productivity

Fathom

Name: Fathom
Availability: InStock
Author: AI Demos

Reliable meeting notes, search, and sharing from live calls, with a caveat on speaker attribution.

Visit Fathom

Unlimited freeAPI + MCP38 languagesGoogle Meet bot

TL;DR — our verdictUpdated August 2026 · 23 test artifacts

Strong meeting memory, with speaker-attribution checks

Where it wins

You want a searchable, shareable meeting record with transcript, summary, and action items.
You want a true free tier with unlimited recordings/transcriptions and a usable API.
You want API or MCP access for downstream automation or agent queries.

Main limitation

You need perfectly reliable speaker attribution in fast-turn multi-speaker calls.

Pricing (verified plans)

Free $0 Free forever.Premium $16 /month / per userTeam $15 /month / per user

Strongest test artifacts

OUTPUT →Meeting Summaries →Action-Item Extraction →

Our take

Fathom came through as a production-ready meeting assistant in this test: it captured a usable transcript, generated a structured summary and real action items, and made the notes searchable, shareable, and accessible through both API and MCP. The main caveat is attribution — a few speaker/name labels were wrong, so fast-turn and named-person outputs should be checked before acting on them.

Screen recording showing Fathom setup/customization.

In-Depth Review

Our detailed analysis of Fathom — features, performance, and real-world testing.

AI Demos Team

Expert Reviewer

Verified Review

Feature-by-Feature Breakdown

Automated Meeting Capture

▾

Test Summary

Feature tested: Automated Meeting Capture

Result: Partial

Feature tested: Automated Meeting Capture

Result: Partial

Expected behavior: Fathom can join a live Google Meet as a bot and keep recording through the call; the tested standup stayed connected for the full meeting.

Test case: Image → Image

Input type: Image

Input used: Input artifact (Image): 31 July 2026 AI Demos daily standup on Google Meet, about 25 minutes long, with one Fathom bot joining alongside other notetakers. — 31-july-meeting-screenshot.png

Observed output: Output artifact (Image): The bot/note-taker appears to have joined the meeting successfully. — fathom-bot-meeting-joined-success.png

Input artifact: Input artifact (Image): 31 July 2026 AI Demos daily standup on Google Meet, about 25 minutes long, with one Fathom bot joining alongside other notetakers. — 31-july-meeting-screenshot.png

Output artifact: Output artifact (Image): The bot/note-taker appears to have joined the meeting successfully. — fathom-bot-meeting-joined-success.png

What changed: Image transformed into Image

Test case: Image → Image

Input type: Image

Observed output: Output artifact (Image): The recording remained available near the end of playback, corroborating full-session capture without a mid-call drop. — fathom-bot-till-end-of-meeting-join-reliability-success.png

Output artifact: Output artifact (Image): The recording remained available near the end of playback, corroborating full-session capture without a mid-call drop. — fathom-bot-till-end-of-meeting-join-reliability-success.png

What changed: Image transformed into Image

Test case: Video file → Image

Input type: Video file

Input used: Input artifact (Video file): 31 July 2026 AI Demos daily standup in Google Meet (~25 min, ~10 active speakers, normal audio). — 31-july-meeting-artifact.webm

Observed output: Output artifact (Image): Fathom generated a readable transcript for the tested standup; most of the content was captured correctly, but one visible name error rendered "Mahreen" as "Meryl." — fathom-meeting-transcript-generated-success.png

Input artifact: Input artifact (Video file): 31 July 2026 AI Demos daily standup in Google Meet (~25 min, ~10 active speakers, normal audio). — 31-july-meeting-artifact.webm

Output artifact: Output artifact (Image): Fathom generated a readable transcript for the tested standup; most of the content was captured correctly, but one visible name error rendered "Mahreen" as "Meryl." — fathom-meeting-transcript-generated-success.png

What changed: Video file transformed into Image

Why it matters / Conclusion: Join reliability looked solid in this normal standup.

Fathom can join a live Google Meet as a bot and keep recording through the call; the tested standup stayed connected for the full meeting.

audio

Input artifact for "Automated Meeting Capture" test: 31 July 2026 AI Demos daily standup on Google Meet, about 25 minutes long, with one Fathom bot joining alongside other notetakers., 31-july-meeting-screenshot.png

31 July 2026 AI Demos daily standup on Google Meet, about 25 minutes long, with one Fathom bot joining alongside other notetakers.

↓→

image

The bot/note-taker appears to have joined the meeting successfully.

audio

31 July 2026 AI Demos daily standup on Google Meet, about 25 minutes long, with one Fathom bot joining alongside other notetakers.

↓→

image

Output artifact for "Automated Meeting Capture" test: The recording remained available near the end of playback, corroborating full-session capture without a mid-call drop., fathom-bot-till-end-of-meeting-join-reliability-success.png

The recording remained available near the end of playback, corroborating full-session capture without a mid-call drop.

audio

31 July 2026 AI Demos daily standup in Google Meet (~25 min, ~10 active speakers, normal audio).

↓→

image

Output artifact for "Automated Meeting Capture" test: Fathom generated a readable transcript for the tested standup; most of the content was captured correctly, but one visible name error rendered "Mahreen" as "Meryl.", fathom-meeting-transcript-generated-success.png

Fathom generated a readable transcript for the tested standup; most of the content was captured correctly, but one visible name error rendered "Mahreen" as "Meryl."

Bottom Line

Join reliability looked solid in this normal standup.

From our researchAI Meeting Notetakers — Capture Accurate Transcripts, Summaries & Action Items From Live Callsearlier research

Speech-to-Text Transcription

▾

Test Summary

Feature tested: Speech-to-Text Transcription

Result: Partial

Fathom produces a readable meeting transcript for the tested standup, capturing most names, jargon, and numbers correctly while still allowing some name-level errors.

audio

31 July 2026 standup with typical engineering jargon, names, and numbers.

↓→

image

Output artifact for "Speech-to-Text Transcription" test: Fathom generated a readable transcript for the standup, with the transcript pane showing a coherent conversation thread., fathom-meeting-transcript-generated-success.png

Fathom generated a readable transcript for the standup, with the transcript pane showing a coherent conversation thread.

Bottom Line

Mostly accurate, but name-level corrections are still needed.

From our researchAI Meeting Notetakers — Capture Accurate Transcripts, Summaries & Action Items From Live Calls

Speaker Attribution

▾

Test Summary

Feature tested: Speaker Attribution

Result: Partial

Fathom labels speakers in multi-speaker transcripts, mostly attributing turns correctly even when fast transitions can merge or mislabel turns.

audio

31 July standup with ~10 active speakers and frequent speaker transitions.

↓→

image

Output artifact for "Speaker Attribution" test: Fathom applied speaker labels across the transcript and handled most turns correctly on the multi-speaker standup., fathom-speaker-diariazation-success.png

Fathom applied speaker labels across the transcript and handled most turns correctly on the multi-speaker standup.

Bottom Line

Good overall, with a real attribution risk in fast-turn segments.

From our researchAI Meeting Notetakers — Capture Accurate Transcripts, Summaries & Action Items From Live Callsearlier research

Meeting Summaries

▾

Test Summary

Feature tested: Meeting Summaries

Result: Passed

Feature tested: Meeting Summaries

Result: Passed

Expected behavior: Fathom turns the meeting into a structured, scannable summary with purpose, key takeaways, and named topic sections that mirror the meeting flow.

Test case: Video file → Image

Input type: Video file

Input used: Input artifact (Video file): 31 July 2026 standup with multiple discussion threads and action points. — 31-july-meeting-artifact.webm

Observed output: Output artifact (Image): Fathom produced a structured summary with a meeting purpose and key takeaways, showing that it can create a skimmable recap after the call. — fathom-meeting-summary-generated-success.png

Input artifact: Input artifact (Video file): 31 July 2026 standup with multiple discussion threads and action points. — 31-july-meeting-artifact.webm

Output artifact: Output artifact (Image): Fathom produced a structured summary with a meeting purpose and key takeaways, showing that it can create a skimmable recap after the call. — fathom-meeting-summary-generated-success.png

What changed: Video file transformed into Image

Test case: Video file → Image

Input type: Video file

Input used: Input artifact (Video file): Same standup, checked for topic breaks in the generated notes. — 31-july-meeting-artifact.webm

Observed output: Output artifact (Image): The summary is segmented into named topics rather than one undifferentiated block, which makes the output easier to scan and navigate. — fathom-topic-segmentation-success.png

Input artifact: Input artifact (Video file): Same standup, checked for topic breaks in the generated notes. — 31-july-meeting-artifact.webm

Output artifact: Output artifact (Image): The summary is segmented into named topics rather than one undifferentiated block, which makes the output easier to scan and navigate. — fathom-topic-segmentation-success.png

What changed: Video file transformed into Image

Test case: Image → Image

Input type: Image

Input used: Input artifact (Image): INPUT: Same standup, checked for the detailed topic structure inside the summary. — 31-july-meeting-screenshot.png

Observed output: Output artifact (Image): Topic section snippet showing structured subheadings and bullet structure. — fathom-meeting-summary-topics-section-snippet.png

Input artifact: Input artifact (Image): INPUT: Same standup, checked for the detailed topic structure inside the summary. — 31-july-meeting-screenshot.png

Output artifact: Output artifact (Image): Topic section snippet showing structured subheadings and bullet structure. — fathom-meeting-summary-topics-section-snippet.png

What changed: Image transformed into Image

Why it matters / Conclusion: Clear, structured, and useful for quick review.

Fathom turns the meeting into a structured, scannable summary with purpose, key takeaways, and named topic sections that mirror the meeting flow.

audio

31 July 2026 standup with multiple discussion threads and action points.

↓→

image

Output artifact for "Meeting Summaries" test: Fathom produced a structured summary with a meeting purpose and key takeaways, showing that it can create a skimmable recap after the call., fathom-meeting-summary-generated-success.png

Fathom produced a structured summary with a meeting purpose and key takeaways, showing that it can create a skimmable recap after the call.

audio

Same standup, checked for topic breaks in the generated notes.

↓→

image

Output artifact for "Meeting Summaries" test: The summary is segmented into named topics rather than one undifferentiated block, which makes the output easier to scan and navigate., fathom-topic-segmentation-success.png

The summary is segmented into named topics rather than one undifferentiated block, which makes the output easier to scan and navigate.

audio

INPUT: Same standup, checked for the detailed topic structure inside the summary.

↓→

OUTPUT

Topic section snippet showing structured subheadings and bullet structure.

Bottom Line

Clear, structured, and useful for quick review.

From our researchAI Meeting Notetakers — Capture Accurate Transcripts, Summaries & Action Items From Live Callsearlier research

Action-Item Extraction

▾

Test Summary

Feature tested: Action-Item Extraction

Result: Passed

Fathom extracts commitments from the meeting and attaches owners to them, including action-item lists pulled from the tested call.

audio

INPUT: 25-minute standup with real commitments made by named participants.

↓→

OUTPUT

Single extracted action item with a checkbox-style task entry.

Bottom Line

Captured the real commitments without obvious false positives.

From our researchAI Meeting Notetakers — Capture Accurate Transcripts, Summaries & Action Items From Live Callsearlier research

Editable Meeting Outputs

▾

Test Summary

Feature tested: Editable Meeting Outputs

Result: Passed

Fathom lets users edit extracted action items inline and refine summaries with instructions before regenerating the output.

text

Post-meeting Fathom outputs for action items and summary.

↓→

image

An action item row shows an edit pencil, indicating the extracted task can be edited inline.

text

Post-meeting Fathom outputs for action items and summary.

↓→

image

The same action item is shown in edit mode with the caret active inside the text.

text

Post-meeting Fathom outputs for action items and summary.

↓→

image

The summary can be customized by giving the AI instructions and regenerating the template.

Bottom Line

Outputs are correctable before sharing.

From our researchAI Meeting Notetakers — Capture Accurate Transcripts, Summaries & Action Items From Live Callsearlier research

Transcript Search

▾

Test Summary

Feature tested: Transcript Search

Result: Passed

Fathom supports keyword search across the transcript, returning timestamped matches and surfacing relevant moments; one combined card also paired this with grounded meeting Q&A.

text

Search the meeting transcript with keywords

↓→

image

Output artifact for "Transcript Search" test: The transcript tab shows search controls and a searchable meeting transcript, confirming in-call lookup is available., fathom-meeting-transcript-search-success.png

The transcript tab shows search controls and a searchable meeting transcript, confirming in-call lookup is available.

Bottom Line

Fast, precise lookup on the tested meeting.

From our researchAI Meeting Notetakers — Capture Accurate Transcripts, Summaries & Action Items From Live Callsearlier research

Meeting Q&A Assistant

▾

Test Summary

Feature tested: Meeting Q&A Assistant

Result: Passed

Fathom answers plain-language questions about the meeting using grounded references to the transcript, including schedule and action-item questions.

text

Ask Fathom: for what date was a call scheduled related to a person's work?

↓→

image

Output artifact for "Meeting Q&A Assistant" test: Ask Fathom answers that the call was scheduled for 6th August and cites the related meeting mention., fathom-chat-query-and-answer-success.png

Ask Fathom answers that the call was scheduled for 6th August and cites the related meeting mention.

text

Ask Fathom: what were the action items for the meeting?

↓→

image

Output artifact for "Meeting Q&A Assistant" test: The Ask Fathom pane shows a successful question-and-answer interaction about meeting action items., fathom-meeting-chat-query-success.png

The Ask Fathom pane shows a successful question-and-answer interaction about meeting action items.

Bottom Line

Good for post-call questions on the tested meeting.

From our researchAI Meeting Notetakers — Capture Accurate Transcripts, Summaries & Action Items From Live Calls

Shareable Notes Links

▾

Test Summary

Feature tested: Shareable Notes Links

Result: Passed

Fathom generates shareable links so published notes can be opened without requiring a Fathom account.

text

Open a shared Fathom recording link as a teammate who does not have an account.

↓→

image

The share modal shows link-based access controls, including 'Anyone with the link can view'.

Bottom Line

External viewers can open shared content without a signup wall.

From our researchAI Meeting Notetakers — Capture Accurate Transcripts, Summaries & Action Items From Live Callsearlier research

Meeting Data API

▾

Test Summary

Feature tested: Meeting Data API

Result: Passed

Fathom exposes REST API endpoints that returned successful responses for summary, transcript, and meeting-listing data in the test script.

text

Review Fathom REST API docs for authentication and rate limits.

↓→

image

The API docs describe Fathom's REST API, including API-key authentication and rate limiting.

text

Call the Fathom REST API summary endpoint with an API key and print the response.

↓→

image

A Python script and terminal output show a successful Fathom API summary response.

INPUT

Inspect the API access screen for key generation or OAuth app setup.

↓→

image

Output artifact for "Meeting Data API" test: The API access screen shows the option to generate an API key or build an OAuth app, confirming that programmatic access is intentionally supported., fathom-api-docs-2.png

The API access screen shows the option to generate an API key or build an OAuth app, confirming that programmatic access is intentionally supported.

Bottom Line

Programmatic access worked in the test.

From our researchAI Meeting Notetakers — Capture Accurate Transcripts, Summaries & Action Items From Live Callsearlier research

Claude MCP Integration

▾

Test Summary

Feature tested: Claude MCP Integration

Result: Partial

Fathom connects to Claude through MCP so meeting-specific questions can be asked from that environment, though one named-person query was misattributed in testing.

text

Connect Fathom to Claude and ask about a meeting-specific question through MCP.

↓→

image

Claude shows the Fathom Search Meetings tool as connected and ready for approval.

Bottom Line

Useful integration, but named-person answers need verification.

From our researchAI Meeting Notetakers — Capture Accurate Transcripts, Summaries & Action Items From Live Callsearlier research

Recording Mode Controls

▾

Test Summary

Feature tested: Recording Mode Controls

Result: Partial

Fathom provides capture-mode settings such as all meetings, external meetings, internal meetings, or manual recording, and it also notes a botless option in beta.

text

Review Fathom's meeting recording settings and capture-mode options.

↓→

image

Output artifact for "Recording Mode Controls" test: The settings menu shows capture-mode choices including all meetings, external meetings, internal meetings, and manual recording., fathom-meeting-options.png

The settings menu shows capture-mode choices including all meetings, external meetings, internal meetings, and manual recording.

Bottom Line

Good control over how the bot joins, though botless is still beta.

From our researchAI Meeting Notetakers — Capture Accurate Transcripts, Summaries & Action Items From Live Calls

Native App Integrations

▾

Test Summary

Feature tested: Native App Integrations

Result: Passed

Fathom’s integrations gallery includes major tools and meeting platforms such as Claude, ChatGPT, HubSpot, Asana, Slack, and Salesforce.

text

Review Fathom's integrations gallery for supported apps and platforms.

↓→

image

Output artifact for "Native App Integrations" test: The integrations gallery lists apps such as Claude, ChatGPT, HubSpot, Asana, Slack, Salesforce, Google Meet, Zoom, and Microsoft Teams., fathom-integrations-available.png

The integrations gallery lists apps such as Claude, ChatGPT, HubSpot, Asana, Slack, Salesforce, Google Meet, Zoom, and Microsoft Teams.

Bottom Line

Breadth looks strong across collaboration, CRM, and meeting platforms.

From our researchAI Meeting Notetakers — Capture Accurate Transcripts, Summaries & Action Items From Live Calls

Multilingual Transcription

▾

Test Summary

Feature tested: Multilingual Transcription

Result: Passed

Fathom documents transcript support for 38 languages; the evaluation itself only exercised English, so the multilingual claim is documentation-backed here.

text

Check Fathom's help documentation for supported transcript languages.

↓→

image

The help article states that Fathom currently supports transcripts in 38 languages.

Bottom Line

Broad language coverage on paper, though only English was exercised here.

From our researchAI Meeting Notetakers — Capture Accurate Transcripts, Summaries & Action Items From Live Callsearlier research

Pricing to supercharge every meeting

Free

$0 Free forever.

Unlimited recordings, instant call summaries, integrations with LLMs & more.

Premium

$16 /month / per user

Everything from Free; advanced call summaries, AI-generated action items, conversational meeting assistant.

Team

$15 /month / per user (2 user min)

Everything from Premium; global search across calls, playlists of highlights from meetings, collaboration using comments, folders, keyword alerts.

Annual billing saves 25%+; paid plans show a 90-day guarantee.

✓ Use This If

●You want a searchable, shareable meeting record with transcript, summary, and action items.

●You want a true free tier with unlimited recordings/transcriptions and a usable API.

●You want API or MCP access for downstream automation or agent queries.

✕ Skip This If

●You need perfectly reliable speaker attribution in fast-turn multi-speaker calls.

●You need production-ready botless capture today rather than a beta/manual option.

●You need named-person MCP answers to be safe without verification.

productivityplanning-toolsspeechFounderOther

It was mostly accurate and readable for the standup, with most jargon, tools, and numbers captured correctly. The clearest confirmed error was a name-level miss: 'Mahreen' appeared as 'Meryl' in one transcript snippet.

Mostly yes, but not perfectly. The transcript handled the multi-speaker standup well overall, yet one rapid transition was merged into a single speaker block and one transcript line showed an attribution issue.

Yes. The test produced a structured summary with Meeting Purpose, Key Takeaways, and Topics, plus an action-items list with owner attribution.

Yes. A keyword search returned a matching transcript snippet with an exact timestamp and surfaced the related action item.

Yes. Ask Fathom returned grounded answers that referenced the meeting content rather than generic model memory.

Yes. The share modal shows link-based access and the evaluation confirmed recipients did not hit a signup wall when opening shared content.

Yes. The report successfully tested the REST API, and it also connected Fathom to Claude through MCP. The caution is that one named-person MCP answer misattributed ownership, so verification is still wise.

The pricing screenshots show a Free plan at $0 forever, Premium at $16 per user per month, and Team at $15 per user per month with a 2-user minimum. The free plan includes unlimited recordings and instant call summaries.

The help article says transcripts are supported in 38 languages. English was the only language directly tested in this evaluation.

Yes. Action items can be edited inline, and summaries can be customized by giving the AI instructions and regenerating the template.

Banner Preview

How the embed badge will look on your site

Embed HTML

Copy this code to your website source

Quick Integration Guide

1Copy the HTML code block above.
2Paste it into your site's HTML or CMS editor.
3Banner appears instantly on your page.
4Links back to your tool profile here.

Similar Tools

Discover more AI tools like Fathom to enhance your workflow.

Bot-based meeting capture with strong transcript quality, summaries, search, AskFred, and MCP.

AI Tool

🤖

Fellow

Accurate, searchable meeting notes from live calls — but the useful plans are paid.

AI Tool

🤖

Granola

Privacy-first botless meeting notes with strong search, sharing, API/MCP access, and action-item workflows — but unreliable transcript accuracy on standups.

AI Tool

🤖

Notta

Reliable live-meeting capture, summaries, and action items for teams that can live with plan limits and no public API.

AI Tool

🤖

MeetGeek

Reliable standup capture with strong summaries, search, API, and MCP — but speaker attribution and chat need review.

AI Tool

🤖

HappyScribe

Strong live meeting capture, searchable notes, and API/MCP access, but attribution errors still need a human check.

AI Tool

🤖

Otter.ai

Otter.ai captures meetings cleanly and answers questions later, but speaker attribution and action-item ownership break down on active standups.

Comments (0)

Built by FutureSmart AI — the team behind AI Demos

Need a custom AI solution for this use case?

If you are looking to build a custom meeting transcription, call summarization, or meeting notes assistant for your business or internal workflow, email us at contact@futuresmart.ai.

Get a custom build

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.

Fathom

Our take

In-Depth Review

Feature-by-Feature Breakdown

Pricing to supercharge every meeting

Promote Fathom

Banner Preview

Embed HTML

Quick Integration Guide

Similar Tools

Comments (0)

Need a custom AI solution for this use case?