---
title: "Best AI Meeting Notetakers for Accurate Transcripts, Summaries, and Action Items"
type: "Ranking"
url: "https://aidemos.com/best/ai-meeting-notetakers"
description: "If you need a bot to join live calls and turn them into usable meeting memory, this ranking compares eight notetakers on the same multi-speaker Google Meet standup, checking transcript accuracy, speaker attribution, summaries, action items, search, sharing, API/MCP access, and pricing."
readTime: "14 min read"
tested: "Fellow vs Fathom vs MeetGeek vs Notta vs HappyScribe vs Otter.ai vs Granola vs Fireflies.ai"
testedDate: "2026-07-31"
category: "productivity"
published: "2026-08-12T03:45:45.634048+00:00"
updated: "2026-08-12T08:05:24.254133+00:00"
evidenceCount: 101
verifiedCount: 94
coverage: "dense"
---

# Best AI Meeting Notetakers for Accurate Transcripts, Summaries, and Action Items

`8 Tools Tested` · `Same Meeting` · `Live Call` · `Transcript + Summary + Action Items` · `Search, Chat, API & MCP`

**Tested:** Fellow vs Fathom vs MeetGeek vs Notta vs HappyScribe vs Otter.ai vs Granola vs Fireflies.ai · 2026-07-31

> If you need a bot to join live calls and turn them into usable meeting memory, this ranking compares eight notetakers on the same multi-speaker Google Meet standup, checking transcript accuracy, speaker attribution, summaries, action items, search, sharing, API/MCP access, and pricing.

## Our Verdict

**#1 pick: Fellow** (Best) — Excellent speaker attribution and open sharing, but the free tier is too restrictive and API/MCP sit behind paid plans.

- #2 Fathom — Strong at capture, search, and sharing; weaker on speaker attribution and MCP person grounding.
- #3 MeetGeek — Strong on capture, notes, search, API, and MCP; weaker on speaker attribution, chat reliability, and open sharing.
- #4 Notta — Strong at capture, summaries, and action items; weaker on speaker attribution, search jumping, and automation access.
- #5 HappyScribe — Strong on capture, chat, API, and sharing; weaker on speaker attribution, owner accuracy, and direct search.
- #6 Otter.ai — Strong on capture, summaries, chat, and collaboration; weak on speaker accountability.
- #7 Granola — Best at grounded downstream workflows, but weakest on raw transcript fidelity.
- #8 Fireflies.ai — Very strong at live meeting capture, transcripts, notes, search, and AskFred — with weak editing, sharing, and API access.

## How We Tested

We tested eight AI meeting notetakers in parallel on the same real 31 July 2026 Google Meet standup (~25 minutes, 14 attendees, ~10 active speakers, normal audio). Each tool captured the identical live conversation, and we compared their transcripts, summaries, action-item extraction, search/chat behavior, sharing behavior, API or MCP access, pricing limits, integrations, language support, and privacy posture using the same source meeting so the tool itself was the variable.

**What we evaluated:**

| Criterion | Description |
| --- | --- |
| Transcription Accuracy | Word accuracy on the shared call, especially names, tools, numbers, and jargon. |
| Speaker Diarization | Correctly attributes who said what across a multi-speaker standup. |
| Join Method & Reliability | Bot-in-call vs. botless/local capture, and reliability of joining Zoom, Meet, or Teams. |
| Summary Quality | Captures decisions and key points, is structured and skimmable, and drops nothing important. |
| Action-Item Extraction | Finds the real action items, ideally with owner and due date, with low false positives. |
| Topic Segmentation | Breaks long multi-topic meetings into useful sections instead of one blob. |
| Editability | Can you fix a wrong summary or action item before sharing? |
| Search Across Notes | Finds a moment or decision by keyword or question. |
| Chat with Notes / Ask Questions | Gives grounded answers with the cited moment and admits when unknown. |
| Sharing Without Registration | Can teammates open the notes or recording via link without creating an account, or is signup forced? |
| API Availability | Whether there is a real API to pull transcripts, summaries, and action items programmatically. |
| MCP Support | Whether it exposes an MCP server so agents like Claude or ChatGPT can query meeting data. |

## The Ranking

8 tools tested head-to-head on the same input.

### 1. Fellow — Best

*Strongest at clean capture and downstream retrieval, with only a minor action-item miss and partial editability limits.*

Excellent speaker attribution and open sharing, but the free tier is too restrictive and API/MCP sit behind paid plans.

### 2. Fathom — Usable

*Strong at capture, search, and sharing; weaker on speaker attribution and MCP person grounding.*

Very capable across capture, notes, sharing, API, and search, with only modest speaker-level caveats.

### 3. MeetGeek — Usable

*Strong on capture, notes, search, API, and MCP; weaker on speaker attribution, chat reliability, and open sharing.*

Good summaries, action items, API, and MCP, but login-gated sharing and mixed diarization keep it behind the leaders.

### 4. Notta — Usable

*Strong at capture, summaries, and action items; weaker on speaker attribution, search jumping, and automation access.*

Accurate capture and strong editability, but direct search is limited and no public API was available in this cycle.

### 5. HappyScribe — Usable

*Strong on capture, chat, API, and sharing; weaker on speaker attribution, owner accuracy, and direct search.*

Structured notes and editing work well, but summary/name hallucinations and no direct transcript search reduce confidence.

### 6. Otter.ai — Usable

*Strong on capture, summaries, chat, and collaboration; weak on speaker accountability.*

The transcript and chat are strong, but diarization and action-item ownership failures make it unreliable for standup accountability.

### 7. Granola — Usable

*Best at grounded downstream workflows, but weakest on raw transcript fidelity.*

Privacy-friendly and feature-rich, but poor transcription and hidden speaker tagging hurt baseline usability.

### 8. Fireflies.ai — Usable

*Very strong at live meeting capture, transcripts, notes, search, and AskFred — with weak editing, sharing, and API access.*

Strongest overall balance of transcript quality, summaries, action items, search, chat, and MCP in this test.

## Full Breakdown

### Fellow

Strong diarization and open sharing, but the free tier is too limited and paid access is required for meaningful programmatic use.

![Fellow screenshot showing Shared meeting input](https://cdn.futuresmart.ai/public/aidemos/19f0fc64803b41d081cd239b60526b4e.png?v=1)
*Screenshot — Shared 31 July 2026 Google Meet standup used as the common test input across all eight tools, with a live recorded call and multiple notetaker bots present.*

**What worked:**
- Fellow's biggest strength is speaker attribution: it labeled all speakers cleanly, and the transcript, summary, search, chat, and sharing flow were all polished. It is a good fit when speaker accountability matters and you can live on a paid plan.

**Where it struggled:**
- The free tier is not viable for recurring team use, and API/MCP access is effectively paid-only. One action item was also misplaced in the evaluation, so task ownership still needs a quick human check.

**What came out:**

![Fellow output showing Transcript output](https://cdn.futuresmart.ai/public/aidemos/5fdfc1fa3cde4398a821f03a4468df70.png?v=1)
*Output — Fellow generated a readable meeting transcript with speaker-labeled dialogue and transcript controls.*

![Fellow output showing Speaker attribution output](https://cdn.futuresmart.ai/public/aidemos/680388795b554a4c86a1ca654881abfa.png?v=1)
*Output — The transcript shows named speaker attribution across turns, indicating excellent diarization in this cycle.*

![Fellow output showing Join confirmation](https://cdn.futuresmart.ai/public/aidemos/1ba3eed84fdb4162b7273a8bbd0c0533.png?v=1)
*Output — A Google Meet tile confirms the Fellow notetaker joined the call.*

![Fellow output showing Summary output](https://cdn.futuresmart.ai/public/aidemos/76b8e62129ad40b193c6d970e97439ac.png?v=1)
*Output — The AI Note tab shows the meeting summary section, indicating Fellow generated recap notes for the call.*

![Fellow output showing Action items output](https://cdn.futuresmart.ai/public/aidemos/86f2967680604cfa846e4327e0ace54d.png?v=1)
*Output — The suggested action-items panel shows extracted follow-ups, but the report notes one task was assigned to the wrong person in this cycle.*

![Fellow output showing Topic segmentation output](https://cdn.futuresmart.ai/public/aidemos/88141f90f736445096311aa17d591251.png?v=1)
*Output — The meeting is split into topic sections, showing good organization of the call content.*

![Fellow output showing Editability view](https://cdn.futuresmart.ai/public/aidemos/a2ee7d9d103e400bb14323817c92bc8a.png?v=1)
*Output — The Topics view shows an Edit control, confirming the summary/topic content can be adjusted before sharing.*

![Fellow output showing Search result](https://cdn.futuresmart.ai/public/aidemos/75389b02741846b38a6a5eb14548afb5.png?v=1)
*Output — Searching the transcript for API returns a highlighted match, confirming keyword search works inside the meeting.*

![Fellow output showing Chat answer](https://cdn.futuresmart.ai/public/aidemos/ef507c8bb4be426a9d057cb25b8d9fa7.png?v=1)
*Output — Fellow answered a meeting question about access status in chat, showing grounded retrieval over the meeting record.*

![Fellow output showing Sharing dialog](https://cdn.futuresmart.ai/public/aidemos/c0317ec3fd434c5e9faadf4e2d658014.png?v=1)
*Output — The share/send modal offers link sharing and delivery options without forcing recipients to sign in.*

![Fellow output showing API docs](https://cdn.futuresmart.ai/public/aidemos/6f7a9af3c1ff404ebde59d2531b25d58.png?v=1)
*Output — The developer API documentation explains how to use Fellow's API for meeting data, transcripts, and structured notes.*

![Fellow output showing MCP docs](https://cdn.futuresmart.ai/public/aidemos/a9a2789eaf364d9aa638866c447872ca.png?v=1)
*Output — The MCP documentation explains how to connect AI assistants to meeting notes, action items, transcripts, and calendar events.*

![Fellow output showing Pricing page](https://cdn.futuresmart.ai/public/aidemos/f27ef03033194062978a88cfe72552e5.png?v=1)
*Output — The pricing page shows Free, Team, Business, and Enterprise plans, including the very limited free tier and the paid upgrade path.*

![Fellow output showing Integrations page](https://cdn.futuresmart.ai/public/aidemos/be0a0f57dbd34c6683c686921db0dd85.png?v=1)
*Output — The integrations directory shows apps such as Android App, Asana, BambooHR, and Box.*

![Fellow output showing Language support](https://cdn.futuresmart.ai/public/aidemos/abd4371942884b2fb5a41de0e825ed82.png?v=1)
*Output — The supported-languages page states Fellow's AI Note Taker can transcribe 100 languages and notes speaker-identification caveats for some languages.*

![Fellow output showing Privacy policy](https://cdn.futuresmart.ai/public/aidemos/5d60149294dc49adaeb262d4e86e3451.png?v=1)
*Output — The privacy page summarizes Fellow's data-handling and security posture in plain language.*

### Fathom

Bot-based assistant with strong summaries, open sharing, working API access, and a valuable free tier.

![Fathom screenshot showing Shared meeting input](https://cdn.futuresmart.ai/public/aidemos/9feea2699803487a98514660f7075887.png?v=1)
*Screenshot — Shared 31 July 2026 Google Meet standup used as the common test input across all eight tools, with a live recorded call and multiple notetaker bots present.*

**What worked:**
- Fathom performed very well across the workflow: capture, notes, action items, search, chat, open sharing, API access, and integrations all came through cleanly. Its free-tier story is also much more generous than many competitors, which makes it one of the best value options here.

**Where it struggled:**
- Speaker attribution was not perfect in the broader report, and the MCP workflow had a grounding risk on the tested query. It is excellent, but not flawless on speaker-level accountability.

**What came out:**

![Fathom output showing Transcript output](https://cdn.futuresmart.ai/public/aidemos/dd7be0ce49cc41a1ba45aeba88e6715f.png?v=1)
*Output — Fathom generated a readable transcript thread for the meeting, showing timestamped dialogue in the transcript tab.*

![Fathom output showing Speaker attribution output](https://cdn.futuresmart.ai/public/aidemos/67da3742bbe8422aa0e13d16c796d867.png?v=1)
*Output — The transcript shows speaker-labeled turns, demonstrating working diarization on the tested meeting.*

![Fathom output showing Join confirmation](https://cdn.futuresmart.ai/public/aidemos/af9452a0c9424a32b2243f807f8c30e9.png?v=1)
*Output — A participant tile confirms the Fathom bot joined the meeting successfully.*

![Fathom output showing Summary output](https://cdn.futuresmart.ai/public/aidemos/0dafac33a3bd4d35a6bc7a13beef8cd5.png?v=1)
*Output — Fathom produced a structured summary with a meeting-purpose block and key takeaways section.*

![Fathom output showing Action items output](https://cdn.futuresmart.ai/public/aidemos/c16a4895225a4bf7b4607a549145ef0e.png?v=1)
*Output — The action-items section shows extracted tasks with owner attribution and a checkbox-style task row.*

![Fathom output showing Topic segmentation output](https://cdn.futuresmart.ai/public/aidemos/7635825fac144c0a95a84c044e375d18.png?v=1)
*Output — The summary is broken into a named topic section, confirming useful topic segmentation.*

![Fathom output showing Editability view](https://cdn.futuresmart.ai/public/aidemos/f017abe4416a40aea844f9d1fff9c1fe.png?v=1)
*Output — The action item row shows an edit affordance, confirming inline task editing is available.*

![Fathom output showing Editability view](https://cdn.futuresmart.ai/public/aidemos/510978e3e9204e2d95a377c1987f6e97.png?v=1)
*Output — The customization modal lets the user instruct the AI to regenerate the summary, showing the summary can be tuned before sharing.*

![Fathom output showing Search result](https://cdn.futuresmart.ai/public/aidemos/217ccd5a423a45789651873169f21404.png?v=1)
*Output — Transcript search found the term API and highlighted the match in context, confirming keyword search across the meeting record.*

![Fathom output showing Chat answer](https://cdn.futuresmart.ai/public/aidemos/d6e26a36815f46eb99ed03422a23d165.png?v=1)
*Output — Ask Fathom answered a meeting question with a concrete scheduled date and a supporting transcript mention.*

![Fathom output showing Sharing dialog](https://cdn.futuresmart.ai/public/aidemos/3dab16b7a16b4df490036fc5afc67e9d.png?v=1)
*Output — The share modal shows link-sharing controls and open access behavior, supporting no-login viewing for recipients.*

![Fathom output showing API response](https://cdn.futuresmart.ai/public/aidemos/8321848227434dd8887c9931002ff0a1.png?v=1)
*Output — The API test returned meeting note data successfully in the terminal, confirming the REST API works in this cycle.*

![Fathom output showing MCP connector](https://cdn.futuresmart.ai/public/aidemos/0b70efe5fb17468690a64895b71b30e1.png?v=1)
*Output — Claude's connector settings show Fathom connected, confirming MCP-style access is available in the product surface.*

![Fathom output showing Pricing page](https://cdn.futuresmart.ai/public/aidemos/7d8adc9f91144955bc63b90fec98e12f.png?v=1)
*Output — The pricing page shows Free, Premium, and Team plans, including the free tier that makes Fathom especially attractive for lighter usage.*

![Fathom output showing Integrations page](https://cdn.futuresmart.ai/public/aidemos/e0a2e4497d094cfb8eee39c7ba5f387c.png?v=1)
*Output — The integrations gallery includes Claude, ChatGPT, HubSpot, Asana, Slack, Salesforce, and major meeting platforms.*

![Fathom output showing Language support](https://cdn.futuresmart.ai/public/aidemos/4a9c60c486364eab960b9eb34f41cb4b.png?v=1)
*Output — The help article states Fathom supports transcripts in 38 languages.*

### MeetGeek

Solid summary and action-item output with useful analytics, but sharing was login-gated and diarization was only mixed.

![MeetGeek screenshot showing Shared meeting input](https://cdn.futuresmart.ai/public/aidemos/e57ffd56f5ec4bf98467a444de41dd92.png?v=1)
*Screenshot — Shared 31 July 2026 Google Meet standup used as the common test input across all eight tools, with a live recorded call and multiple notetaker bots present.*

**What worked:**
- MeetGeek produced structured notes, usable action items, search, chat, and a clear analytics layer. The API and MCP surfaces are also real, which makes it a credible operations-focused notetaker.

**Where it struggled:**
- Speaker diarization was only partial, shared links required login, and the plan limits are not especially generous for recurring team use. It is good, but the sharing friction keeps it below the top tier.

**What came out:**

![MeetGeek output showing Transcript output](https://cdn.futuresmart.ai/public/aidemos/9b8715e4533e4478b340c2748f00b5cd.png?v=1)
*Output — MeetGeek opened the meeting on the Conversation tab with transcript content visible under the player.*

![MeetGeek output showing Speaker attribution output](https://cdn.futuresmart.ai/public/aidemos/4000fee0fd6b46a4baf237080f616a84.png?v=1)
*Output — The transcript excerpt shows partial diarization, including an 'Unknown speaker' label alongside an identified speaker, which is usable but not perfect.*

![MeetGeek output showing Join confirmation](https://cdn.futuresmart.ai/public/aidemos/b8159d5a66bd4767bfaa0f2bc6e02218.png?v=1)
*Output — The participant list shows MeetGeek's notetaker bot in the meeting, confirming a successful join.*

![MeetGeek output showing Summary output](https://cdn.futuresmart.ai/public/aidemos/0f55639f0b024964933e6d8b9b0442bf.png?v=1)
*Output — The Summary tab shows a generated meeting summary with a next-steps section.*

![MeetGeek output showing Action items output](https://cdn.futuresmart.ai/public/aidemos/061bbe8a9cec405793de857d23b3f216.png?v=1)
*Output — The Next Steps section lists an extracted action item with a timestamp, confirming follow-up extraction worked.*

![MeetGeek output showing Topic segmentation output](https://cdn.futuresmart.ai/public/aidemos/e17c8cc9eaff4d909d0e4b7962b70f59.png?v=1)
*Output — The topics/highlights view breaks the meeting into numbered sections, showing useful topic segmentation.*

![MeetGeek output showing Editability view](https://cdn.futuresmart.ai/public/aidemos/464c87bc5d38441784e248ffb44a6f72.png?v=1)
*Output — The summary screen shows editing affordances, indicating the generated notes can be revised.*

![MeetGeek output showing Search result](https://cdn.futuresmart.ai/public/aidemos/95bea4badd5d41bb8f6684144eb71007.png?v=1)
*Output — Searching for API highlights the matched term in the transcript, showing searchable notes.*

![MeetGeek output showing Chat answer](https://cdn.futuresmart.ai/public/aidemos/e923e617611c4b96ac5254a5659e1412.png?v=1)
*Output — The Ask MeetGeek panel returned a specific meeting-date answer, showing working chat over the notes.*

![MeetGeek output showing Sharing dialog](https://cdn.futuresmart.ai/public/aidemos/cedf791ed3a84735996190bac3a07230.png?v=1)
*Output — The share modal exposes access controls and link sharing for the meeting recap.*

![MeetGeek output showing Login gate](https://cdn.futuresmart.ai/public/aidemos/810cf50a723f41ebb62df1da1705adc7.png?v=1)
*Output — A shared MeetGeek link opened a login screen, confirming external viewers are pushed to sign in.*

![MeetGeek output showing API response](https://cdn.futuresmart.ai/public/aidemos/74916c6af4d94bf48fefddb2f06d53f0.png?v=1)
*Output — The API request returned meeting highlights in the terminal, confirming programmatic access works.*

![MeetGeek output showing MCP connector](https://cdn.futuresmart.ai/public/aidemos/8bc4721910584001b634d7e08c1b95de.png?v=1)
*Output — Claude's permission screen shows the MeetGeek tool chain and a request to allow summary retrieval, confirming MCP connectivity.*

![MeetGeek output showing Pricing page](https://cdn.futuresmart.ai/public/aidemos/53dfabde3aa64f23bfe571dd2ae97845.png?v=1)
*Output — The pricing grid shows Basic, Pro, Business, and Enterprise plans with transcription limits and feature tiers.*

![MeetGeek output showing Integrations page](https://cdn.futuresmart.ai/public/aidemos/77d7af68d0864c1bbf2cc4cff9d69b19.png?v=1)
*Output — The integrations page groups supported apps across conferencing, CRM, collaboration, and automation categories.*

![MeetGeek output showing Language support](https://cdn.futuresmart.ai/public/aidemos/805dbd5630824e4da0f182f503198258.png?v=1)
*Output — The supported-languages page shows a multi-language support table for transcription, AI Summary & Insights, and AI Chat.*

![MeetGeek output showing Privacy policy](https://cdn.futuresmart.ai/public/aidemos/549f0e0f8c224f1ea3fd36a11f3fee3b.png?v=1)
*Output — The privacy policy page describes MeetGeek's data handling and limits on recording access.*

### Notta

Accurate capture with strong note editing and action-item extraction, but weak direct search and no public API in this cycle.

![Notta screenshot showing Shared meeting input](https://cdn.futuresmart.ai/public/aidemos/28980d4363994134bc0e8dfba38eb6ec.png?v=1)
*Screenshot — Shared 31 July 2026 Google Meet standup used as the common test input across all eight tools, with a live recorded call and multiple notetaker bots present.*

**What worked:**
- Notta captured the call cleanly, produced strong summaries and action items, and let users edit the outputs inline. The chat answer was also grounded and specific, which makes the app useful for quick post-meeting recall.

**Where it struggled:**
- Direct transcript search was limited, the public API was not available in this cycle, and diarization had some attribution problems plus a contamination concern in the evaluation. It is capable, but not the cleanest all-round team choice.

**What came out:**

![Notta output showing Transcript output](https://cdn.futuresmart.ai/public/aidemos/04d5579c26a1402597f90ed310d89842.png?v=1)
*Output — Notta produced a meeting transcript view with the record open in Transcript mode.*

![Notta output showing Speaker attribution output](https://cdn.futuresmart.ai/public/aidemos/0dfaa8e71f364886b25e8b8fad5d0e93.png?v=1)
*Output — The transcript shows mostly correct speaker attribution, but the report notes some mismatches and cross-tool contamination concerns on specific lines.*

![Notta output showing Join confirmation](https://cdn.futuresmart.ai/public/aidemos/f5c5b3d8549a4575a4f46db5d0e70dff.png?v=1)
*Output — The Google Meet participant panel shows Notta Bot present, confirming the bot joined the call.*

![Notta output showing Summary output](https://cdn.futuresmart.ai/public/aidemos/15ea1eaf67334c559c00982a7971fb72.png?v=1)
*Output — Notta generated a structured AI Notes summary with a project-sync heading and a task-management section.*

![Notta output showing Action items output](https://cdn.futuresmart.ai/public/aidemos/a67668b670864992821ce3ec522ef9d2.png?v=1)
*Output — The action-items panel lists real follow-ups with assignees/details, showing correct extraction in the tested meeting.*

![Notta output showing Topic segmentation output](https://cdn.futuresmart.ai/public/aidemos/e5804b6f935c495fb918d38003a44b23.png?v=1)
*Output — The notes are organized into topic-level sections, with the AI Notes and Notta Brain layout visible side by side.*

![Notta output showing Editability view](https://cdn.futuresmart.ai/public/aidemos/f2b14243fe5847028aa46fcbd413b8c4.png?v=1)
*Output — The summary text is shown in an editable state, confirming inline correction is available before sharing.*

![Notta output showing Search result](https://cdn.futuresmart.ai/public/aidemos/7fb40207e7134136b93478dc73b831b0.png?v=1)
*Output — The transcript view shows no dedicated search control in this screen, matching the report's finding that direct transcript search was limited.*

![Notta output showing Chat answer](https://cdn.futuresmart.ai/public/aidemos/a0065541283e4ae8b79d7327ddcf389e.png?v=1)
*Output — Notta Brain answered a question about a scheduled call with a specific date, demonstrating usable chat over the meeting record.*

![Notta output showing Sharing dialog](https://cdn.futuresmart.ai/public/aidemos/70b91f8699e446d1aa296b322f47c526.png?v=1)
*Output — The sharing settings allow public link access, with transcript and AI notes selectable for sharing.*

![Notta output showing API absence](https://cdn.futuresmart.ai/public/aidemos/4453f2a2c8d24c37abfd1ebf713faa9a.png?v=1)
*Output — The search result states that Notta does not publish a public self-serve REST API for standard tiers, which matches the report's API finding.*

![Notta output showing MCP package](https://cdn.futuresmart.ai/public/aidemos/15484eee1e0d49878a5db884c64ddd12.png?v=1)
*Output — The npm package page shows Notta's MCP server package, indicating a model-context integration path exists.*

![Notta output showing Pricing page](https://cdn.futuresmart.ai/public/aidemos/98c335ed1d79449aa01fe8a7629813e5.png?v=1)
*Output — The pricing page shows Free, Pro, Business, and Enterprise tiers with transcription and AI-summary limits.*

![Notta output showing Integrations page](https://cdn.futuresmart.ai/public/aidemos/c753323d03fe4fd582f781d501c71c21.png?v=1)
*Output — The integrations page lists services such as Zoom, Google Meet, Microsoft Teams, Slack, Notion, Salesforce, HubSpot, and Zapier.*

![Notta output showing Language support](https://cdn.futuresmart.ai/public/aidemos/3c29492510314972becf03f1ac7fee4f.png?v=1)
*Output — The help-center article lays out Notta's transcription and translation language support categories.*

![Notta output showing Privacy policy](https://cdn.futuresmart.ai/public/aidemos/06ad37fe7f324b2490eacfa3635176ef.png?v=1)
*Output — The privacy/security article says files are transcribed by machines and not seen by a human.*

### HappyScribe

Clear note structure and editing, but trust drops because of name hallucinations and limited direct search.

![HappyScribe screenshot showing Shared meeting input](https://cdn.futuresmart.ai/public/aidemos/42a4bb060f684c7c9e35f70dfcf6878f.png?v=1)
*Screenshot — Shared 31 July 2026 Google Meet standup used as the common test input across all eight tools, with a live recorded call and multiple notetaker bots present.*

**What worked:**
- HappyScribe can produce structured summaries, action items, editing controls, sharing, API access, and an MCP path. The note layout is clear and the outputs are easy to scan.

**Where it struggled:**
- The summary showed a name hallucination tied to transcription error, speaker attribution was not fully dependable, and direct transcript search was weak or absent in the tested UI. It is usable, but the trust layer needs work.

**What came out:**

![HappyScribe output showing Transcript output](https://cdn.futuresmart.ai/public/aidemos/fa81d17b16074d16952b38aca01a3a1e.png?v=1)
*Output — HappyScribe produced a generated transcript excerpt with speaker labeling and timestamps.*

![HappyScribe output showing Join confirmation](https://cdn.futuresmart.ai/public/aidemos/318f753c50084fdab85e38fd48164384.png?v=1)
*Output — The meeting header shows the HappyScribe AI Notetaker present, confirming the bot joined successfully.*

![HappyScribe output showing Summary output](https://cdn.futuresmart.ai/public/aidemos/3c090d3911b14059a46f133e62d9dbfe.png?v=1)
*Output — The summary page shows topic sections such as tool testing and access tracking, confirming structured note generation.*

![HappyScribe output showing Summary hallucination](https://cdn.futuresmart.ai/public/aidemos/55b2175d5e104e0aaee6d348094c6458.png?v=1)
*Output — The summary artifact shows a name error that the report traced to a misheard speaker name, which undermines trust in the generated recap.*

![HappyScribe output showing Action items output](https://cdn.futuresmart.ai/public/aidemos/6c8a81118abe4b47861829e7e9d0a9f4.png?v=1)
*Output — The action-items page lists a concrete follow-up task, showing the tool can extract real commitments.*

![HappyScribe output showing Topic segmentation output](https://cdn.futuresmart.ai/public/aidemos/7efe080286db46009d9e6d9fcc0bff0f.png?v=1)
*Output — The summary is split into topic sections, confirming a useful skimmable structure.*

![HappyScribe output showing Editability view](https://cdn.futuresmart.ai/public/aidemos/1d58ea4fa10e4cd39fe8723edecb2d5e.png?v=1)
*Output — The transcript or summary text is shown in an editable state, confirming inline correction is available.*

![HappyScribe output showing Search failure](https://cdn.futuresmart.ai/public/aidemos/c96fbc13d89f4a7a8388260c71f330d7.png?v=1)
*Output — The transcript screen shows no search affordance in this view, matching the report's finding that direct transcript search was not available.*

![HappyScribe output showing Chat answer](https://cdn.futuresmart.ai/public/aidemos/d0661ac7f145459d861464602f3dd603.png?v=1)
*Output — HappyScribe's AI chat answered a question about a scheduled call date with a direct answer of 6th August.*

![HappyScribe output showing Sharing dialog](https://cdn.futuresmart.ai/public/aidemos/bfb1f677d8c54c08b7af24a46f640bc4.png?v=1)
*Output — The share dialog shows workspace and link-sharing permissions, including an edit-capable public-link option.*

![HappyScribe output showing API response](https://cdn.futuresmart.ai/public/aidemos/47b0a2630bcd414280873e80ad4130b9.png?v=1)
*Output — The API request returned a 200 status and transcript listing data, confirming the API worked in this cycle.*

![HappyScribe output showing MCP prompt](https://cdn.futuresmart.ai/public/aidemos/8bc55dd2273f460c8b94cafa1f6981ba.png?v=1)
*Output — Claude's permission screen shows HappyScribe's search-transcriptions tool and an approval prompt, confirming the MCP path exists.*

![HappyScribe output showing Pricing page](https://cdn.futuresmart.ai/public/aidemos/27e6ac3ad28244ba94504dd2b06febfb.png?v=1)
*Output — The pricing page shows Free, Basic, Pro, and Business tiers and the minutes/chats included in each plan.*

![HappyScribe output showing Integrations page](https://cdn.futuresmart.ai/public/aidemos/cd93112e08ec4ff294305aff2466ae20.png?v=1)
*Output — The integrations directory includes MCP Server, Claude, ChatGPT, Zoom, Google Meet, and Microsoft Teams.*

![HappyScribe output showing Language support](https://cdn.futuresmart.ai/public/aidemos/2af9e8b123854e4e83a297dc8e530da1.png?v=1)
*Output — The language matrix lists supported transcription and subtitle languages across AI and human workflows.*

![HappyScribe output showing Privacy policy](https://cdn.futuresmart.ai/public/aidemos/d04857989b434a5f8d764763c3b1a1a1.png?v=1)
*Output — The privacy policy page introduces HappyScribe's data-protection posture and GDPR context.*

![HappyScribe output showing Search interface](https://cdn.futuresmart.ai/public/aidemos/3c73e73b7bc046bbadbf2fae6f20c202.png?v=1)
*Output — The transcript page shows review/export/share controls but no dedicated search option, reinforcing the limited-search finding.*

### Otter.ai

Feature-rich note surface with good transcript and chat surfaces, but diarization and action-item ownership fail too hard for standup accountability.

![Otter.ai screenshot showing Shared meeting input](https://cdn.futuresmart.ai/public/aidemos/13130835905a4f668ff6cc4d9084bc93.png?v=1)
*Screenshot — Shared 31 July 2026 Google Meet standup used as the common test input across all eight tools, with a live recorded call and multiple notetaker bots present.*

**What worked:**
- Otter captured a readable transcript, produced a structured summary, answered questions in chat, and exposed API/MCP and sharing surfaces. The product surface is broad and the transcript itself was not the weak point.

**Where it struggled:**
- The report's critical failure is accountability: diarization was too weak for a multi-speaker standup, and action-item owner assignment cascaded from that weakness. That makes the notes hard to trust for team follow-through.

**What came out:**

![Otter.ai output showing Transcript output](https://cdn.futuresmart.ai/public/aidemos/5d7cf53320d5450e98a6e81106e326e1.png?v=1)
*Output — Otter's transcript view shows speaker-labeled turns and the main transcript tab, confirming transcript capture succeeded.*

![Otter.ai output showing Speaker attribution output](https://cdn.futuresmart.ai/public/aidemos/2719bfbefd714164b16c0fe8252e7dab.png?v=1)
*Output — The transcript screen shows generic speaker buckets rather than clean attribution for the live standup, which is the core diarization failure in the report.*

![Otter.ai output showing Join confirmation](https://cdn.futuresmart.ai/public/aidemos/052591018a8041a2a3666646dfd4dc3d.png?v=1)
*Output — A Meet participant tile shows Mahreen's Notetaker, confirming the Otter bot joined the call.*

![Otter.ai output showing Join reliability evidence](https://cdn.futuresmart.ai/public/aidemos/5d4a3deb043f42ac92f4e27b1c5d6803.png?v=1)
*Output — The end-of-transcript evidence confirms the meeting reached its close without a bot dropout.*

![Otter.ai output showing Summary output](https://cdn.futuresmart.ai/public/aidemos/7d9e23f5e5a742728df303e4f1493b00.png?v=1)
*Output — The Summary tab shows an outline with named sections, confirming structured note generation.*

![Otter.ai output showing Action items output](https://cdn.futuresmart.ai/public/aidemos/3517b6d86f6c421bb0692139a4150daf.png?v=1)
*Output — The action-item section shows an extracted task, but the broader report says most action items lacked reliable owner assignment.*

![Otter.ai output showing Topic segmentation output](https://cdn.futuresmart.ai/public/aidemos/b2b2a8bb2a86416e9ee6c2e55008b9a0.png?v=1)
*Output — The outline is broken into topic sections such as issue/task assignments and error resolution, showing useful segmentation.*

![Otter.ai output showing Editability view](https://cdn.futuresmart.ai/public/aidemos/9683fdcd555d43dea5d541b21e67d060.png?v=1)
*Output — The summary bullet appears in an editable field, showing that the generated recap can be corrected before sharing.*

![Otter.ai output showing Search result](https://cdn.futuresmart.ai/public/aidemos/48980b8966f3428a83b129353a66d270.png?v=1)
*Output — Otter AI Chat answered a natural-language timestamp question with a precise time, showing note retrieval through chat.*

![Otter.ai output showing Sharing dialog](https://cdn.futuresmart.ai/public/aidemos/d83449cb0d8e40999ca5bcea863de57b.png?v=1)
*Output — The share dialog shows restricted access controls and copy-link/post-to-Slack actions for sharing the meeting record.*

![Otter.ai output showing API docs](https://cdn.futuresmart.ai/public/aidemos/498169362f5f4ba8a0ea243419a4d604.png?v=1)
*Output — The help-center page says the public API is available on the Enterprise plan and may require admin permissions.*

![Otter.ai output showing MCP connector](https://cdn.futuresmart.ai/public/aidemos/25898a105ffc42d7981a85b97318e191.png?v=1)
*Output — Claude's connector settings show Otter.ai connected, confirming a MCP-style connector path exists.*

![Otter.ai output showing Pricing page](https://cdn.futuresmart.ai/public/aidemos/e7f9e07e01c540f3965618ab24818fc5.png?v=1)
*Output — The pricing page shows Basic, Pro, Business, and Enterprise tiers and the features available on each plan.*

![Otter.ai output showing Integrations page](https://cdn.futuresmart.ai/public/aidemos/1f10068f6dc64ddcbe21de12306a3bcf.png?v=1)
*Output — The integrations directory shows connectors such as Airtable, Amazon S3, Android, and Asana.*

![Otter.ai output showing Language support](https://cdn.futuresmart.ai/public/aidemos/ac2ca2601c214845bb4b1c99561aff8a.png?v=1)
*Output — The language article states Otter supports English, Spanish, French, German, Japanese, and Chinese (Simplified), with extra accent-handling notes.*

![Otter.ai output showing Privacy policy](https://cdn.futuresmart.ai/public/aidemos/889aac8373114a1a91e40c57ea322365.png?v=1)
*Output — The privacy policy page documents Otter's handling of personal information and enterprise processing context.*

### Granola

Botless desktop capture with strong privacy posture and useful downstream outputs, but transcript quality and default speaker tagging were not good enough for this standup.

![Granola screenshot showing Shared meeting input](https://cdn.futuresmart.ai/public/aidemos/bb40ed8967e84872b4f6572ab55cc2f4.png?v=1)
*Screenshot — Shared 31 July 2026 Google Meet standup used as the common test input across all eight tools, with a live recorded call and multiple notetaker bots present.*

**What worked:**
- Granola's botless capture is appealing, and the downstream notes features are useful: the summary is structured, action items are real, search works, the chat answer is grounded, and the API/MCP paths are real. It is also privacy-forward and convenient because no meeting bot has to join the call.

**Where it struggled:**
- The transcript quality was poor enough to break accountability, and speaker tagging ships off by default, which makes the baseline experience easy to miss for first-time users. That is a serious problem for a meeting-notetaker category where attribution matters.

**What came out:**

![Granola output showing Transcript output](https://cdn.futuresmart.ai/public/aidemos/a13551b2e9b44231a2b3be6d5d747007.png?v=1)
*Output — Granola rendered the call as transcript-like text blocks, but the broader report judged the transcript quality unreliable across the meeting.*

![Granola output showing Speaker attribution output](https://cdn.futuresmart.ai/public/aidemos/ef6a874fb4da4a3ab3498cd4aafb621d.png?v=1)
*Output — The speaker-tags setting is shown in the off state, which explains why the default transcript had no speaker attribution.*

![Granola output showing Join confirmation](https://cdn.futuresmart.ai/public/aidemos/562e538ac2bb413fa813f5a0143af621.png?v=1)
*Output — A desktop banner says a meeting was detected in Chrome and offers Take notes, confirming the botless local capture flow.*

![Granola output showing End-of-transcript output](https://cdn.futuresmart.ai/public/aidemos/bd70f29023af46cca5d9ecc79f6ada68.png?v=1)
*Output — The end of the transcript shows the closing turns of the meeting, confirming full-duration capture in the local app.*

![Granola output showing Summary output](https://cdn.futuresmart.ai/public/aidemos/e2a9de216633445a99e4fe0818820b7a.png?v=1)
*Output — The summary/document view shows section headings for use-case status and tool research, indicating a structured note output.*

![Granola output showing Action items output](https://cdn.futuresmart.ai/public/aidemos/1e8af184699f41d482d7b0f408d92525.png?v=1)
*Output — The extracted next-step note shows a real follow-up task, including the instruction to create a subtask and tag the owner once done.*

![Granola output showing Topic segmentation output](https://cdn.futuresmart.ai/public/aidemos/c1219da09aec450a909ac21ee5e4ee36.png?v=1)
*Output — The notes are split into a named topic section, showing that Granola can segment meeting content into useful headings.*

![Granola output showing Editability view](https://cdn.futuresmart.ai/public/aidemos/fcf4453c15b9490aa755d9c3753295dd.png?v=1)
*Output — The summary view is shown in a minimal editable state, confirming note text can be adjusted.*

![Granola output showing Search result](https://cdn.futuresmart.ai/public/aidemos/56410184e9074485807cc846923578d7.png?v=1)
*Output — The in-notes search interface found the term API and highlighted the match, showing search across the note content worked.*

![Granola output showing Chat answer](https://cdn.futuresmart.ai/public/aidemos/b18093276e714a3ea4ffb6c62b5ff39b.png?v=1)
*Output — The chat snippet answers a question about tool access and suggests the next step was to rerun testing, showing grounded note Q&A.*

![Granola output showing Sharing dialog](https://cdn.futuresmart.ai/public/aidemos/4e58355a0aa146c5908558dbc7b3cdd3.png?v=1)
*Output — The share-notes dialog shows link sharing with anyone-with-link access and a copy-link control.*

![Granola output showing API response](https://cdn.futuresmart.ai/public/aidemos/b269b55ec77041949518fea881afecaa.png?v=1)
*Output — The API request succeeded and returned note JSON, confirming Granola's public API access worked in this cycle.*

![Granola output showing MCP answer](https://cdn.futuresmart.ai/public/aidemos/8d2a5dee775f4be297561634ff88a2c9.png?v=1)
*Output — The MCP Q&A result shows the Granola integration was used and the access status answer was returned from the meeting data.*

![Granola output showing Pricing page](https://cdn.futuresmart.ai/public/aidemos/f6ceed3b99e14772a9a5e63198b48158.png?v=1)
*Output — The pricing page shows Basic, Business, and Enterprise plan columns with their prices and actions.*

![Granola output showing Integrations page](https://cdn.futuresmart.ai/public/aidemos/54d6de146bfe45f8bcb3f0e44c798f55.png?v=1)
*Output — The help-center integrations page lists Zapier, Notion, Slack, CRMs, MCP, and Granola API resources.*

![Granola output showing Language support](https://cdn.futuresmart.ai/public/aidemos/a9c8a508b58e4359a2008541b5777da4.png?v=1)
*Output — The help-center article explains multilingual support and lists the supported transcription and summary languages.*

![Granola output showing Privacy policy](https://cdn.futuresmart.ai/public/aidemos/70b4555855784d319a9d21729e0d7699.png?v=1)
*Output — The privacy policy page shows the effective date and the opening policy text describing Granola's AI-powered notetaking services.*

### Fireflies.ai

Bot-based meeting assistant with strong transcript, summary, action-item, search, chat, and MCP surfaces, plus broad integrations and a generous free tier.

![Fireflies.ai screenshot showing Shared meeting input](https://cdn.futuresmart.ai/public/aidemos/36a9e821288a46fb84fdbd3d1e1daba6.png?v=1)
*Screenshot — Shared 31 July 2026 Google Meet standup used as the common test input across all eight tools, with a live recorded call and multiple notetaker bots present.*

**What worked:**
- Fireflies handled the core meeting-memory workflow very well: the transcript was accurate, speaker labeling was strong, summaries and action items were structured, search and AskFred worked, and the Claude/MCP path was present in the product surface. The integrations and language coverage are also broad, which makes it practical for teams.

**Where it struggled:**
- Sharing still pushed recipients toward login, the API test produced an error in this cycle, and editing the generated notes was limited on the tested screen. It is strong, but not frictionless on every operational axis.

**What came out:**

![Fireflies.ai output showing Transcript output](https://cdn.futuresmart.ai/public/aidemos/646a8d7a4bff412e965d9f23e86fc62a.png?v=1)
*Output — Fireflies generated a readable transcript for the standup, with timestamped turns visible in the transcript panel.*

![Fireflies.ai output showing Speaker attribution output](https://cdn.futuresmart.ai/public/aidemos/033cd05477174bdfa7d623a7a7ee10b8.png?v=1)
*Output — Speaker labeling worked on the tested transcript, with consecutive turns attributed to different team members rather than collapsing into a single unlabeled block.*

![Fireflies.ai output showing Join confirmation](https://cdn.futuresmart.ai/public/aidemos/9486a3566db14c02a31d0632b6b390b8.png?v=1)
*Output — The Fireflies bot appeared in the meeting participant list, confirming it joined the Google Meet call successfully.*

![Fireflies.ai output showing Summary output](https://cdn.futuresmart.ai/public/aidemos/31235f51f90647dca9bf55240e83206c.png?v=1)
*Output — Fireflies produced a structured meeting summary with a clear notes section and a topic heading summarizing the call.*

![Fireflies.ai output showing Action items output](https://cdn.futuresmart.ai/public/aidemos/d2b48f7c215848608e9695c40b2e08e9.png?v=1)
*Output — The action-items panel grouped follow-ups by owner and included timestamped references, showing the extracted tasks were tied to the meeting content.*

![Fireflies.ai output showing Topic segmentation output](https://cdn.futuresmart.ai/public/aidemos/4a0e08d8aacd43a19b35751d2c8314bf.png?v=1)
*Output — The notes were organized into topic-level sections rather than a single block, making the standup easy to scan.*

![Fireflies.ai output showing Editability view](https://cdn.futuresmart.ai/public/aidemos/64699d6aa730406cb495ee6eaacf7219.png?v=1)
*Output — The summary view exposed copy-only controls, which means the generated notes were not directly editable from this screen on the tested tier.*

![Fireflies.ai output showing Search result](https://cdn.futuresmart.ai/public/aidemos/d8c8ab24fc6740739a97e9b88a0cb9fb.png?v=1)
*Output — Transcript search found the queried term and highlighted the match in context, confirming keyword search worked within the meeting.*

![Fireflies.ai output showing Chat answer](https://cdn.futuresmart.ai/public/aidemos/10dddd5d43a34695bcd53a1688fa8c15.png?v=1)
*Output — AskFred answered a natural-language question with a specific date, showing grounded chat over the meeting record.*

![Fireflies.ai output showing Sharing dialog](https://cdn.futuresmart.ai/public/aidemos/a5441e74a5be463299a72bdb40ee8ae9.png?v=1)
*Output — The share modal exposed link-sharing and access controls, but the separate login-gate evidence shows recipients still have to authenticate to view the meeting.*

![Fireflies.ai output showing Login gate](https://cdn.futuresmart.ai/public/aidemos/d0ffe8162ff643ca960dfde42e4bfedb.png?v=1)
*Output — A shared Fireflies link presented a sign-in modal, confirming that external viewers were not able to open the meeting without logging in.*

![Fireflies.ai output showing API output](https://cdn.futuresmart.ai/public/aidemos/76be14a5edb54647a2d2fab758a60973.png?v=1)
*Output — The API test returned a 'User not found' error in the terminal, so API access was not cleanly validated in this cycle.*

![Fireflies.ai output showing MCP connector](https://cdn.futuresmart.ai/public/aidemos/31ab38b4b9484ebbadecdbf04861b55c.png?v=1)
*Output — The MCP evidence slide shows a working Claude-style query over Fireflies meeting data, returning a grounded answer about access status.*

![Fireflies.ai output showing Pricing page](https://cdn.futuresmart.ai/public/aidemos/a51f66b729684f9498d71c304e68fe46.png?v=1)
*Output — The pricing page shows Free, Pro, Business, and Enterprise tiers, including the free plan and the feature ladder used in the evaluation.*

![Fireflies.ai output showing Integrations page](https://cdn.futuresmart.ai/public/aidemos/da7c94f9254e46eba60646c423c3bbbc.png?v=1)
*Output — The integrations directory shows Fireflies connectors such as Google Meet, Dropbox, HubSpot, and Microsoft Teams, supporting broad workflow coverage.*

![Fireflies.ai output showing Language support](https://cdn.futuresmart.ai/public/aidemos/f46cfdcad0a84a3cb3125c89264ece07.png?v=1)
*Output — The language-support page states Fireflies supports transcription and summaries in 100+ languages, including auto-detect and multi-language meeting support.*

![Fireflies.ai output showing Privacy policy](https://cdn.futuresmart.ai/public/aidemos/0052174989794562880562d833f5bb0d.png?v=1)
*Output — The privacy policy page states how Fireflies processes user information and documents its data-handling posture.*

## Evidence (first-party, tested)

*101 tested cells · 94/101 artifact-verified. Cite a cell by its Evidence ID, e.g. `ev:fathom·ai-demos-daily-standup-31-july-2026·action-item-extraction`.*

| Tool | Criterion | Scenario | Verdict | Proof | Evidence ID |
| --- | --- | --- | --- | --- | --- |
| Fathom | Action-Item Extraction | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/9e8a3e7fb669448e959de254a7dcec48.png?v=1) | `ev:fathom·ai-demos-daily-standup-31-july-2026·action-item-extraction` |
| Fathom | API Availability | cross-scenario | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/8a41e1e272224a109d4219fe9559f64c.png?v=1) | `ev:fathom·cross·api-availability` |
| Fathom | Chat with Notes / Ask Questions | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/cc0b830d51cd48a2a6c9121c331b1626.png?v=1) | `ev:fathom·ai-demos-daily-standup-31-july-2026·chat-with-notes-ask-questions` |
| Fathom | Editability | cross-scenario | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/250e06aaf20943829541f1fa853bc1fc.png?v=1) | `ev:fathom·cross·editability` |
| Fathom | Editability | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/f95956fe337641f39d3a44becd5b4378.png?v=1) | `ev:fathom·ai-demos-daily-standup-31-july-2026·editability` |
| Fathom | Join Method & Reliability | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/ca085ba7495e48de8bcb08c818d56a36.png?v=1) | `ev:fathom·ai-demos-daily-standup-31-july-2026·join-method-reliability` |
| Fathom | Join Method & Reliability | cross-scenario | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/f28078c69a7b406aabd62475b8f0ee41.png?v=1) | `ev:fathom·cross·join-method-reliability` |
| Fathom | MCP Support | cross-scenario | ◐ mixed | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/04d5d851757f4b708f0a3ddcf2b48ed2.png?v=1) | `ev:fathom·cross·mcp-support` |
| Fathom | MCP Support | AI Demos Daily Standup — 31 July 2026 | ◐ mixed | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/faa59fe85b0f44c888541a26e2b56521.png?v=1) | `ev:fathom·ai-demos-daily-standup-31-july-2026·mcp-support` |
| Fathom | Search Across Notes | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/1c674d0990a14c97bed5954bfe61cd0d.png?v=1) | `ev:fathom·ai-demos-daily-standup-31-july-2026·search-across-notes` |
| Fathom | Sharing Without Registration | cross-scenario | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/0ce6c776c62142ef8a96c28e9c63d489.png?v=1) | `ev:fathom·cross·sharing-without-registration` |
| Fathom | Sharing Without Registration | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/7ee71ea221914e3784453ada73809c72.png?v=1) | `ev:fathom·ai-demos-daily-standup-31-july-2026·sharing-without-registration` |
| Fathom | Speaker Diarization | AI Demos Daily Standup — 31 July 2026 | ◐ mixed | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/e6b553ff015d4c309c161b0bb12a1660.png?v=1) | `ev:fathom·ai-demos-daily-standup-31-july-2026·speaker-diarization` |
| Fathom | Speaker Diarization | cross-scenario | ◐ mixed | 👁 observed | `ev:fathom·cross·speaker-diarization` |
| Fathom | Summary Quality | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/7a6b7c6bede341b1a0f8d941ba783549.png?v=1) | `ev:fathom·ai-demos-daily-standup-31-july-2026·summary-quality` |
| Fathom | Topic Segmentation | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/1d727e487573413ab4068646f67ff0af.png?v=1) | `ev:fathom·ai-demos-daily-standup-31-july-2026·topic-segmentation` |
| Fathom | Transcription Accuracy | AI Demos Daily Standup — 31 July 2026 | ◐ mixed | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/19b062d60f8e45a6a5fa364948287362.png?v=1) | `ev:fathom·ai-demos-daily-standup-31-july-2026·transcription-accuracy` |
| Fathom | Transcription Accuracy | cross-scenario | ◐ mixed | 👁 observed | `ev:fathom·cross·transcription-accuracy` |
| Fellow | Action-Item Extraction | AI Demos Daily Standup — 31 July 2026 | ◐ mixed | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/2f38a356a98d4395b43001c1d7f6c5b3.png?v=1) | `ev:fellow·ai-demos-daily-standup-31-july-2026·action-item-extraction` |
| Fellow | API Availability | cross-scenario | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/df451104d29b472097079e271286c5f1.png?v=1) | `ev:fellow·cross·api-availability` |
| Fellow | Chat with Notes / Ask Questions | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/553c970153974fc9a12d999ec7a3b888.png?v=1) | `ev:fellow·ai-demos-daily-standup-31-july-2026·chat-with-notes-ask-questions` |
| Fellow | Editability | AI Demos Daily Standup — 31 July 2026 | ◐ mixed | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/14b659d136f643cca123a6031ecfdee2.png?v=1) | `ev:fellow·ai-demos-daily-standup-31-july-2026·editability` |
| Fellow | Editability | cross-scenario | ◐ mixed | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/b040bd918523461c937d91e952aeff8d.png?v=1) | `ev:fellow·cross·editability` |
| Fellow | Join Method & Reliability | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/3a01097650fc49a2827e4f746d150bec.png?v=1) | `ev:fellow·ai-demos-daily-standup-31-july-2026·join-method-reliability` |
| Fellow | MCP Support | cross-scenario | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/f51e407115b74031a505eefaab5492e9.png?v=1) | `ev:fellow·cross·mcp-support` |
| Fellow | Search Across Notes | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/8f4552025cb34119b9efac386b0c3c65.png?v=1) | `ev:fellow·ai-demos-daily-standup-31-july-2026·search-across-notes` |
| Fellow | Sharing Without Registration | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/bf43c36b12f3477d8e3fb105a5757de5.png?v=1) | `ev:fellow·ai-demos-daily-standup-31-july-2026·sharing-without-registration` |
| Fellow | Sharing Without Registration | cross-scenario | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/9fbf3d36dea1488cb50441241d1f168f.png?v=1) | `ev:fellow·cross·sharing-without-registration` |
| Fellow | Speaker Diarization | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/f6c095044c7a4498b82f300b683163dd.png?v=1) | `ev:fellow·ai-demos-daily-standup-31-july-2026·speaker-diarization` |
| Fellow | Summary Quality | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/e4f4726c9f8b4d8e9ec8e58a8faf8a3a.png?v=1) | `ev:fellow·ai-demos-daily-standup-31-july-2026·summary-quality` |
| Fellow | Topic Segmentation | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/e9cfe1cbf2b543a3a9ce3c67d83d21ff.png?v=1) | `ev:fellow·ai-demos-daily-standup-31-july-2026·topic-segmentation` |
| Fellow | Transcription Accuracy | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/7476ae6a377e411094940f94f34b21f9.png?v=1) | `ev:fellow·ai-demos-daily-standup-31-july-2026·transcription-accuracy` |
| Granola | Action-Item Extraction | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/0a21e41b32d648b1b5206dc034a58201.png?v=1) | `ev:granola·ai-demos-daily-standup-31-july-2026·action-item-extraction` |
| Granola | API Availability | cross-scenario | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/588de32058dc4da6bebffdac070696fa.png?v=1) | `ev:granola·cross·api-availability` |
| Granola | Chat with Notes / Ask Questions | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/8b17d8ed350645bfb4a942128365177f.png?v=1) | `ev:granola·ai-demos-daily-standup-31-july-2026·chat-with-notes-ask-questions` |
| Granola | Editability | AI Demos Daily Standup — 31 July 2026 | ◐ mixed | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/d2e0703f86a040cea768260e1ddae9f4.png?v=1) | `ev:granola·ai-demos-daily-standup-31-july-2026·editability` |
| Granola | Join Method & Reliability | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/7fcbe8cb7009429e98b95bdcabbaddcd.png?v=1) | `ev:granola·ai-demos-daily-standup-31-july-2026·join-method-reliability` |
| Granola | MCP Support | cross-scenario | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/897b3b8f239a4964ad154607853d3348.png?v=1) | `ev:granola·cross·mcp-support` |
| Granola | Search Across Notes | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/2f0d48a2a7b443c2bba29bfa9da76c9f.png?v=1) | `ev:granola·ai-demos-daily-standup-31-july-2026·search-across-notes` |
| Granola | Sharing Without Registration | cross-scenario | ◐ mixed | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/e0c00226f6a340e3b1144f0c1539417b.png?v=1) | `ev:granola·cross·sharing-without-registration` |
| Granola | Speaker Diarization | AI Demos Daily Standup — 31 July 2026 | ✗ failed | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/ecd4eac38c5f4f1f9e0d671349a4f71a.png?v=1) | `ev:granola·ai-demos-daily-standup-31-july-2026·speaker-diarization` |
| Granola | Summary Quality | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/eed3e4d591e047e2990d9327d214923e.png?v=1) | `ev:granola·ai-demos-daily-standup-31-july-2026·summary-quality` |
| Granola | Topic Segmentation | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/5343a488a2ad46a48618165dac495d7c.png?v=1) | `ev:granola·ai-demos-daily-standup-31-july-2026·topic-segmentation` |
| Granola | Transcription Accuracy | AI Demos Daily Standup — 31 July 2026 | ✗ failed | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/d3ec039f484d40328057665e51ba34de.png?v=1) | `ev:granola·ai-demos-daily-standup-31-july-2026·transcription-accuracy` |
| HappyScribe | Action-Item Extraction | AI Demos Daily Standup — 31 July 2026 | ◐ mixed | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/672943809b1d40c483bdee07651023c2.png?v=1) | `ev:happyscribe·ai-demos-daily-standup-31-july-2026·action-item-extraction` |
| HappyScribe | Action-Item Extraction | cross-scenario | ◐ mixed | 👁 observed | `ev:happyscribe·cross·action-item-extraction` |
| HappyScribe | API Availability | cross-scenario | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/bdb13fdfec4f460ebfee0017b9ad0d8c.png?v=1) | `ev:happyscribe·cross·api-availability` |
| HappyScribe | API Availability | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/10669edaf2084560962ad04281c80df8.png?v=1) | `ev:happyscribe·ai-demos-daily-standup-31-july-2026·api-availability` |
| HappyScribe | Chat with Notes / Ask Questions | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/04410e535a0e4f95add9df3c5cd7e79c.png?v=1) | `ev:happyscribe·ai-demos-daily-standup-31-july-2026·chat-with-notes-ask-questions` |
| HappyScribe | Editability | cross-scenario | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/7e4630d765ea4655a9734da07932d7ef.png?v=1) | `ev:happyscribe·cross·editability` |
| HappyScribe | Editability | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/433723b3549846b8af78c4f2c45860bc.png?v=1) | `ev:happyscribe·ai-demos-daily-standup-31-july-2026·editability` |
| HappyScribe | Join Method & Reliability | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/e8b523853472416086473489ced66b2b.png?v=1) | `ev:happyscribe·ai-demos-daily-standup-31-july-2026·join-method-reliability` |
| HappyScribe | MCP Support | cross-scenario | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/ed2f14b587ec4ae1ae671a0b5ed09cbb.png?v=1) | `ev:happyscribe·cross·mcp-support` |
| HappyScribe | MCP Support | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/391b9a028a0749e398e6b5c45d5d3efb.png?v=1) | `ev:happyscribe·ai-demos-daily-standup-31-july-2026·mcp-support` |
| HappyScribe | Search Across Notes | AI Demos Daily Standup — 31 July 2026 | ⚠ struggled | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/b87b83ed4ad74856afc0e0c56a9589af.png?v=1) | `ev:happyscribe·ai-demos-daily-standup-31-july-2026·search-across-notes` |
| HappyScribe | Sharing Without Registration | cross-scenario | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/9af04d3474cf438bb74f4d19fb3142f6.png?v=1) | `ev:happyscribe·cross·sharing-without-registration` |
| HappyScribe | Sharing Without Registration | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/8c2c046734bc49489f457e4a2f9b72ce.png?v=1) | `ev:happyscribe·ai-demos-daily-standup-31-july-2026·sharing-without-registration` |
| HappyScribe | Speaker Diarization | AI Demos Daily Standup — 31 July 2026 | ◐ mixed | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/64855a79c2564122bac93c704797f365.png?v=1) | `ev:happyscribe·ai-demos-daily-standup-31-july-2026·speaker-diarization` |
| HappyScribe | Summary Quality | AI Demos Daily Standup — 31 July 2026 | ◐ mixed | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/ac13349047eb4335af7ffc2f1f798d7e.png?v=1) | `ev:happyscribe·ai-demos-daily-standup-31-july-2026·summary-quality` |
| HappyScribe | Summary Quality | cross-scenario | ◐ mixed | 👁 observed | `ev:happyscribe·cross·summary-quality` |
| HappyScribe | Topic Segmentation | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/bf864a8106ca4c3db1b8e79103b8e4f2.png?v=1) | `ev:happyscribe·ai-demos-daily-standup-31-july-2026·topic-segmentation` |
| HappyScribe | Transcription Accuracy | AI Demos Daily Standup — 31 July 2026 | ◐ mixed | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/768c8f926786411984134803cb5396a5.png?v=1) | `ev:happyscribe·ai-demos-daily-standup-31-july-2026·transcription-accuracy` |
| MeetGeek | Action-Item Extraction | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/39691e7c2e9141b3aac5a412f3bb110d.png?v=1) | `ev:meetgeek·ai-demos-daily-standup-31-july-2026·action-item-extraction` |
| MeetGeek | API Availability | cross-scenario | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/f28c0980d50f403b97cbcf8c336f2367.png?v=1) | `ev:meetgeek·cross·api-availability` |
| MeetGeek | Chat with Notes / Ask Questions | AI Demos Daily Standup — 31 July 2026 | ◐ mixed | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/e84f2cebcccd4af092aff8637baab736.png?v=1) | `ev:meetgeek·ai-demos-daily-standup-31-july-2026·chat-with-notes-ask-questions` |
| MeetGeek | Chat with Notes / Ask Questions | cross-scenario | ◐ mixed | 👁 observed | `ev:meetgeek·cross·chat-with-notes-ask-questions` |
| MeetGeek | Editability | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/c7ceaf56115b45bb86e3c89f78424ea4.png?v=1) | `ev:meetgeek·ai-demos-daily-standup-31-july-2026·editability` |
| MeetGeek | Join Method & Reliability | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/75a3b9a7a5a744e48ae416a4c23cd654.png?v=1) | `ev:meetgeek·ai-demos-daily-standup-31-july-2026·join-method-reliability` |
| MeetGeek | MCP Support | cross-scenario | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/56e50078302342ba9710546c1d1a1f48.png?v=1) | `ev:meetgeek·cross·mcp-support` |
| MeetGeek | Search Across Notes | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/f6217f0566a14109934b7cab87f91009.png?v=1) | `ev:meetgeek·ai-demos-daily-standup-31-july-2026·search-across-notes` |
| MeetGeek | Sharing Without Registration | cross-scenario | ✗ failed | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/e2c372baa01141b5b9bed2f378d807f9.png?v=1) | `ev:meetgeek·cross·sharing-without-registration` |
| MeetGeek | Speaker Diarization | AI Demos Daily Standup — 31 July 2026 | ◐ mixed | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/a909947fdbe64d41b2bc10dca0f44193.png?v=1) | `ev:meetgeek·ai-demos-daily-standup-31-july-2026·speaker-diarization` |
| MeetGeek | Summary Quality | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/348b3c4459f34fe8b0214debd8e0998b.png?v=1) | `ev:meetgeek·ai-demos-daily-standup-31-july-2026·summary-quality` |
| MeetGeek | Topic Segmentation | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/802e2202df02459b9fe81b754b08212f.png?v=1) | `ev:meetgeek·ai-demos-daily-standup-31-july-2026·topic-segmentation` |
| MeetGeek | Transcription Accuracy | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/f3fbc51e628c42c19595b82c5e481a25.png?v=1) | `ev:meetgeek·ai-demos-daily-standup-31-july-2026·transcription-accuracy` |
| Notta | Action-Item Extraction | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/22b60eda9b5e4b05b6ad287c471e5e3c.png?v=1) | `ev:notta·ai-demos-daily-standup-31-july-2026·action-item-extraction` |
| Notta | API Availability | cross-scenario | ✗ failed | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/71917796a9a74f56b5145907765c61ae.png?v=1) | `ev:notta·cross·api-availability` |
| Notta | Chat with Notes / Ask Questions | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/a822259cb8bf403a96a2d5814c1b7036.png?v=1) | `ev:notta·ai-demos-daily-standup-31-july-2026·chat-with-notes-ask-questions` |
| Notta | Editability | cross-scenario | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/8f9e1241db444bf298f144d498c17aa8.png?v=1) | `ev:notta·cross·editability` |
| Notta | Join Method & Reliability | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/00e1f74ee5b94aa7ac2fcb3fa7e3cf49.png?v=1) | `ev:notta·ai-demos-daily-standup-31-july-2026·join-method-reliability` |
| Notta | MCP Support | cross-scenario | ◐ mixed | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/6d806ef077a8411eabae8bb80c0357b2.png?v=1) | `ev:notta·cross·mcp-support` |
| Notta | Search Across Notes | AI Demos Daily Standup — 31 July 2026 | ◐ mixed | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/6f2c790bed6f4cb1bc9611c89957bda8.png?v=1) | `ev:notta·ai-demos-daily-standup-31-july-2026·search-across-notes` |
| Notta | Sharing Without Registration | cross-scenario | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/04e04021d3ba41f88ea9502065434726.png?v=1) | `ev:notta·cross·sharing-without-registration` |
| Notta | Speaker Diarization | AI Demos Daily Standup — 31 July 2026 | ⚠ struggled | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/665e1b0e39b34e40bb46f742733e7db7.png?v=1) | `ev:notta·ai-demos-daily-standup-31-july-2026·speaker-diarization` |
| Notta | Speaker Diarization | cross-scenario | ◐ mixed | 👁 observed | `ev:notta·cross·speaker-diarization` |
| Notta | Summary Quality | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/6868c5d787734044ae365ce951e12c06.png?v=1) | `ev:notta·ai-demos-daily-standup-31-july-2026·summary-quality` |
| Notta | Topic Segmentation | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/84691bb3ad6d4ac19dec1c5f25f3a0a9.png?v=1) | `ev:notta·ai-demos-daily-standup-31-july-2026·topic-segmentation` |
| Notta | Transcription Accuracy | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/86173a6a33c7494b96de40392d0740b6.png?v=1) | `ev:notta·ai-demos-daily-standup-31-july-2026·transcription-accuracy` |
| Otter.ai | Action-Item Extraction | AI Demos Daily Standup — 31 July 2026 | ⚠ struggled | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/7e0a51fd2a4744aea043309aec25fc20.png?v=1) | `ev:otter-ai·ai-demos-daily-standup-31-july-2026·action-item-extraction` |
| Otter.ai | Action-Item Extraction | cross-scenario | ✗ failed | 👁 observed | `ev:otter-ai·cross·action-item-extraction` |
| Otter.ai | API Availability | cross-scenario | ◐ mixed | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/0def4a3b2a4b42c3a2874e1ead7f117b.png?v=1) | `ev:otter-ai·cross·api-availability` |
| Otter.ai | Chat with Notes / Ask Questions | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/e5aefd3684124771a718737d9c8d1754.png?v=1) | `ev:otter-ai·ai-demos-daily-standup-31-july-2026·chat-with-notes-ask-questions` |
| Otter.ai | Editability | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/a0307003e47741e68372b1f2d17c91ec.png?v=1) | `ev:otter-ai·ai-demos-daily-standup-31-july-2026·editability` |
| Otter.ai | Join Method & Reliability | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/0db8542fb3c24e2e936f4fb01433dcac.png?v=1) | `ev:otter-ai·ai-demos-daily-standup-31-july-2026·join-method-reliability` |
| Otter.ai | MCP Support | cross-scenario | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/e71fdd943fd247c5b79479868abbf930.png?v=1) | `ev:otter-ai·cross·mcp-support` |
| Otter.ai | Search Across Notes | AI Demos Daily Standup — 31 July 2026 | ◐ mixed | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/6b991fb73fbd4561a65d6a295b698f7e.png?v=1) | `ev:otter-ai·ai-demos-daily-standup-31-july-2026·search-across-notes` |
| Otter.ai | Sharing Without Registration | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/8abd6d08b2304fd7ac5ff846bb428173.png?v=1) | `ev:otter-ai·ai-demos-daily-standup-31-july-2026·sharing-without-registration` |
| Otter.ai | Speaker Diarization | AI Demos Daily Standup — 31 July 2026 | ✗ failed | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/c2c822851e2f4e71a1ae3b14247c11d3.png?v=1) | `ev:otter-ai·ai-demos-daily-standup-31-july-2026·speaker-diarization` |
| Otter.ai | Summary Quality | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/fb9982daf5bf4988b132cda637bf3053.png?v=1) | `ev:otter-ai·ai-demos-daily-standup-31-july-2026·summary-quality` |
| Otter.ai | Topic Segmentation | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/22a79c0f848e403c82d0a767e9e00914.png?v=1) | `ev:otter-ai·ai-demos-daily-standup-31-july-2026·topic-segmentation` |
| Otter.ai | Transcription Accuracy | AI Demos Daily Standup — 31 July 2026 | ✓ worked | 🧾 [proof](https://cdn.futuresmart.ai/public/aidemos/0db942cf24dd4c3f9383ac4f5208e840.png?v=1) | `ev:otter-ai·ai-demos-daily-standup-31-july-2026·transcription-accuracy` |

> 🧾 = artifact-verified (proof captured) · 👁 = observed (noted, no artifact) · verdicts: worked / mixed / struggled / failed.

## Final Take

Fellow is the overall winner here because the page ranks it #1 and its decisive-check profile is the most consistently strong: it scores 5.0 on join method & reliability, speaker diarization, summary quality, and transcription accuracy, with only a minor miss on action-item extraction at 4.0. The main trade-off is that it is not the strongest on every downstream workflow feature: editability is only 3.0, so teams that need heavier post-meeting editing may prefer another tool. Fireflies.ai is the main alternative if action-item extraction is the priority, since it scores 5.0 there and also matches Fellow on capture, summaries, and transcription, but it gives up diarization quality, editing, sharing without registration, and API access. Notta is a good fit when you want strong capture, summaries, and action items plus full editability, but its weaker speaker diarization and search across notes make it less rounded than Fellow. Fathom stands out for capture, search, sharing, and a strong overall workflow, but its lower diarization and MCP score keep it behind the top tier. MeetGeek is appealing for capture, notes, search, API, and MCP, though chat with notes and sharing without registration are weak points. Otter.ai fits users who value capture, summaries, chat, and collaboration, but it is held back sharply by weak action-item extraction and speaker diarization. HappyScribe is more of a generalist for capture, chat, API, and sharing, but its lower summary, transcription, and search scores make it less compelling for this job. Granola is best when you care about grounded downstream workflows, but its very weak transcription accuracy and diarization make it the riskiest choice for raw meeting fidelity.

Tested as of 2026-08-12 · re-verified monthly.

## Need a custom AI solution for this use case?

If you are looking to build a custom meeting transcription, summarization, or action item extraction system for your business or internal workflow, email us at [contact@futuresmart.ai](mailto:contact@futuresmart.ai).

### Found something inaccurate or missing?

We try to keep our AI research accurate and useful. If you found outdated information, an issue, or have a suggestion, email us at [collaborate@aidemos.com](mailto:collaborate@aidemos.com).