Skip to content
How-To Guides

How to Use AI in Meetings: Agendas, Live Notes, Follow-Ups and Action Items (With Prompts)

An end-to-end guide to using AI in meetings: prepare agendas, take live notes, write follow-ups and track action items, with ready prompts and templates.

How to Use AI in Meetings: Agendas, Live Notes, Follow-Ups and Action Items (With Prompts)
On this page
  1. Key takeaways
  2. The AI meeting workflow at a glance
  3. Stage 1: prepare the meeting with AI
  4. Stage 2: take notes live with AI
  5. Stage 3: turn notes into follow-ups
  6. Stage 4: track action items until they are done
  7. More prompts for different meetings
  8. Best practices for AI in meetings
  9. Common mistakes
  10. Illustrative example: a weekly marketing meeting
  11. AI meeting checklist
  12. How Digital Tatva can help
  13. Frequently asked questions

Last updated: 20 November 2026

Quick answer: To use AI in meetings end to end, work in four stages. Before the meeting, ask a chatbot such as ChatGPT, Claude, Gemini or Copilot to draft an agenda, a pre-read and the decisions you need. During the meeting, turn on a note taker (Gemini “Take notes for me” in Google Meet, Copilot or Intelligent Recap in Microsoft Teams, Zoom AI Companion, or a third-party tool such as Otter, Fireflies, Fathom or tl;dv) after telling attendees, and ask it for catch-up summaries if you join late. After the meeting, have AI turn the notes into a recap email, a decision log and a task list with owners and dates. Then push action items into your project tool and ask AI for a weekly status check. Always review AI notes before sharing, because they can miss context or get names and numbers wrong.

Key takeaways

  • AI helps most before and after meetings, not just during them.
  • A clear agenda with decisions to make produces much better AI notes.
  • Tell participants before you record or transcribe, and follow your company policy.
  • Turn notes into owners, deadlines and a follow-up email within the hour.
  • Check AI output for errors, especially names, numbers and commitments.

The AI meeting workflow at a glance

Stage What AI does Tools
Before Agenda, pre-read summary, briefing on attendees and past decisions ChatGPT, Claude, Gemini, Microsoft Copilot
During Transcription, live notes, catch-up summaries, answering questions about what was said Gemini in Meet, Copilot in Teams, Zoom AI Companion, Otter, Fireflies, Fathom, tl;dv
After Summary, decisions, action items, follow-up email Built-in recaps or a chatbot with the transcript
Follow-through Tasks in project tools, reminders, weekly status checks Asana, ClickUp, monday.com, Jira, Notion, automation tools

Compare note-taking tools in best AI meeting assistants and free AI note takers.

Stage 1: prepare the meeting with AI

Decide whether you need a meeting at all

Before you send an invite, ask AI to help you decide whether an email, document or async video would work instead.

Prompt: “I want to discuss [topic] with [people]. The goal is [goal]. Should this be a meeting, an email or a shared document? If a meeting, how long should it be and who must attend?”

Draft an agenda with outcomes

AI note takers produce better summaries when the meeting has a clear structure. Ask for an agenda built around decisions, not just topics.

Prompt: “Create a 30-minute agenda for a meeting about [topic]. Attendees: [roles]. We need to decide [decisions]. Include timings, an owner for each item and the question each item must answer.”

Time Item Owner Outcome needed
0 to 5 min Context and goal Meeting lead Shared understanding of the problem
5 to 15 min Options review Project owner Pros and cons of each option
15 to 25 min Decision Decision maker Chosen option and reason
25 to 30 min Next steps Meeting lead Owners and deadlines

Summarise pre-reads

Upload documents, reports or long email threads to your AI assistant and ask for a short briefing.

Prompt: “Summarise these documents into a one-page pre-read: background, key numbers, open questions and the decision we need. Use plain language and bullet points.”

Built-in assistants such as Gemini in Google Workspace and Copilot in Microsoft 365 can draw on your files and emails directly; see how to use Gemini in Gmail and Sheets and Microsoft Copilot review.

Prepare for client or sales meetings

Prompt: “I am meeting [company] about [product]. Based on their public website, summarise what they do, likely priorities and five smart questions I should ask. Flag anything you are unsure about.”

Check facts yourself before the meeting; AI can be wrong about companies and people.

Stage 2: take notes live with AI

Choose your note taker

Meeting platform Built-in AI option Third-party options
Google Meet Gemini “Take notes for me” on eligible Workspace and Google AI plans; see our Google Meet guide Otter, Fireflies, Fathom, tl;dv, Read AI
Microsoft Teams Copilot, Facilitator and Intelligent Recap with the right licences; see our Teams Copilot guide Otter, Fireflies, Fathom, tl;dv
Zoom Zoom AI Companion on paid plans; see Zoom AI Companion Otter, Fireflies, Fathom, tl;dv, Read AI
In person Phone or laptop recording apps Otter, Granola and other apps that capture audio

Built-in tools are usually the simplest choice if your company already pays for them. Third-party tools work across platforms and often add search, CRM sync and analytics. For a head-to-head, see Otter vs Fireflies.

Set up the meeting properly

  1. Tell people in the invite that AI notes or a transcript will be used, and why.
  2. Announce it at the start and give people a chance to object.
  3. Set the meeting language if your tool supports more than one.
  4. Use good audio: headsets, one speaker at a time and names when speaking help accuracy.
  5. Say decisions out loud: “We have decided to…” and “Priya will do X by Friday” are easy for AI to capture.

For legal and etiquette details, read AI note taker privacy, consent and etiquette.

Use AI during the meeting

  • Catch up: “Summarise what has been discussed so far” if you join late.
  • Check alignment: “What decisions have we made so far, and what is still open?”
  • Find disagreements: “Where do participants disagree?”
  • Keep time: “Which agenda items have we not covered yet?”

Copilot in Teams and Gemini in Meet offer catch-up summaries in eligible plans; third-party tools vary.

Stage 3: turn notes into follow-ups

Clean up the summary

Review the AI notes as soon as the meeting ends. Fix names, numbers and anything said off the record. If your tool’s summary is weak, paste the transcript into a chatbot (only if your company allows it) and use a structured prompt.

Prompt: “From this transcript, produce: 1) a five-line summary, 2) decisions made with reasons, 3) action items with owner and due date, 4) open questions, 5) risks mentioned. If an owner or date was not stated, write ‘not stated’ rather than guessing.”

Meeting notes template

Section What to include
Meeting Title, date, attendees, absent
Purpose One sentence on why the meeting happened
Summary Three to five bullet points
Decisions Decision, reason, who decided
Action items Task, owner, due date, status
Open questions Question and who will answer it
Next meeting Date and focus

Write the follow-up email

Prompt: “Write a short, friendly follow-up email to attendees based on these notes. Start with thanks, then decisions, then action items in a table with owners and dates, then next steps. Keep it under 200 words.”

For client meetings, add: “Use a professional tone and do not include internal discussion points.”

Keep a decision log

Decisions get lost in long notes. Keep a simple shared log with date, decision, reason, owner and link to the notes. Ask AI each week: “List all decisions from this week’s meeting notes and flag any that conflict with earlier decisions.”

Stage 4: track action items until they are done

  1. Send tasks to your project tool. Many note takers and built-in recaps can create tasks, or you can use automation to send action items to Asana, ClickUp, monday.com, Jira, Trello or Notion. See how to automate tasks with Zapier and AI.
  2. Assign one owner per task with a clear due date.
  3. Set reminders a day or two before deadlines.
  4. Run a weekly AI status check: “From these meeting notes and the task list, which action items are overdue, which are at risk and what should we discuss next time?”
  5. Open the next meeting with open actions before new topics.

Project tools with built-in AI can help; see ClickUp Brain, monday.com AI and best AI project management tools.

More prompts for different meetings

Meeting type Prompt
Team stand-up “Summarise each person’s update as done, doing next and blockers. List blockers that need a decision.”
Client discovery “Extract the client’s goals, pain points, budget signals, timeline, decision makers and objections.”
Sales call “List buying signals, objections, competitors mentioned and agreed next steps. Draft a CRM note.”
Interview “Summarise the candidate’s answers against these criteria: [criteria]. Quote evidence and avoid judging personal characteristics.”
Project review “List what went well, what went wrong, root causes and three improvements with owners.”
Board or leadership “Summarise decisions, approvals, risks and follow-ups in formal minutes style.”
Brainstorm “Group the ideas into themes, remove duplicates and rank them by impact and effort.”
Training session “Turn this session into a one-page guide with key steps, tips and common mistakes.”

Best practices for AI in meetings

  • Be transparent: tell people when AI is recording, transcribing or summarising.
  • Respect sensitive meetings: turn off AI for HR, legal, medical or confidential conversations unless approved.
  • Review before sharing: AI can misattribute who said what.
  • Limit access: share notes only with people who need them.
  • Set retention: delete recordings and transcripts you no longer need.
  • Do not replace attention: AI notes are not a reason to multitask through meetings.
  • Use multilingual support carefully: many tools support only certain languages, and mixed-language meetings may not work well; translation tools such as DeepL can help afterwards.

Common mistakes

  1. No agenda: AI summaries of unstructured meetings are vague.
  2. Not announcing AI notes: this damages trust and may break laws or policies.
  3. Sending raw AI notes to clients: always edit first.
  4. Action items without owners: nothing happens.
  5. Too many tools: multiple bots joining one call confuses everyone.
  6. Ignoring data policies: pasting confidential transcripts into unapproved AI tools can leak data.

Illustrative example: a weekly marketing meeting

This is an illustrative scenario, not a client story.

A five-person marketing team meets every Monday on Google Meet. On Friday, the lead asks Gemini to draft an agenda from last week’s notes and the open task list. The invite says Gemini will take notes. During the meeting, people state decisions clearly and name owners. Afterwards, the lead reviews the Gemini notes, fixes a misheard campaign name and asks a chatbot to turn them into a follow-up email and a task list. Tasks are added to ClickUp with due dates. On Thursday, the lead asks AI which tasks are at risk and adds them to the top of next Monday’s agenda.

AI meeting checklist

  • Meeting needed and goal clear.
  • Agenda with decisions, timings and owners.
  • Pre-read summarised and shared.
  • Attendees told about AI notes in the invite and at the start.
  • Note taker set up with the right language and permissions.
  • Decisions and owners stated out loud.
  • Notes reviewed and corrected within an hour.
  • Follow-up email sent with action items.
  • Tasks added to the project tool with due dates.
  • Weekly AI status check on open actions.

How Digital Tatva can help

Digital Tatva helps businesses adopt AI practically, from internal workflows to making sure customers find them through AI search. See our AEO services and read what is AI search visibility.

Want customers to find you through AI assistants? Request an AEO audit or try the free AI visibility checker.

Frequently asked questions

How can I use AI in meetings?

Use AI to draft agendas and pre-reads, take notes and transcripts during the meeting, create summaries and follow-up emails afterwards and track action items in your project tools.

What is the best AI tool for meeting notes?

If your company uses Google Workspace, Microsoft 365 or Zoom, start with their built-in AI. Otherwise, tools like Otter, Fireflies, Fathom and tl;dv work across platforms.

Can AI write meeting minutes?

Yes. AI can draft minutes from a transcript, but a person should review them for accuracy, especially for formal or legal minutes.

Do I need to tell people I am using an AI note taker?

Yes, as good practice, and in many places it is legally required. Tell attendees in the invite and at the start of the meeting.

How do I get better AI meeting summaries?

Use a clear agenda, good audio, one speaker at a time, and state decisions and owners explicitly. Then use a structured prompt to refine the summary.

Can AI track action items from meetings?

Many tools extract action items and some can create tasks automatically. You can also use automation tools to send tasks to your project management app.

Can AI take notes in Google Meet?

Yes. Gemini’s “Take notes for me” works on eligible Google Workspace and Google AI plans.

Can AI take notes in Microsoft Teams?

Yes. Copilot, Facilitator and Intelligent Recap work with the right Microsoft licences, and third-party tools also support Teams.

Is it safe to paste meeting transcripts into ChatGPT?

Only if your company policy allows it and the plan you use protects business data. Remove sensitive details when in doubt.

Can AI help with in-person meetings?

Yes. Recording apps on phones or laptops can transcribe and summarise in-person meetings, with everyone’s knowledge and consent.

Product details, prices and features are checked at the time of writing and can change. Sources are linked in the text so you can check them yourself.

Written by

Ketan Parmar

Ketan Parmar has spent more than 15 years in digital marketing, helping brands grow through SEO, Google Ads, Meta Ads, content strategy and social media. Today he focuses on AI search visibility: how businesses get found and recommended in ChatGPT, Gemini, Perplexity and Google's AI answers.

How does your brand appear in AI answers?

We test the questions your buyers ask across the major AI assistants and search engines, then show you what to fix first.

Get an AI Visibility Audit

Your brand deserves to be in the answer.

We look at how search engines and AI assistants describe your business today, then build a practical plan to improve it.