{"observation":{"id":"21587b7d-dc6e-4aa1-8ee4-4e5f13641a3e","tool":"fellow","tool_name":"Fellow","criterion":"action-item-extraction","criterion_name":"Action-Item Extraction","criterion_definition":"Finds the real action items, ideally with owner and due date, with low false positives.","criterion_evidence_type":"transformation","criterion_rank_role":"decisive","criterion_rank_role_reason":"Finding the real action items is a central outcome readers are hiring the tool for, not just a convenience feature. (3 of 3 judges)","scenario":"ai-demos-daily-standup-31-july-2026","scenario_name":"AI Demos Daily Standup — 31 July 2026","group_tag":"ai-meeting-notetaker","scenario_description":"A real 25-minute technical engineering daily standup with 14 attendees and about 10 active speakers, used as the single parallel-capture meeting for evaluating AI meeting notetakers on transcription, diarization, summaries, action items, search/chat, and collaboration features.","modality":"image","input_text":null,"input_artifact_refs":[{"alt":null,"url":"https://cdn.futuresmart.ai/public/aidemos/547dd6f13e4a420fa8ad7bf2c88c7612.png?v=1","role":"input","filename":"31-july-meeting-screenshot.png"}],"stresses":["Transcription accuracy for real names, tool names, numbers, and technical jargon","Speaker diarization across multiple active speakers","Robustness to overlapping speech, crosstalk, and rapid turn-taking","Join reliability for bot-based and botless capture","Summary quality on identical source material","Action-item extraction with correct owners and commitments","Topic segmentation of standup updates","Search and chat grounded in the meeting content","Sharing, API, MCP, integrations, plan limits, languages, and privacy feature coverage"],"verdict":"mixed","score":null,"score_total":null,"note":"Action-item extraction was mostly correct, with real commitments and proper owner assignment for most items, but one real action item was misplaced from Mahreen to Anshika; the report states a 95%+ capture rate.","evidence_state":"verified","source":null,"artifacts":[{"url":"https://cdn.futuresmart.ai/public/aidemos/2f38a356a98d4395b43001c1d7f6c5b3.png?v=1","role":"output","alt":null}],"run_id":"ace58582-3d1e-48ee-996c-9b3cd03f27a2","study_title":"AI Meeting Notetakers — Capture Accurate Transcripts, Summaries & Action Items From Live Calls","study_kind":"generation","research_task":"86baxegnv","tested_at":null,"completeness":"input-and-output","input":{"state":"files","text":null,"files":[{"url":"https://cdn.futuresmart.ai/public/aidemos/547dd6f13e4a420fa8ad7bf2c88c7612.png?v=1","filename":"31-july-meeting-screenshot.png","alt":"AI Demos Daily Standup — 31 July 2026","role":"input"}],"modality":"image","stresses":["Transcription accuracy for real names, tool names, numbers, and technical jargon","Speaker diarization across multiple active speakers","Robustness to overlapping speech, crosstalk, and rapid turn-taking","Join reliability for bot-based and botless capture","Summary quality on identical source material","Action-item extraction with correct owners and commitments","Topic segmentation of standup updates","Search and chat grounded in the meeting content","Sharing, API, MCP, integrations, plan limits, languages, and privacy feature coverage"]},"tool_page_slug":"fellow","tool_url":"https://aidemos.com/tools/fellow","permalink":"https://aidemos.com/evidence/21587b7d-dc6e-4aa1-8ee4-4e5f13641a3e","api_url":"https://ai.aidemos.com/v1/observations/21587b7d-dc6e-4aa1-8ee4-4e5f13641a3e"},"peers":[{"id":"8340c79d-88de-4721-9c3e-c58e057071fb","tool":"fathom","tool_name":"Fathom","verdict":"worked","score":null,"score_total":null,"note":"Fathom extracts real commitments into an ACTION ITEMS section with owner attribution; the published output shows timestamped tasks and a named owner on the item.","artifact_count":2,"thumbnail":"https://cdn.futuresmart.ai/public/aidemos/9e8a3e7fb669448e959de254a7dcec48.png?v=1","evidence_url":"https://aidemos.com/evidence/8340c79d-88de-4721-9c3e-c58e057071fb"},{"id":"737cde44-da22-4dd6-b985-62104c4d14a8","tool":"fireflies-ai","tool_name":"Fireflies.ai","verdict":"worked","score":null,"score_total":null,"note":"Grouped action items by owner, attributed them to the correct team member, and exposed a clickable source timestamp (19:13) for at least one item.","artifact_count":1,"thumbnail":"https://cdn.futuresmart.ai/public/aidemos/bd68b01019254a5380cf234ec8ab89a1.png?v=1","evidence_url":"https://aidemos.com/evidence/737cde44-da22-4dd6-b985-62104c4d14a8"},{"id":"9af81f24-1ac8-4a6a-b543-1959eafc22b0","tool":"granola","tool_name":"Granola","verdict":"worked","score":null,"score_total":null,"note":"Granola extracts concrete next steps with ownership: the visible action item says to create a subtask and add details, names Mahreen Fathima as owner, and marks the item for same-day follow-up; the report says the extracted list contained no false positives.","artifact_count":2,"thumbnail":"https://cdn.futuresmart.ai/public/aidemos/0a21e41b32d648b1b5206dc034a58201.png?v=1","evidence_url":"https://aidemos.com/evidence/9af81f24-1ac8-4a6a-b543-1959eafc22b0"},{"id":"a3626a76-5657-440e-9460-f49e46ad880b","tool":"happyscribe","tool_name":"HappyScribe","verdict":"mixed","score":null,"score_total":null,"note":"It extracted the real action item about updating logs, but owner attribution was wrong because the misheard name cascaded into the action item and showed 'Nadine' instead of 'Mahreen.'","artifact_count":2,"thumbnail":"https://cdn.futuresmart.ai/public/aidemos/672943809b1d40c483bdee07651023c2.png?v=1","evidence_url":"https://aidemos.com/evidence/a3626a76-5657-440e-9460-f49e46ad880b"},{"id":"dfeacb85-baa0-4c8b-a44e-e4a7b005d54f","tool":"meetgeek","tool_name":"MeetGeek","verdict":"worked","score":null,"score_total":null,"note":"It extracts real commitments as action items rather than noise; the report says all extracted items had correct ownership and timing, and the visible note includes an owned action item with timestamp 19:51.","artifact_count":1,"thumbnail":"https://cdn.futuresmart.ai/public/aidemos/39691e7c2e9141b3aac5a412f3bb110d.png?v=1","evidence_url":"https://aidemos.com/evidence/dfeacb85-baa0-4c8b-a44e-e4a7b005d54f"},{"id":"1b0800ea-3b02-4159-88c1-ed437a0b12a7","tool":"notta","tool_name":"Notta","verdict":"worked","score":null,"score_total":null,"note":"The action-item list extracted the real commitments from the call, formatted them as checkbox items with @mentions, and the report says owner assignment was correct with no false positives.","artifact_count":1,"thumbnail":"https://cdn.futuresmart.ai/public/aidemos/22b60eda9b5e4b05b6ad287c471e5e3c.png?v=1","evidence_url":"https://aidemos.com/evidence/1b0800ea-3b02-4159-88c1-ed437a0b12a7"},{"id":"914e606f-d39b-4941-8b42-0f3b3653ee40","tool":"otter-ai","tool_name":"Otter.ai","verdict":"struggled","score":null,"score_total":null,"note":"Otter extracted action items, including at least one due-today API-related task with an assignee, but the report says most items were left without an owner and duplicate entries also appeared, so the output needed manual cleanup before delegation.","artifact_count":2,"thumbnail":"https://cdn.futuresmart.ai/public/aidemos/7e0a51fd2a4744aea043309aec25fc20.png?v=1","evidence_url":"https://aidemos.com/evidence/914e606f-d39b-4941-8b42-0f3b3653ee40"}],"other_criteria":[{"id":"b8fe3f78-ae35-4ccf-b887-6188bac5ad2f","criterion":"chat-with-notes-ask-questions","criterion_name":"Chat with Notes / Ask Questions","rank_role":"context","verdict":"worked","score":null,"score_total":null,"note":"Ask Fellow returned grounded answers to natural-language questions against the meeting notes, and the tested query produced a cited response rather than an unsupported hallucination.","artifact_count":1,"evidence_url":"https://aidemos.com/evidence/b8fe3f78-ae35-4ccf-b887-6188bac5ad2f"},{"id":"a47ae17f-dfbd-439c-9ba9-e3153156cca8","criterion":"editability","criterion_name":"Editability","rank_role":"context","verdict":"mixed","score":null,"score_total":null,"note":"Summary and action items are editable inline before sharing, but the transcript itself is locked for audit-trail purposes, so editing is only partial.","artifact_count":1,"evidence_url":"https://aidemos.com/evidence/a47ae17f-dfbd-439c-9ba9-e3153156cca8"},{"id":"5d9e1f80-f657-4216-8e73-e81047b806aa","criterion":"join-method-reliability","criterion_name":"Join Method & Reliability","rank_role":"decisive","verdict":"worked","score":null,"score_total":null,"note":"The bot auto-joined via Google Calendar integration and stayed connected for the full meeting capture without drops or disconnections, including the late closing portion of the call.","artifact_count":3,"evidence_url":"https://aidemos.com/evidence/5d9e1f80-f657-4216-8e73-e81047b806aa"},{"id":"a8a8aaba-e0e1-44b8-9cce-4fa91c5e5213","criterion":"search-across-notes","criterion_name":"Search Across Notes","rank_role":"context","verdict":"worked","score":null,"score_total":null,"note":"Transcript search works with exact timestamp retrieval for matching terms, letting users jump to precise moments within the meeting; the report notes cross-meeting search was untested.","artifact_count":2,"evidence_url":"https://aidemos.com/evidence/a8a8aaba-e0e1-44b8-9cce-4fa91c5e5213"},{"id":"b22ab795-4b0c-4939-80f6-1f9daca9f6e6","criterion":"sharing-without-registration","criterion_name":"Sharing Without Registration","rank_role":"context","verdict":"worked","score":null,"score_total":null,"note":"Shared recap links can be opened by anyone with the link without creating a Fellow account, and the share modal also offers optional password protection for viewers outside the workspace.","artifact_count":2,"evidence_url":"https://aidemos.com/evidence/b22ab795-4b0c-4939-80f6-1f9daca9f6e6"},{"id":"5928b605-59f7-4372-a4b7-914060123f74","criterion":"speaker-diarization","criterion_name":"Speaker Diarization","rank_role":"decisive","verdict":"worked","score":null,"score_total":null,"note":"The transcript attributed speaker turns correctly across the standup, with all ~10 speakers labeled by name and no attribution errors or generic labels reported.","artifact_count":2,"evidence_url":"https://aidemos.com/evidence/5928b605-59f7-4372-a4b7-914060123f74"},{"id":"d89239cd-ae18-4967-90d8-e968276badf1","criterion":"summary-quality","criterion_name":"Summary Quality","rank_role":"decisive","verdict":"worked","score":null,"score_total":null,"note":"The meeting recap was reported as clearly structured and complete, with the key decisions and discussion points preserved and nothing important dropped.","artifact_count":1,"evidence_url":"https://aidemos.com/evidence/d89239cd-ae18-4967-90d8-e968276badf1"},{"id":"b69458f2-5b1d-4915-bf53-39ab59f059d2","criterion":"topic-segmentation","criterion_name":"Topic Segmentation","rank_role":"context","verdict":"worked","score":null,"score_total":null,"note":"The tool broke the standup into logical topic sections with clear headers and separated discussion points, rather than leaving the meeting as one undifferentiated blob.","artifact_count":1,"evidence_url":"https://aidemos.com/evidence/b69458f2-5b1d-4915-bf53-39ab59f059d2"},{"id":"d4b52ad2-b536-4580-99fa-644f4b772a93","criterion":"transcription-accuracy","criterion_name":"Transcription Accuracy","rank_role":"decisive","verdict":"worked","score":null,"score_total":null,"note":"The transcript was near-clean: the tool captured nearly all names, technical jargon, and numbers correctly, with no significant misheard terms or hallucinations observed in the tested meeting.","artifact_count":2,"evidence_url":"https://aidemos.com/evidence/d4b52ad2-b536-4580-99fa-644f4b772a93"}],"appears_in":[{"page_type":"ranking","slug":"ai-meeting-notetakers","title":"Best AI Meeting Notetakers for Accurate Transcripts, Summaries, and Action Items","url":"https://aidemos.com/best/ai-meeting-notetakers","binding":"run"}],"same_scenario":[{"id":"8340c79d-88de-4721-9c3e-c58e057071fb","tool":"fathom","tool_name":"Fathom","verdict":"worked","score":null,"score_total":null,"note":"Fathom extracts real commitments into an ACTION ITEMS section with owner attribution; the published output shows timestamped tasks and a named owner on the item."},{"id":"737cde44-da22-4dd6-b985-62104c4d14a8","tool":"fireflies-ai","tool_name":"Fireflies.ai","verdict":"worked","score":null,"score_total":null,"note":"Grouped action items by owner, attributed them to the correct team member, and exposed a clickable source timestamp (19:13) for at least one item."},{"id":"9af81f24-1ac8-4a6a-b543-1959eafc22b0","tool":"granola","tool_name":"Granola","verdict":"worked","score":null,"score_total":null,"note":"Granola extracts concrete next steps with ownership: the visible action item says to create a subtask and add details, names Mahreen Fathima as owner, and marks the item for same-day follow-up; the report says the extracted list contained no false positives."},{"id":"a3626a76-5657-440e-9460-f49e46ad880b","tool":"happyscribe","tool_name":"HappyScribe","verdict":"mixed","score":null,"score_total":null,"note":"It extracted the real action item about updating logs, but owner attribution was wrong because the misheard name cascaded into the action item and showed 'Nadine' instead of 'Mahreen.'"},{"id":"dfeacb85-baa0-4c8b-a44e-e4a7b005d54f","tool":"meetgeek","tool_name":"MeetGeek","verdict":"worked","score":null,"score_total":null,"note":"It extracts real commitments as action items rather than noise; the report says all extracted items had correct ownership and timing, and the visible note includes an owned action item with timestamp 19:51."},{"id":"1b0800ea-3b02-4159-88c1-ed437a0b12a7","tool":"notta","tool_name":"Notta","verdict":"worked","score":null,"score_total":null,"note":"The action-item list extracted the real commitments from the call, formatted them as checkbox items with @mentions, and the report says owner assignment was correct with no false positives."},{"id":"914e606f-d39b-4941-8b42-0f3b3653ee40","tool":"otter-ai","tool_name":"Otter.ai","verdict":"struggled","score":null,"score_total":null,"note":"Otter extracted action items, including at least one due-today API-related task with an assignee, but the report says most items were left without an owner and duplicate entries also appeared, so the output needed manual cleanup before delegation."}]}