, search, and export work offline. Sensitive conversations can stay on the Mac instead of becoming a server-side meeting archive.
• A full day is not metered. With no transcription-minute quota or per-conversation limit, long workshops and back-to-back calls do not require rationing minutes or splitting recordings.
• The transcript connects to whole-workday context. Extract todos pulls out action items; the Day Journal preserves the work surrounding the meeting; Second Brain and recall connect recurring people, files, projects, and topics; and Ask the Book answers with sources pointing to exact frames.
• Your exported Otter VTT transcript history can come with you. Import Transcript turns Otter VTT exports into real meetings in the daybook without requiring the original audio. It imports transcript content, not Otter audio or screenshots.

That combination makes Everpaper the stronger personal meeting-memory system on a supported Mac. Otter is still the better fit when shared editing, CRM automation, centralized administration, or broad device access is the priority.

Join the Everpaper private-beta waitlist if that local-first workflow fits your work.

How we compared

Every Otter claim here comes from Otter's own pricing page, privacy policy (effective June 16, 2026), help center, and public court reporting on the class action — all checked 2026-08-04. Every Everpaper claim comes from the shipped meeting docs and privacy docs. Claims we couldn't verify were omitted.

Head-to-head

Capture: bot in the call vs no bot at all

Verdict: Everpaper for discretion, Otter for hands-off auto-join.

Otter's main capture path is OtterPilot (also called Otter Notetaker): a bot that auto-joins your Zoom, Google Meet, or Teams call as a visible participant, records the audio, captures slides, and emails a summary afterward. Everyone in the meeting sees "OtterPilot" in the participant list. That visibility is honest in one sense — but the consent model has burned people. A 2022 Hacker News report describes the bot joining two confidential meetings on a user's behalf, recording everything, and emailing all attendees an inaccurate outline. Reddit has threads with titles like "Do not join Otter.Ai unless you want your whole company spammed."

Everpaper never joins anything. When a meeting starts, you press ⇧⌘M (or use Calendar Auto-Start with a 5-second cancellable countdown) and your Mac records its own microphone and system audio as separate time-aligned tracks. No participant appears and no email goes out. That makes disclosure the recorder's responsibility: get the consent your participants and jurisdiction require before recording.

Privacy: cloud training vs nothing leaves the Mac

Verdict: Everpaper, decisively. This is the core of the comparison.

Otter's privacy policy (effective June 16, 2026) states that audio, transcripts, and OtterPilot screenshots go to AWS in the US, and that Otter may use your data for "training our proprietary AI technology on de-identified audio recordings and on transcriptions (which may contain Personal Information)." It also discloses sharing with data-labeling and AI service providers. Retention lasts "as long as necessary to fulfill the purposes set out in this Policy." There is no on-device processing mode; every AI feature runs in Otter's cloud and nothing works offline.

This is also now a legal question, not just a policy one. In re Otter.AI Privacy Litigation — a consolidated federal class action in the Northern District of California, combining four suits filed August–September 2025 — alleges recording without all-party consent and using recordings to train AI, under ECPA, CIPA, and Illinois's BIPA. The motion-to-dismiss hearing was held May 20, 2026 before Judge Eumi K. Lee. Otter claimed 35M+ users as of December 2025. The case is unresolved, but if your meetings contain client data, health information, or anything privileged, "we'll see how the litigation goes" is a real line item.

Everpaper's model is structurally different: capture, OCR, transcription (a speech model on the Apple Neural Engine), storage, the knowledge graph, and search all run locally with no account at all. The docs put it as "no provider, no traffic" — AI features like Ask the Book and meeting summaries simply wait until you configure a provider, and you can point them at a fully local model (Ollama, LM Studio, oMLX with Qwen3.5 4B, or Apple Foundation Models) so nothing leaves the machine even then. Cloud providers (Anthropic, OpenAI, Gemini, OpenRouter) exist but sit behind a toggle — Allow cloud AI to analyze your journal — that is off by default. On top of that, an offline redaction pass scrubs API keys, tokens, card numbers, and SSNs at capture time and again before any outbound AI call.

Limits: metered minutes vs no caps

Verdict: Everpaper. Otter's free tier is a demo with anxiety built in.

Otter's free Basic plan gives you 300 transcription minutes a month, a hard 30-minute cap per conversation, and 3 lifetime audio/video file imports. Deleted conversations still count toward the monthly limit. Pro raises that to 1,200 minutes a month and 90 minutes per meeting; Business goes to 4 hours per meeting. A Capterra reviewer put the lived experience plainly: "Having to track usage and stay within a set number of transcription minutes added unnecessary stress." If you bill by the hour and your day is back-to-back calls, 300 minutes is about two and a half workdays.

Everpaper has no minute quotas. Record a six-hour workshop, transcribe a folder of old audio with everpaper-cli transcribe, keep screenshots for 30 days (or 7, 90, or Forever — your call). The only ceiling is your disk.

Transcription quality and speaker ID

Verdict: Tie, with different failure modes.

Otter has a decade of ASR work behind it and identifies speakers on every tier, with taggable speakers from Pro up. But Capterra and G2 reviews regularly call accuracy "hit or miss," and the generic "Speaker 1 / Speaker 2" labels often need manual correction.

Everpaper transcribes on-device with Parakeet (parakeet-tdt-0.6b-v3) by default, or optionally MOSS (MOSS-Transcribe-Diarize, ~2 GB download, happiest on 16 GB machines), which does transcription and diarization in one model. Live transcripts show speaker labels as you talk, and ⌘4 hides your own voice or any speaker from the record. Honest edges from the docs: word-level timestamps within a sentence are approximate, and a voice silent for a long stretch may re-enter under a new label. Neither tool is magic in a noisy room.

Collaboration and team features

Verdict: Otter, clearly — this is its home turf.

Real-time co-editing of transcripts, highlights, comments, shared workspaces, cross-meeting AI Chat over your whole archive, admin analytics, activity logs, Salesforce/HubSpot/Zapier integrations, and an Enterprise API make Otter a team product. Everpaper's supplied docs describe a personal daybook and do not list shared workspaces, transcript comments, or CRM sync. What Everpaper does have is the Second Brain — a cross-day knowledge graph of people, files, projects, and topics from your days and your meetings.

Platform coverage

Verdict: Otter, and it's not close.

Otter runs on the web, macOS, Windows, iOS, Android, and as a Chrome extension. Everpaper runs on macOS 14.2 or later, Apple Silicon only — no Intel Macs, no iPhone app, no Windows, no web. If your team is mixed-platform or you review notes from your phone, Otter is the only one of the two that exists for you.

Beyond meetings

Verdict: Everpaper for whole-workday context; Otter for meeting-centered workflows.

Otter keeps its post-call tools centered on meeting content. Everpaper extends the transcript into whole-workday context: Extract todos pulls action items from the transcript, while the Day Journal places the conversation beside the rest of your work through periodic screenshots of your front window (default every 2 minutes), on-device OCR, and an AI day narrative.

That whole-day context feeds Second Brain and recall across people, files, projects, and topics. ⌘K opens the Index, ⇧⌘A asks the Book with sources pointing back to exact frames, and ⌥Space summons the Concordance search panel over any app. You can export the whole journal as an Obsidian vault or a GraphML knowledge graph. If the goal is remembering your work, meetings are one input, not the whole product.

Use cases

Otter fits: sales and CS teams who need calls synced into Salesforce, orgs standardized on Windows or mixed devices, journalists and researchers who collaborate on transcripts with colleagues, anyone whose company has already accepted cloud meeting tools.

Everpaper fits: solo founders, consultants, developers, and lawyers on Apple Silicon Macs who handle sensitive conversations; people in all-party-consent jurisdictions who want recording to be a deliberate act; anyone who hit the 300-minute wall; anyone who wants their meeting notes wired into a private memory of their whole day.

Honest pros and cons

Otter.ai — pros: broad platform coverage including web and mobile; genuine real-time collaboration on transcripts; deep business integrations; speaker ID on every tier; a real free tier for light users; cross-meeting AI Chat over a big archive.

Otter.ai — cons: privacy policy reserves training on your audio and transcripts; consolidated BIPA/ECPA/CIPA class action pending; bot auto-join behavior has a documented history of consent failures; 300 min/mo and 30 min/conversation on free; deleted conversations still count against your quota; no offline mode at all.

Everpaper — pros: nothing leaves the Mac by default; no minute caps or per-conversation limits; no bot enters your calls; redaction scrubs secrets before any AI call; local-provider path means AI features can run entirely offline; meetings feed a full day journal and second brain.

Everpaper — cons: macOS 14.2+ and Apple Silicon only; no mobile, no web, no Windows; still in private beta; AI features (summaries, Ask, day narrative) wait until you configure a provider; Second Brain has a readiness gate (5 days, 300 frames, 60 distinct things, 3 recurring threads) before it shows insights; single-user only — no team workspaces.

Pricing

Otter.ai (per user, checked 2026-08-04, from otter.ai/pricing):

On monthly billing, Pro costs $203.88 per user per year and Business $360. For a 10-person team on Business annual, that's $2,880 a year. Gotchas: the free tier's 30-minute conversation cap kills real meetings, file imports are capped (3 lifetime on free, 10/mo on Pro), and Business's "unlimited file imports" carries an asterisk on the pricing page.

Everpaper: in private beta with limited spots, with a founding-license option for early supporters — not a monthly subscription. No public price, and we won't invent one here.

FAQ

Does Otter train on your meetings? Yes, per its own privacy policy (effective June 16, 2026): Otter may use your data for "training our proprietary AI technology on de-identified audio recordings and on transcriptions (which may contain Personal Information)." The policy also discloses sharing with data-labeling and AI service providers. "De-identified" is doing a lot of work in that sentence — the policy itself notes transcripts may contain personal information.

Is Otter free? There is a permanent free tier: 300 transcription minutes per month, a 30-minute cap per conversation, and 3 lifetime file imports (checked 2026-08-04). In August 2022 Otter cut the free allowance from 600 to 300 minutes. Deleted conversations still count toward the monthly limit, so "free" in practice means light, careful use.

What is the Otter.ai lawsuit about? In re Otter.AI Privacy Litigation is a consolidated federal class action (N.D. Cal.) combining four suits filed August–September 2025. It alleges Otter recorded meetings without all-party consent and used recordings to train AI, under the federal ECPA, California's CIPA, and Illinois's BIPA. The motion-to-dismiss hearing was held May 20, 2026; the case is ongoing.

Does Everpaper work on Windows or iPhone? No. macOS 14.2 or later, Apple Silicon only. There is no mobile app, no web app, and no Windows version.

Can Everpaper summarize meetings without the cloud? Yes. Transcription always runs on-device. Summaries and Ask need an AI provider, but you can use a local one — Ollama, LM Studio, oMLX (which offers one-click setup with Qwen3.5 4B), or Apple Foundation Models — so the whole pipeline stays on your Mac.

Can I move from Otter to Everpaper? Everpaper 1.11.0 added Import Transcript: it accepts .vtt exports from Zoom, Teams, and Otter and turns them into real meetings in your daybook without requiring the original audio. This brings over exported VTT transcript history, not Otter audio or screenshots.

Final recommendation

Choose Otter if your work depends on shared transcripts, CRM sync, centralized team workflows, or access from phones, the web, and Windows. It is the more mature collaboration product, and you should adopt it with open eyes about its privacy policy, training clause, cloud-only operation, and pending class action.

For an individual using macOS 14.2+ on Apple Silicon, Everpaper is our decisive recommendation when privacy, no-bot capture, offline operation, unmetered local transcription, and whole-day memory matter more than team collaboration. It keeps core capture and recall on your Mac, turns meetings into todos and sourced answers, and imports exported Otter VTT transcript history.

Everpaper is in private beta — join the waitlist at everpaper.app. For the full meeting-recording feature set, see The Minutes in the docs, and for a wider field of options, read our best Otter.ai alternatives for 2026.