by admin | Jul 13, 2026 | AI Tools
You just left a long online meeting, and now you are staring at a few messy lines you typed while half-listening. Who agreed to do what? What was that deadline? If you have ever tried to write notes and follow the conversation at the same time, you know it rarely works well.
This is where AI meeting assistants come in. These tools can join your call, write down every word, and hand you a clean summary with the action items already pulled out. In this guide I will explain what they are, which ones you may already have, how to use them, and the privacy side that most people forget to check.
What is an AI meeting assistant?
An AI meeting assistant is a tool that listens to your meeting, turns the speech into text, and then writes a short summary for you. Most of them give you three things: a full transcript of what was said, a summary of the main points, and a list of action items or next steps.
There are two main types. Some are built right into the meeting apps you already use, like Google Meet, Microsoft Teams, and Zoom. Others are separate tools that send a small bot into your call to take notes. Both do a similar job, so the right choice usually depends on which apps your team already lives in.
The AI note-takers you may already have
Before you sign up for anything new, check the apps you use every week. The big three all have their own note-taker built in.
- Google Meet has a feature called Take notes for me. Powered by Gemini, it captures the meeting in a Google Doc and adds a summary and action items. It is available on Google AI Pro and Ultra plans and eligible Workspace accounts.
- Microsoft Teams uses Copilot and its intelligent recap to summarize who said what and suggest action items. You need live transcription turned on, plus a Teams Premium or Microsoft 365 Copilot license.
- Zoom has AI Companion, which creates a meeting summary with key points, decisions, and next steps a few minutes after the call ends. An account admin needs to switch it on first.
Standalone tools like Otter
If your team jumps between different meeting apps, a standalone assistant can be easier. Otter is one of the best known. You connect your calendar, and its bot joins your Zoom, Google Meet, or Teams calls automatically. It writes a live transcript, builds a summary, and even lets you ask questions like “what were my action items?” after the call.
Otter is not the only one. Tools like Fireflies and Fathom do similar work, and several offer a free tier for light use. Free limits change often, so check the current plan on the tool’s own site before you rely on it. For a wider look at everyday AI tools, see our guide to useful AI tools for daily work and study.
How to use an AI meeting assistant
The setup is usually simple. Here is the basic flow that works for almost any tool.
- Turn on the note-taker before the meeting, or connect your calendar so it joins on its own.
- Tell everyone on the call that it is being recorded and transcribed.
- Run the meeting as normal and let the tool listen in the background.
- After the call, open the summary, read it, and fix anything the AI got wrong.
- Share the clean notes and assign the action items.
Where these tools really help
From my own experience running websites and online projects, the transcript is nice, but the real time saver is the action-item list. Instead of rewatching a call to find one decision, I get a short list I can drop straight into my task app. If you want to build on that, our guide on using AI for time management shows how to turn those items into a real plan.
Students get a lot out of these tools too. You can record a lecture (with permission), get a full transcript, then drop it into a tool like NotebookLM to turn it into a study guide or a quick quiz.
The privacy part people skip
Here is the part I always slow down on, because it touches privacy and security. An AI meeting assistant records and stores everything that is said, and sometimes that includes sensitive information. A few simple habits keep you safe.
First, always tell people they are being recorded. In many places you are legally required to get consent before recording a conversation, so do not treat it as optional. Second, avoid recording calls where people share private or confidential details unless everyone agrees. Third, check what the tool does with your data. Zoom, for example, says it does not use your meeting content to train its AI models, while Microsoft says Teams recap data is stored based on your organization’s own admin policy.
Quick tip: Treat every AI meeting summary as a first draft, not the final record. The AI can mishear names, numbers, and decisions, so read it once and correct it before you send it to anyone.
If you want a fuller checklist, our guide on how to use AI safely and protect your privacy covers what to share and what to keep out of any AI tool.
Common questions about AI meeting assistants
Are AI meeting assistants free?
Some standalone tools offer a free tier for a small number of meetings, while the built-in features in Google Meet, Teams, and Zoom usually need a paid plan. Because free limits and prices change often, check the current plan on each tool’s official site.
Do AI meeting notes make mistakes?
Yes. AI transcription is good but not perfect. It can get names, numbers, and technical words wrong, and it can misread who said what. Always review the summary before you treat it as an official record.
Is it legal to record a meeting with an AI note-taker?
It depends on where you and the other people are. Many regions require you to tell participants and get their consent before recording. The safe habit is to always announce that the meeting is being recorded and let people opt out. This is general information, not legal advice, so check your local rules for important calls.
Final takeaway
AI meeting assistants can save you real time by handling the boring part of meetings, so you can actually pay attention. Start with whatever is already built into Google Meet, Teams, or Zoom, or try a free standalone tool like Otter. Just remember the two rules that matter most: tell people they are being recorded, and always check the summary before you trust it. Do that, and you get the time back without the headaches.
by admin | Jul 12, 2026 | AI Guides
Ever asked ChatGPT a long, messy question and been a little surprised at how well it understood what you actually meant? That’s not magic. It’s a specific piece of engineering called the transformer, and it’s the reason today’s AI chatbots feel so much more capable than the clunky “AI” tools from just a few years ago.
You don’t need a computer science degree to understand it. This guide breaks the transformer down in plain English, using the same kind of everyday examples I’d use explaining it to a friend, not a textbook.
What is a transformer, in plain English?
A transformer is a type of neural network design that reads a piece of text (or image, or audio) and works out how every word or piece relates to every other word or piece, all at the same time. It’s the architecture that powers ChatGPT, Gemini, Claude, and pretty much every well known AI tool built after 2018.
Before transformers, AI language models read text one word at a time, left to right, a bit like reading with your finger under each word. That worked, but it struggled with long sentences and lost track of context. If a sentence started with “the bank” and ended forty words later with “river,” older models often failed to connect the two.
Transformers fixed this with something called self-attention, which lets the model look at an entire sentence (or an entire document) at once and decide which words matter most to each other, regardless of how far apart they are.
The paper that started it all
The transformer was introduced in a 2017 research paper from Google called “Attention Is All You Need.” The title says it all: the researchers showed that a network built entirely around an attention mechanism could beat the older, more complicated designs, while also being faster to train.
According to Google’s own writeup, the Transformer needed less computation to train than earlier models and was a much better fit for modern hardware like GPUs, built for doing many calculations in parallel rather than one step at a time. That efficiency is a big part of why AI progress sped up so fast after 2017.
How self-attention actually works (a simple example)
Take the sentence: “I arrived at the bank after crossing the river.” A human instantly knows “bank” means a riverbank, not a financial building, because of the word “river” later in the sentence.
A transformer does something similar. For every word, it compares that word to every other word in the sentence and assigns an “attention score,” a number showing how relevant each other word is. When the model processes “bank,” it can give a high attention score to “river,” which tells it to lean toward the riverbank meaning. This comparison happens for every word, all at once, which is what makes transformers both accurate and fast.
Stack this process across many layers, and the model builds up a rich understanding of meaning, tone, and context before it ever generates a reply.
Why transformers replaced older AI models
According to IBM’s explanation of transformer models, the design uses mechanisms like attention, self-attention, parallel processing, and positional encoding to understand context across large amounts of data and predict the right output for a prompt. A few reasons this mattered so much:
- They handle long documents far better, since attention connects distant words directly instead of passing information step by step.
- They train faster on modern chips (GPUs and TPUs), because attention calculations can run in parallel instead of one word at a time.
- They scale well, meaning bigger models trained on more data keep getting noticeably better, which is exactly what fueled the jump from early chatbots to today’s LLMs.
This scaling behavior explains a lot of the last few years in AI. Once researchers had an architecture that improved reliably with more data and compute, progress became a matter of resources as much as new ideas.
Where you’re already using transformers
You’ve probably used transformer-based tools today without thinking about it:
- Chatbots like ChatGPT, Gemini, and Claude, which generate text one token at a time using a transformer.
- Translation tools, which use the same attention idea to match meaning across languages.
- Search engines and autocomplete, which use transformer models to understand intent, not just keywords.
- Image tools like Midjourney or ChatGPT’s image generator, which borrow the same attention mechanism adapted for pixels instead of words.
Quick tip: if you want to see attention in action, ask a chatbot a question with a pronoun far from what it refers to, for example “My laptop crashed after I updated the software, can you help me fix it?” A good transformer-based model will correctly figure out that “it” means the laptop, not the update.
Transformers vs neural networks vs LLMs: how they connect
These three terms get mixed up a lot, so here’s the simple version:
- A neural network is the general family of brain-inspired models that learn from data.
- A transformer is one specific, very successful type of neural network design, built around self-attention.
- A large language model (LLM) is what you get when you train a transformer on massive amounts of text so it can predict and generate language.
So when ChatGPT generates a reply, it’s an LLM, built on a transformer, predicting the most likely next token based on everything that came before it. From my own experience testing different AI writing and research tools while building websites, understanding this one idea makes it easier to guess why a model sometimes loses track of a long conversation. It’s still predicting based on attention, and attention has limits.
Common Questions
Do I need to understand transformers to use AI tools?
No. You can use ChatGPT, Gemini, or Claude without knowing any of this. But knowing roughly how attention works helps you write clearer prompts and understand why AI sometimes misreads context in long conversations.
Is a transformer the same as a large language model?
Not exactly. The transformer is the underlying architecture. An LLM is the finished, trained model built using that architecture on huge amounts of text.
Are transformers only used for text?
No. The same attention idea has been adapted for images, audio, video, and even protein folding research, which is part of why the 2017 paper had such a lasting impact across AI, not just chatbots.
Final takeaway
The transformer isn’t a mysterious black box, it’s a specific idea: let every word pay attention to every other word, all at once, instead of reading one at a time. That one shift, first proposed by Google researchers in 2017, is the foundation almost every modern generative AI tool is built on. The next time a chatbot understands your rambling question perfectly, you’ll know exactly what’s happening behind the scenes.
by admin | Jul 11, 2026 | Opportunities
Everyone keeps telling you to “learn AI skills” for your next job or promotion. Fair enough. But learn what, exactly? Where do you even start when there are thousands of AI courses, tutorials, and certificates competing for your attention online?
Microsoft built a free tool for exactly this problem. It is called AI Skills Navigator, and instead of dumping another course catalog on you, it asks a few questions about your goals and role, then builds you a personal learning path. No cost, no account required to browse, and it is open to anyone in the world, not just Microsoft employees or students.
What is the Microsoft AI Skills Navigator?
AI Skills Navigator is a free learning platform at aiskillsnavigator.microsoft.com. Microsoft describes it as a way to help “every organization, every role, and every learner” find the right AI learning path, and it is listed as one of the main free resources on Microsoft’s own AI skills and training page.
Instead of browsing a giant list of Microsoft Learn courses and guessing what matters, you tell the tool your goals, current skill level, interests, and how you like to learn. It then pulls together role-based playlists made from Microsoft Learn courses, hands-on labs, short videos, and even podcasts, all aimed at getting you from “I know nothing about this” to “I can actually do this.”
How it’s different from a normal course catalog
Most free learning platforms answer the question “what can I learn?” Microsoft designed this one to answer a more useful question: what do you need to learn next, and why does it matter for your actual job. According to Microsoft’s own Inside Track blog, the platform is built around four ideas: playlists tied to real roles, hands-on skills rather than just watching videos, learning formats that fit into a busy workday, and credentials that actually prove what you can do.
That last part matters if you are job hunting. A stack of half-finished course certificates does not impress anyone. A credential tied to a role, like “AI fundamentals for business analysts,” is a lot easier to explain in an interview.
What you actually get for free
- Role-based learning playlists (student, career switcher, developer, business user, manager, and more)
- Microsoft Learn courses and structured learning paths
- Hands-on labs so you practice instead of just watching
- Short videos and audio content you can use during a commute or lunch break
- Credentials and certification prep tied to specific skills, not just course completion
None of this requires a paid Microsoft 365 subscription to get started, and it works whether you are a complete beginner or already work in tech and just want to add AI skills to what you know.
Quick tip: Pick one specific goal before you open the tool, like “I want to use AI safely at my current job” or “I want an entry-level AI credential.” A focused goal gets you a much more useful playlist than going in with “I want to learn AI” in general.
How to start using it today
- Go to aiskillsnavigator.microsoft.com
- Answer the short prompts about your role, goals, and current skill level
- Review the playlist it builds for you, it is fine to skip sections you already know
- Work through the hands-on labs, not just the videos, since that is where the real skill-building happens
- Check back every few weeks. Microsoft updates the playlists regularly, so returning learners keep seeing new, relevant content
If you already have a favorite starting point, our guide on how to learn AI for free covers other no-cost options too, and pairs well with this tool if you want a broader view before picking a path.
Who this is really for
This is not just for developers. Microsoft built the playlists around roles like business analysts, project managers, educators, and IT support staff, not only engineers. If you are worried about being left behind at work because “everyone else understands AI,” this is a low-pressure way to catch up without committing to a paid bootcamp.
From my own experience helping people set up websites and online tools, the biggest barrier is rarely the AI itself. It is not knowing where to start or which course is worth your evening. A tool that just tells you “start here, based on your goal” removes that decision fatigue completely.
If a paid credential is your real goal, it is worth comparing this against our list of free AI certifications you can get online, since some of those overlap with what AI Skills Navigator recommends.
A word on AI Skills Fest, and why the tool outlasts it
Microsoft also runs an annual event called AI Skills Fest to get people started on the platform in one focused week. The 2025 edition brought together more than 126,000 participants in a single day and set a Guinness World Record for AI skilling participation, according to Microsoft’s Inside Track blog. The 2026 edition ran in June, but here is the part that matters if you are reading this after the event ended: AI Skills Navigator itself is not a one-week event. It is a standing, ongoing platform, and Microsoft has said the goal now is “sustaining long-term engagement” rather than treating the Fest as the only entry point.
In plain terms, you have not missed anything by finding this in July. The tool works exactly the same whether you show up during a big event week or on a random Tuesday.
Building AI skills also matters for your paycheck, not just your resume. Our post on why AI skills now pay more breaks down the data on that if you want the bigger picture. And if coding has been the thing holding you back, you can learn AI without coding too, no computer science degree required.
Common Questions
Is Microsoft AI Skills Navigator really free?
Yes. It is listed as a free public resource on Microsoft’s AI skills and corporate responsibility page, and no paid Microsoft 365 subscription is needed to use the core learning paths.
Do I need to already know how to code?
No. The playlists cover business, education, and non-technical roles as well as developer paths, so beginners have plenty to work with.
Will I get an actual certificate?
Some playlists lead to Microsoft Applied Skills credentials or certification exam prep. Others are shorter skill-building content without a formal credential. Check each playlist description before you start if a certificate matters to you.
Final takeaway
You do not need to figure out AI learning alone, and you definitely do not need to pay for it just to get started. Microsoft AI Skills Navigator takes the guesswork out of “what should I learn first,” and it is built to keep working for you long after any single event ends. Give it fifteen minutes, answer honestly about your goals, and see what path it builds for you.
by admin | Jul 10, 2026 | AI Tools
If you downloaded OpenAI’s Atlas browser last October to let ChatGPT handle your tabs, forms, and bookings, you’re about to get some news. OpenAI announced on July 9, 2026 that it is shutting Atlas down. The browser stops working on August 9, 2026.
This isn’t just a small product update. It’s a good moment to understand what “AI browser agents” actually are, why OpenAI is walking away from its own standalone browser after less than a year, and what your options look like now if you want an AI assistant that can act inside your browser.
What is happening to ChatGPT Atlas?
According to OpenAI’s own help center, Atlas is “scheduled to stop working on August 9, 2026.” OpenAI says it is folding the browser-based agentic features people liked about Atlas (multiple tabs, downloads, navigation, account login support) into the main ChatGPT app instead, along with a ChatGPT Chrome extension and sidebar for people who just want help while they browse in Chrome.
In plain terms: OpenAI decided it doesn’t need a whole separate browser to give people an AI agent that can click, scroll, and fill in forms for them. It would rather build that into ChatGPT itself and into a Chrome extension, which reaches far more people than a standalone app ever could.
What you need to do before August 9
If you have been using Atlas, OpenAI’s guidance is straightforward, but easy to miss if you don’t check your email:
- Export your bookmarks to an HTML file before August 9, then import them into Chrome or another browser.
- Save or copy the URLs of any open tabs you care about. Tabs will not carry over automatically.
- Bookmark or save anything from your browsing history you might need later.
- Treat any exported cookie or session files as sensitive data. Don’t share them with anyone you don’t fully trust.
Quick tip: your ChatGPT conversation history is completely separate from your Atlas browser data, so none of your actual chats are at risk. It’s only the browser-specific stuff (bookmarks, tabs, history, cookies) that needs manual saving.
Why this matters even if you never used Atlas
Atlas launched in October 2025 as OpenAI’s bet that people wanted an entire browser built around ChatGPT. Less than a year later, that bet didn’t pay off the way a standalone product needed to. That’s a useful signal about where this whole “AI browser agent” category is actually heading in 2026: not a handful of separate browsers competing for your default browser slot, but AI assistants that plug into the browser you already use.
From my own experience testing different AI tools for client websites and day to day work, the products that stick are usually the ones that fit into an existing habit rather than asking you to replace one. A Chrome extension is a much smaller ask than “switch your entire browser.”
What are your options now?
If you liked the idea of an AI agent handling browser tasks for you, a few real alternatives exist today, and they take different approaches:
- Perplexity Comet is a full standalone browser, available on Mac, Windows, iOS, and Android, that can research, summarize, and complete multi-step tasks for you. If you want to try the “browsing assistant” idea, this is the most direct like-for-like replacement for what Atlas offered.
- Claude in Chrome (Anthropic) is a Chrome extension rather than a full browser. It can read a page, click buttons, fill in forms, and work across multiple tabs, and it’s currently available to Pro, Team, and Enterprise plan subscribers.
- Gemini in Chrome lives inside Google’s own Chrome browser as a side panel, with an “auto browse” agent mode for multi-step tasks like comparing prices or filling out forms, available to Google AI Pro and Ultra subscribers.
If you’re just getting started and don’t want to commit to a subscription yet, our ChatGPT vs Gemini vs Claude comparison and our guide to what Perplexity AI actually does are good places to see how these tools differ before you pick one.
The privacy question worth asking
Any tool that can see and act on everything you do in a browser, your open tabs, your logged in accounts, your shopping carts, deserves a moment of caution. Before you let an AI agent browse on your behalf, check what permissions it’s asking for, avoid giving it access to banking or health accounts, and review its data settings. We cover this in more detail in how to use AI safely and protect your privacy, which is worth a read regardless of which tool you end up choosing.
This is also a good moment to remember that “agentic” AI is still new. Mistakes happen: an agent might click the wrong button, submit a form early, or misread a page. Keep an eye on what it’s doing, especially for anything involving money or personal information.
Common Questions
Will my ChatGPT conversations be deleted when Atlas shuts down?
No. Your ChatGPT conversation history is stored separately from Atlas browser data, so it stays available in the regular ChatGPT app.
Do I have to switch to a new browser right away?
No, but you should export your Atlas bookmarks and save any important tabs or history before August 9, 2026, since that data won’t transfer automatically.
What replaces Atlas for AI browsing?
OpenAI is moving browser-based agent features into the main ChatGPT desktop app and a ChatGPT Chrome extension. If you want a different option, Perplexity Comet, Claude in Chrome, and Gemini in Chrome all offer similar AI browsing assistance.
Final takeaway
OpenAI killing its own Atlas browser after less than a year isn’t really a failure story, it’s a sign the whole AI browsing space is still figuring itself out. If you used Atlas, take ten minutes this week to export your bookmarks and save what matters. And if you’re curious about AI agents that can act inside your browser, you now have a clearer picture of where each major player stands, and it’s worth reading up before you hand any of them the keys.
by admin | Jul 9, 2026 | Research & Productivity
Ask anyone who has written a thesis which part they underestimated, and you will usually get the same answer: the literature review. You start with one search, and two weeks later you have sixty open tabs, a folder full of unread PDFs, and no clear picture of the field.
AI tools can remove a lot of that pain if you point them at the right jobs. This guide walks through a five step literature review with AI, using free tools, and it stays honest about the parts you still need to do yourself.
Can you really do a literature review with AI?
Partly, yes. AI is genuinely good at three jobs here: finding papers that match your question even when you do not know the perfect keywords, summarizing individual papers quickly, and showing how papers connect to each other.
What it cannot do is judge research quality the way you can, decide why a gap in the field matters, or build your argument. AI models also make mistakes with total confidence. They sometimes invent references or misread a paper\u2019s findings, a problem we explained in AI hallucinations explained. So the workflow below uses AI for speed and keeps the judgement with you.
Someone close to me spends her days in PhD research on machine learning and medical imaging, so I have watched how fast a reading pile can grow. The researchers who cope are not the ones reading faster. They are the ones with a better system.
Step 1: Turn your topic into a real question
\u201cAI in healthcare\u201d is a topic. \u201cHow accurate are deep learning models at detecting brain tumours from MRI scans?\u201d is a question. Every step that follows works better when you start from a question, because modern research tools use semantic search. They match meaning, not just keywords.
Write your question down before you open any tool. If you cannot phrase it yet, that is useful information too. Spend an hour with a general overview or a textbook chapter first, then come back.
Step 2: Find papers with AI search tools
Three tools cover most of the discovery work:
- Elicit searches more than 138 million papers. You type your question and it returns a table of relevant papers with short summaries. The Basic plan is free, and it can import your library from Zotero.
- Semantic Scholar is a free academic search engine from the non-profit Allen Institute for AI. It indexes over 200 million papers and adds short AI generated summaries, called TLDRs, so you can screen results quickly.
- Research Rabbit maps papers visually. You start with one paper you already trust, and it shows similar, earlier, and later works, so you follow the citation trail instead of searching blind.
University libraries have started recommending these tools too. The University of Michigan Library keeps a practical guide on AI in literature reviews if you want a librarian\u2019s take on the same tools.
Tip: run the same question through two different tools. Each one searches differently, and the papers that appear in both lists are usually the ones to read first.
Step 3: Screen and organize what you find
You will collect far more papers than you need, so do not try to read them all. Screen each one by its abstract or TLDR and sort it into three piles: keep, maybe, and drop. Be ruthless with the drop pile.
For the keepers, use Zotero, a free and open source reference manager. It stores your citations, formats them in thousands of styles, and connects with Elicit and Research Rabbit. We covered where it fits in our guide to AI tools for thesis writing.
Step 4: Summarize and compare the papers
For every paper you kept, you want four things: the question it asked, the method it used, what it found, and its limitations. AI can speed this up a lot. Our guide on how to summarize research papers with AI shows practical prompts, and our NotebookLM and Elicit walkthrough covers tools that answer questions only from the sources you upload.
One warning from experience: AI extraction makes mistakes. It can misread a sample size or blur two findings together. Check every number and claim against the actual paper before it goes anywhere near your draft.
Step 5: Write the review yourself
Here is the part no tool can do. A literature review is not a list of summaries. It is an argument about the state of a field: what researchers agree on, where they clash, and which gap your work will fill. That structure has to come from your reading, so group your papers by theme or debate, not by author.
Two rules protect you here. First, verify that every reference exists and says what you claim, because AI generated citations are sometimes fake. Second, check your university\u2019s AI policy and disclose what you used. Most universities now allow AI for searching and summarizing but treat AI written text as misconduct.
From my own work with websites and online tools, the pattern is always the same. Tools that remove boring steps earn their place. Tools that promise to think for you cause trouble later.
Common Questions
Can AI write my literature review for me?
It can produce text that looks like one, but that is the trap. The references may not exist, the synthesis is shallow, and most universities treat submitting it as academic misconduct. Use AI to find, organize, and summarize. Write the argument yourself.
Are these AI research tools free?
Yes, for everything in this workflow. Semantic Scholar is completely free, Elicit has a free Basic plan, Research Rabbit lets you sign up free, and Zotero is free and open source.
How many papers should a literature review include?
It depends on your field and level. A bachelor\u2019s thesis might cover 20 to 40 papers, while a PhD literature review can pass 150. Your supervisor\u2019s guidance beats any general number, so ask early.
Final Takeaway
A literature review with AI is not about outsourcing the reading. It is about shrinking the boring parts: hunting for papers, formatting citations, and writing first pass summaries. Pick one question, run it through Elicit or Semantic Scholar this week, and save what you find into Zotero. The pile gets smaller, the map gets clearer, and the thinking stays yours.
