Teach Yourself How To Use Comet

Teach yourself how to use Comet based upon your browsing and query history. Learn about the most recent feature updates and see five specific and personalized queries that you should try. Surface the most compelling use cases for your business and find prompt examples that you might see on social media.

How Do You Use Cement Based Upon Your Browsing and Query History?

Using Comet with your browsing and query history turns everyday research into a faster, more personalized workflow. Instead of treating each search as a one-off task, Comet builds context over time—learning what you’ve already looked at, what questions you’ve asked, and where you tend to go next.

Start by browsing naturally. As you read articles, open tabs, and run searches, Comet quietly tracks this activity to understand your intent. When you ask a new question, Comet can reference prior queries and visited pages to deliver answers that are more relevant and less repetitive. For example, if you’ve been researching SEO tools, future queries about “best reporting dashboards” will automatically lean toward marketing and analytics use cases rather than generic software lists.

You can also use your history proactively. Ask Comet questions like, “Summarize what I’ve read about this topic,” or “Compare the sources I looked at earlier today.” This is especially useful when you’re juggling multiple projects or returning to research after a break. Instead of retracing steps, Comet surfaces key takeaways and highlights gaps you may want to explore next.

Privacy and control remain important. You can review, pause, or clear your browsing and query history at any time, ensuring the system only retains what you’re comfortable sharing. Used thoughtfully, this balance of memory and control allows Comet to function like a smart research assistant—one that remembers context, reduces friction, and helps you move from questions to decisions with far less effort.

What are Five Personalized and Specific Queries to Try?

  • “Based on the articles I’ve read this week, what am I missing on this topic?” This leverages your recent browsing to identify gaps, opposing viewpoints, or advanced angles you haven’t explored yet.

  • “Summarize the key takeaways from the pages I visited about this problem and give me a recommended next step.” Ideal after a research session—Comet synthesizes across multiple tabs instead of just one source.

  • “Compare the tools I’ve already looked at and tell me which one best fits my use case.” Rather than re-explaining each tool, Comet references what you’ve already reviewed and narrows it down based on intent.

  • “I’m switching back to this project—remind me what conclusions I was leaning toward and why.” Powerful for context recovery when you revisit work after days or weeks.

  • “Given my past searches, what would you recommend I research next if my goal is better results, not more information?” This shifts Comet from passive search assistant to active strategist, prioritizing impact over volume.

  • What are the Best, Most Useful Prompts From Social Media?

  • “Summarize everything I’ve read today like I’m briefing a coworker in 60 seconds.” Popular because it turns scattered tabs into an instant stand-up update.

  • “Explain what I’m researching as if I were new to this topic.” Often shared as a way to sanity-check whether your research actually makes sense.

  • “What conclusion would most people jump to after reading what I’ve read—and is it wrong?” This prompt gets traction because it challenges confirmation bias.

  • “Turn my browsing history into a decision memo with pros, cons, and a recommendation.” Frequently posted by founders and operators as a replacement for long internal docs.

  • “If I had to make a decision today based only on what I’ve read, what should it be?” A favorite for avoiding research paralysis.

  • Exploring Perplexity’s Comet Agentic Browser for Business ROI

    As the AI landscape rapidly evolves, a new category is creating tangible value for businesses: agentic AI browsers. Rather than focusing solely on advanced chatbots or proprietary AI models, companies are now witnessing real productivity gains from tools that seamlessly integrate multiple AI capabilities with web browsing. Perplexity’s Comet, an agentic AI browser, stands out for its hybrid architecture and immediate business applications. Here’s a precise breakdown of what this technology delivers and why it’s poised to influence how modern organizations approach efficiency and growth.

    Agentic AI Browsers: Understanding Their Strategic Position

    The momentum around agentic AI browsers is driven by specific shifts in the competitive landscape. Traditional answers engines have been challenged by deeper research modes from Google and OpenAI, prompting Perplexity’s significant pivot away from direct competition. With Comet, the focus has moved toward running sophisticated AI tasks natively in the browser environment, providing advantages in speed, control, and user context not easily achieved by standard chatbots or cloud-based agents.

    Other major technology players have also signaled their intentions in this area: Google has Project Mariner and is preparing its own agentic browser; Microsoft’s Edge quietly expands its agentic features; and OpenAI is set to enter the field imminently. Each recognizes that browser control offers greater access to user workflows than hosted chatbot portals, establishing a new AI battleground on the very desktops of professionals.

    How Perplexity’s Comet Works: Architecture, Access, and Security

    Comet is built on Chromium, ensuring compatibility with Chrome extensions and offering a frictionless user migration—passwords, bookmarks, and preferences are imported automatically when logging in with Google credentials. This setup is critical for business users who rely on fast onboarding and consistent web experiences.

    Most notably, Comet employs a hybrid AI approach, running simple tasks locally within the browser while delegating more complex processes to Perplexity’s cloud. This local execution avoids many of the pitfalls seen in early computer-vision-based agents, which suffered from slow performance and reliability issues (such as those observed when trying to execute repetitive UI actions in virtual browser sandboxes).

    Access is currently prioritized for those on Perplexity’s top-tier subscription, with gradual rollout for other paid users and eventual plans to open access further. Early adopters benefit from integration features such as connectors to Gmail, Google Drive, and Calendar, with Microsoft 365 integration likely in the pipeline.

    Perplexity Comet Business Use Cases: Automation in Action

    Five distinct business use cases are already transforming day-to-day operations for professionals using Comet:

    1. Personalized, Multi-Step, Multi-Platform Research:
    Comet enables complex research workflows that traverse Gmail, Google Drive, social media feeds, and calendar data—all in one command. For example, it can summarize recent podcast episodes, save drafts automatically to Google Docs, and trigger follow-up actions, such as emailing reports, without manual intervention. This creates an up-to-date “second brain” dynamically synchronized with key business knowledge bases and accessible by other AI tools.

    2. Time Savings on Social Media Analysis:
    Instead of navigating the distracting black holes of LinkedIn and Twitter, Comet pulls trending topics, critical DMs, and actionable insights relevant to the business automatically. This operationalizes social monitoring for executives, saving hours while retaining vital awareness of industry movements.

    3. Executive Email and Calendar Management:
    By connecting to calendar and email platforms, Comet automates event creation, scans for important correspondence, and identifies meeting requests—functionality closely resembling an executive assistant’s capabilities. For organizations frustrated by the fragmentation of digital schedules, automated summarization and proactive response reduce missed opportunities and improve time management.

    4. Automated Market and Competitive Intelligence:
    By informing Comet with details about business roles, competitors, and market context, organizations can automate recurring competitive research. Regular prompts can have Comet gather intel, aggregate news coverage, and store findings in dedicated Google Drive folders—ready for strategic review.

    5. Multi-AI Tool Orchestration:
    Perhaps most uniquely, Comet coordinates across multiple proprietary AI tools. One demonstration included having Comet scan a specific ChatGPT conversation, extract areas needing deeper exploration, and route queries for further research in Google Gemini, producing comprehensive cross-platform intelligence reports. This type of orchestration enables organizations to maximize the potential of every licensed AI platform, with Comet acting as the conductor.

    Productivity Gains Experienced Firsthand

    Early testing has documented a dramatic reduction in manual work. Workflows that previously required several hours—such as deep market analysis, cross-channel communications reviews, or document production—can now be automated or handled in the background, freeing up time for strategic initiatives. The fusion of browser-native control and cloud AI power means barriers common in older agent or workflow tools (like Zapier or Make) are eliminated, offering more context-awareness, reliability, and real-time transparency.

    Implications for Business Leaders and Decision Makers

    The shift toward agentic AI in the browser is not theoretical—it is happening now, with Google, Microsoft, and OpenAI all making investments that signal this will become the standard interface for interacting with AI. Unlike AI chatbots, which require deliberate data connection and yet may be siloed, agentic browsers naturally unify access to all open tabs, extensions, and logged-in resources, delivering a more holistic and actionable intelligence layer.

    For business owners and decision makers, early engagement with agentic AI browsers like Perplexity’s Comet offers a direct path to faster research, smarter communications, and dynamic data-driven discovery, ahead of competitors still relying on more segmented workflows.

    Preparing for the Next Phase of AI Adoption

    Digital workflows are on the cusp of another leap in efficiency and capability. Agentic AI browsers, led by Perplexity Comet’s hybrid, browser-integrated architecture, already provide measurable ROI across research, communication, intelligence, and multi-AI coordination tasks. For companies seeking practical, implementable AI advantage, this new category of browser-based autonomy represents an immediate opportunity worth exploring and adopting proactively.

    For ongoing updates on this subject and practical AI business strategies, staying informed on the latest releases and best practices will help organizations realize the full potential of these new tools as they mature.


    Topics Covered in This Episode:

    1. Rise of Agentic AI Browsers Explained
    2. Perplexity Comet’s Hybrid AI Architecture
    3. Local vs. Virtual Browser Agentic Workflows
    4. Perplexity Comet Key Features and Access
    5. Connecting Google Services with Comet Browser
    6. Agentic AI Browser Live Demo Use Cases
    7. Multi-Platform Personalized Business Research
    8. Automated Market Research and Competitive Analysis
    9. Perplexity Comet Cross-Tool Workflow Automation
    10. Agentic AI Browsers vs. Traditional AI Chatbots


    Keywords:

    Perplexity Comet, agentic AI browser, agentic AI, AI browser, business use cases, ROI, Perplexity app, hybrid AI, cloud AI, on-device AI, Comet browser, Chromium, browser automation, multi-platform research, deep research modes, AI desktop assistant, iOS assistant, Google Chrome, browser workflows, Chrome extensions, Microsoft Edge, computer using agents, virtual browser, computer vision, connector integration, Gmail integration, Google Drive integration, calendar management, task automation, prompt engineering, personalized research, market research, competitive analysis, social media automation, email management, calendar automation, Google Gemini, ChatGPT, Claude, OpenAI operator, Project Mariner, Microsoft Copilot, Edge browse


    Podcast Transcript

    Jordan Wilson [00:00:17]:
    There's a new wave happening in the AI world and it's not specifically about a new AI chatbot or the latest AI model or even agents, but at the same time, it's kind of combining all of those things. So what is it that I'm talking about? It's agentic AI browsers. And on today's show for putting AI to work on Wednesdays, we're going to be looking at one of the more popular ones that is released and available now to most people, which is Perplexity's Comet. And we're going to go over what it is, how it works, and talk about five different business use cases that you could start using today. All right, I hope you're excited. I am. Let's get into it. If you're new here, welcome to Everyday AI.

    Jordan Wilson [00:01:09]:
    This is your daily live stream, podcast and free daily newsletter helping everyday business professionals like you and me not just learn what's happening in the world of AI, but how we can leverage it to grow our companies and our careers. So if that's what you're trying to do, you're definitely in the right place. It starts here with the unedited, unscripted, sometimes off the rails live stream and podcast. But it continues where you can become the smartest person in AI by going to our website, your everydayai.com signing up for the free daily newsletter. We're going to be recapping most important insights from today's conversation as well as keeping you up to date with everything else that you need to know. So make sure, if you haven't already, please go to our website and do that. And if we're going to have the AI news there as well, obviously. So go make sure to check that out.

    Jordan Wilson [00:01:54]:
    So let's talk about this new wave agentic AI browsers. I actually am extremely bullish on this and let me be honest. And you can go back, I have the receipts and the receipts are on our website, more than 560 episodes. You can go back and listen to what I've been saying over the years. But one thing, and I've been correct on, I've been saying for a long time with Perplexity, I haven't been very strong on Perplexity as an answers engine.

    Jordan Wilson [00:02:25]:
    Right.

    Jordan Wilson [00:02:25]:
    I said back in my 2025 AI predictions and roadmap series that Perplexity was either going to have to hard pivot or they would potentially get squashed. Yet here we are. We've had a lot of reports about Perplexity potentially getting acquired. I don't think they are a lot of those have been shot, shut down over the last week or two. But what we've seen from Perplexity is a hard pivot. They can no longer compete as an answers engine, specifically after Google and OpenAI released their extremely impressive deep research modes. So what we've seen out of Perplexity, which I'm very kind of on board with, is this recent pivot specifically, I think in two different areas, try to be an iOS assistant because, you know, Siri and Apple has failed. So they have a great iOS app for that.

    Jordan Wilson [00:03:14]:
    And now one of the first, I would say big tech companies coming to production with an agentic AI browser with Comet. All right, so that's what we're going to be going over today. So let me just recap what we're going to be going over on today's show. So I'm going to talk briefly about why agentic AI browsers are the next wave of generative AI. I'm going to tell you how Perplexity comment works and then because it is putting AI to work. On Wednesdays, our new weekly series, I'm going to go over five business use cases that you can use and give you some examples of how I'm using Perplexity comment as well. Yeah, so even if you listen to our news show on Monday, I kind of alluded to this. I was actually running a perplexity comment in the background as I was recording the, the new show on Monday.

    Jordan Wilson [00:04:07]:
    So it's something I'm already using and I think that there's a lot of cool business use cases for it as well. Hey, shout out live stream audience. Good to see you. I don't know if you can tell I'm actually extremely sick. So if, if this is a little shorter or lacking the usual, the, the usual spice and vigor, I'm, I'm definitely 100% under the weather, just FYI, you know, you don't have to think that there's been an AI avatar takeover. Like, who's this weird, you know, Jordan? That doesn't sound like him. A little raspy. Yes.

    Jordan Wilson [00:04:39]:
    I'm feeling absolutely garbage. But it makes everything better to see you all here. Big bogey, good to see you. Michelle joining on the YouTube machine. Robert joining from Dallas, Texas on LinkedIn. Marie, good to see you. Joining from New Jersey, Jose from Santiago. It's good to see everyone.

    Jordan Wilson [00:04:56]:
    Angie, good morning to you as well. All right, so let's get into talking about perplexities comment but first, why is this important? All right, so if you're listening, especially on the podcast, I'm going to be showing, you know, sharing my screen, but I'm going to do my best to walk you all through it. But here's why it's important and why I think actually agentic AI browsers are the next wave of AI and not necessarily a chatbot. So number one, you have to see the players and where they're deciding to play. So you know, last week on the show we had the reporter from Reuters who broke the story on OpenAI is also coming out with an agentic AI browser in the coming weeks. All right, Google, which we know is kind of the, the king in the browser world, they've been experimenting with their own agentic capabilities in their Project Mariner and then also Microsoft Copilot. One of their, I think under, I guess underutilized or not very talked about features is their browser, which I love the Edge browser from Microsoft, but specifically its vision capabilities. And I think that Microsoft is going to continue to expand and improve what is capable inside of its Edge browsers.

    Jordan Wilson [00:06:24]:
    But let me fast forward. I think the end play here is the browser. I think it's so much more about going to, you know, ChatGPT.com or gemini.google.com or Claude AI.

    Jordan Wilson [00:06:37]:
    Right.

    Jordan Wilson [00:06:37]:
    I think what the big tech companies know and realize, they, they don't want to be front facing AI chatbots.

    Jordan Wilson [00:06:44]:
    Right.

    Jordan Wilson [00:06:45]:
    Ultimately everyone wants to control your desktop but right now only, you know, Microsoft, Windows, sl, Apple can do that. But then the next best thing is controlling your browser.

    Jordan Wilson [00:06:58]:
    Right.

    Jordan Wilson [00:06:58]:
    And there's so many new added capabilities that I guess we can all realize and use when we are using a browser versus when we are using a front end AI chatbot.

    Jordan Wilson [00:07:15]:
    All right.

    Jordan Wilson [00:07:15]:
    And we're going to go over some of those in today's show. But I think that's ultimately what this boils down to. There's so many more agentic capabilities when you are using a browser versus when you are using a front end chatbot.

    Jordan Wilson [00:07:35]:
    All right.

    Jordan Wilson [00:07:35]:
    And I, I, I do admit that the first kind of six to nine months of the computer using agents kind of phase has been a little bit of a flop.

    Jordan Wilson [00:07:49]:
    Right.

    Jordan Wilson [00:07:49]:
    So this goes back to 2024. Anthropic Claude was the first with their computer using Agent OpenAI followed with their operator Agent, which I think when it first came out was not very good because it was powered by the GPT-4O model. Now it's powered by the o3 model, which is so much better. But this is a difference, right? These computer using agents that we've seen, they essentially use a virtual browser, right? A virtual sandbox, and then they use computer vision, which is a little slow, right, and also a little wonky. So as an example, right, I did a couple shows on Operator when it first came out. One thing I was trying to do with Operator was, you know, have it go through my LinkedIn messages because I stink at keeping up. And then anything that was important, essentially just mark with a star, right? But the problem, because it was using a virtual browser, right? And it's using computer vision, essentially it takes a screenshot every so often and then it, you know, kind of uses coordinates like you would think from a map. So it takes a screenshot and then it says, okay, I need to click here.

    Jordan Wilson [00:08:58]:
    So number one, it's a little slow. But also if there's anything wonky in a website's UI ux, it's going to throw it off by a couple of pixels. So what happened in my use case, LinkedIn's inbox is wonky. So as it was trying to do that, it was not able to scroll down. And then the star button and the archive button were like two pixels from each other. So it actually kept archiving messages, which was the exact opposite of what I wanted to do. So the big benefit of moving this to an actual browser versus the combination of using a virtual browser and essentially computer vision is it's running on your local machine. So what we have with Perplexity's Comet is essentially hybrid AI, right? So we're getting some Edge AI capabilities because it's running some things on device in the browser, and then for more complex tasks, it's still, you know, sending it to Perplexities Cloud.

    Jordan Wilson [00:09:55]:
    So let's talk a little bit about how specifically Comet works. So right now it's built on Chromium, as a lot of browsers are, and we'll talk a little bit about why that's important probably in future episodes, right? But Google Chrome has their very popular kind of open source version of Google Chrome, which is Chromium. That's what Microsoft Edge is built on. And that's also what Perplexity Comment is built on as well. So there's pros and there's cons in that, but we'll talk about that a different day. But what that means specifically is when you sign up for Perplexity or when you're using Comet, you will have the ability to to essentially log in with your Google Chrome or your Gmail or your, you know, Google workspace credentials. So it's going to save all your passwords, all of your bookmarks, all of your preferences are going to get imported directly. So that's big, right? Same thing with Chrome extensions.

    Jordan Wilson [00:10:50]:
    Everything just works. Love Chromium and the fact that Google makes that open and available even to their competitors, right? So not only does Microsoft use it, but in this case Perplexity uses it and reportedly OpenAI is going to be using Chromium as well. The other thing about how come it works is it does have this hybrid AI architecture. So some more simple AI tasks run locally, which from a accuracy perspective in my initial testing makes Perplexity Comet much better than some of the other computer using agents that use more of a of a virtual browser slash computer vision approach because that approach is slower. So by running everything locally in the browser that you own.

    Jordan Wilson [00:11:33]:
    Right.

    Jordan Wilson [00:11:33]:
    That just makes it so much better. And then the more complex tasks, you know, essentially it's going to run via the cloud. You know, you obviously have to have a Perplexity account, so it's available immediately. Let's talk about access. Perplexity Comment is available immediately to people on the 200amonth max plan, which is not a lot of people. Aside from that, it's getting rolled out to people who are on the wait list if you have a normal paid account. So that's. I did get access to this last week.

    Jordan Wilson [00:12:04]:
    I'm not on Perplexity Max plan. I'm on most of the max plans for other services. Not for Perplexity, I'm on the normal paid plan. But I did get access to Perplexity's comment browser last week because I did sign up for the waitlist early and I'm on the paid plan. But Perplexity did even say that they will make this available in the future to free users as well. Whether that's in a couple of months or a year, we don't know. But if this is something you want, if you haven't already sign up for the wait list. You should probably go do that if you are on a normal $20 a month perplexity pay plan.

    Jordan Wilson [00:12:41]:
    And we shared about this in our newsletter as well. We told you, hey, go sign up now for the wait list if you want access to it when it comes out. Because sometimes it is like a month's or multiple months gaps right between people who have quote unquote earlier access. If you sign up for a wait list soon versus when it starts rolling out to everyone else. That's why you need to read our newsletter. We keep you up to date with all those things. Another thing to keep in mind with Perplexity's comment is it also has the context of all your open tabs in the browser, which is really cool and a big difference on, you know, how you might be connecting models like Gemini, Claude or ChatGPT to different websites that you may want.

    Jordan Wilson [00:13:22]:
    Right?

    Jordan Wilson [00:13:22]:
    So generally you would have to leave a link or, you know, tell your AI chatbot to browse the web and to go to certain web pages. And if you're running a deep research, it's going to go to hundreds. But how Perplexity works comment is it has the context of everything that you have open as well as you'll probably want to set up Perplexities connectors in your account. So if you wanted to have access to your Gmail, to your calendar, etc. All right, let me take a sip here because I'm you can tell my my voice is about shot already. Hey, at least for those of you that complain that I mumble on too longer, this will be a shorter episode just because I'm literally not going to be able to talk. But let's start live. Let's take a look.

    Jordan Wilson [00:14:17]:
    And as always, there's probably going to be a lot of things that go wrong doing live demos with generative AI, but let's just go ahead and give it a look. So one thing I'm going to start with and we're going to see how this works. All right, so I'm going to start a couple for our podcast audience. I have the the Perplexity comment browser open right now.

    Jordan Wilson [00:14:43]:
    All right.

    Jordan Wilson [00:14:43]:
    It's on my screen and like, like I talked about, it has all my, my Chrome extensions, my bookmarks, everything. So I've already logged into all my, you know, different accounts that I may be demoing here. And again, I've connected it inside my Perplexity accounts to my different connectors so it has access to my Gmail, my calendar, my Google Drive, etc. I hope that Perplexity will open this up to Microsoft's offerings as well their Outlook. I'm sure they will. But here's what I'm going to do and I'm going to give everyone kind of a look around Perplexity comment here, but I'm going to just start a prompt going I'm going to say recap the latest episode of the Everyday AI podcast. Please put that recap in a Google Doc, save it as a PDF, then go in my email and send that recap too. And I put my personal Gmail in there.

    Jordan Wilson [00:15:37]:
    And then I'm just saying please do not ask for my input on anything else. Please execute this task autonomously.

    Jordan Wilson [00:15:43]:
    Right.

    Jordan Wilson [00:15:44]:
    I'm just doing this for demo purposes, y'.

    Jordan Wilson [00:15:46]:
    All.

    Jordan Wilson [00:15:46]:
    I want you to see some use cases on how I'm actually using Perplexity comment and how you can too. But essentially this is just a demo to show you how this works. Okay, so now what's happening? And I just put this in kind of the home page of Comet. So if you're used to using Perplexity, you'll kind of understand. But I have two different tabs here. I have an answer tab and a sources tabs and I can go and see how this is agentically working.

    Jordan Wilson [00:16:15]:
    Right.

    Jordan Wilson [00:16:15]:
    So what it did first is it looked up, it essentially did a search latest episode everyday AI podcast recap, July 2025. And you'll see here different results showed different latest episodes.

    Jordan Wilson [00:16:29]:
    Right.

    Jordan Wilson [00:16:29]:
    So some of these, it brings up 561 for episode number 559. And then others it says 566. So hopefully the comment will know that the latest episode in this case is 566 and not the older ones. So then it's going through. Okay, it did that. Correct. It went to the website for episode 566 from some caching issues. The latest episode is actually 567, but caching issues, sometimes it's going to take a couple of hours.

    Jordan Wilson [00:17:00]:
    All right, so this is interesting. I love doing live demos because I obviously did this this morning before trying this. And this time it says the Google Gmail tools cannot create or convert Google Docs to PDFs directly.

    Jordan Wilson [00:17:14]:
    No.

    Jordan Wilson [00:17:15]:
    Nor can they attach files. It's so funny because I did this earlier and it worked perfectly and it sent the, you know, it recapped it, it created a PDF and then it sent a link to that Google Drive Doc to the email that I told it to. So interesting. It didn't work the second time. So it's just asking me, would you like to proceed with an email containing the recap text or try a different approach? So I'm going to say recap text.

    Jordan Wilson [00:17:41]:
    All right.

    Jordan Wilson [00:17:41]:
    And I'm going to click continue and we'll see what happens here. Hey, live stream audience, let me know. Have any of you use Perplexity's comment yet if you want to see any. Any use cases? I have a couple. I have five business use Cases that I'm going to go over here at the end. But I did want to do a couple of these live demos so you can see well the pros and the cons as well. So you'll see here after doing that now on the right hand side I can actually see what it's doing. So I can go ahead and open this Google Doc that it's working on right now.

    Jordan Wilson [00:18:16]:
    So I'm going to go ahead and click that and click Open page. Okay, so comment went through, it went to that website, it recapped that episode, episode 566 which was our Monday recap. It recapped every single story here, did a really good job and it created that document inside my, my Google Drive, which is great, a Google Doc, right? So I can click on that kind of right hand side where it's telling me what it's doing and I can go in and see. So that's great for a couple of reasons. One, transparency. You need to be able to see what's going on under the hood as we talk about agentic AI browsers, right? Sounds big and scary. It's not. But another couple reasons why this is good.

    Jordan Wilson [00:19:01]:
    And it actually pretty big cheat code, right? Having an agentic AI browser create Google Docs for you is actually pretty huge because you can personalize them, right? For your own perspective. But not just that so many of these AI chatbots, right, obviously including Google OpenAI and Claude now can connect to your Google Drive, right, and do deep research on your Google Drive. So the fact that now I can have a browser go personalize research for me in this use case, it wasn't personalized per se, even though it was about my podcast, right? But I could go have it, you know, every single day, research, I don't know certain companies, competitors, etc, create a Google Doc that automatically gets saved in my account and then through those connectors via those other AI chatbots, it's like literally having an up to date second brain brain that you can then access via any other kind of AI chatbot, right? So this is one way that I'm using Perplexity Comment now is just to personalize research. Save it in my Google Drive, right? And then when I need to, I can pull from that information. All right, so there we go. It says, okay, so it already saved that PDF and you probably can't see it, but it saved it in my browser here. So okay, there we go. So you can see it saved the PDF.

    Jordan Wilson [00:20:33]:
    I have it open here on my screen and now it's going through and trying to create an email and send that email to myself as that requested. So interesting. Obviously, you know, on this one it ran into an error, although last time it worked perfectly twice. All right, let's do one other one or two other live you use cases. All right, and then I'll go over at the end what I think are five pretty cool business use cases. All right, so now let's go ahead and do another example. All right, I'm going to change this. When it comes to AI, it's all about trust.

    Jordan Wilson [00:21:24]:
    And the same thing can be said for how you operate your business and accept payments from your customers. Whether you're sending invoices on the go from your laptop, needing business management solutions, or trying to open another location for your local business, it's all about having the right tools that people trust. That's where Square has your back. Running a business means making 100 tiny decisions a day. And Square helps make those decisions smarter, easier and more efficient. I remember using Square's card reader for mobile devices more than 15 years ago, and I've trusted Square ever since. Whenever I see that Square logo out in my hometown of Chicago, I know that I can trust that business to handle my transaction without problem. Square keeps up so you don't have to slow down, get everything you need to run and grow your business without any long term commitments.

    Jordan Wilson [00:22:11]:
    And why wait? Right now you can get up to $200 off square hardware at square.com go jordan. That's s q U-A-R-E.com g o j o r dash dash a n. Run your business smarter with Square. Get started today. Give me a second here, y'. All. I want to try to stop this other one from going. We'll see if this works.

    Jordan Wilson [00:22:46]:
    Okay, so here's what I said this time and, and think of how this, this may not sound super impressive to you, but I said look at my social media, Twitter, LinkedIn, calendar, Gmail and give me a brief overview of what's happening. Don't worry about connection requests. Mainly look at what's trending in social media feeds and any important DMS messages that don't look like spam or promotion, right? So because I have my Twitter and my LinkedIn logged in and saved, and that's another huge benefit of running these things locally right on your browser is it's going to be able to have access to everything. So it went through, it looked at my calendar, it looked at my Gmail, and then it went through and looked at my Twitter and LinkedIn feeds. As well. So I don't need to necessarily waste time.

    Jordan Wilson [00:23:41]:
    Right.

    Jordan Wilson [00:23:42]:
    Even though I do go on Twitter fairly often.

    Jordan Wilson [00:23:45]:
    Right.

    Jordan Wilson [00:23:45]:
    And I go there obviously to research to see what people are talking about in AI. And if I see, oh, this topic is hot, then I'll probably plan an episode around it after I learn a little bit.

    Jordan Wilson [00:23:55]:
    Right.

    Jordan Wilson [00:23:56]:
    But it can be a very time consuming task.

    Jordan Wilson [00:23:59]:
    Right.

    Jordan Wilson [00:23:59]:
    Even going through your email, all those things, it's so easy to get distracted. And I think this is almost a great way to start your day. I would obviously personalize this a little bit more, give comment a little more information on things you care about, things you don't.

    Jordan Wilson [00:24:16]:
    Right.

    Jordan Wilson [00:24:17]:
    But this is just a great example here. It went through. It found some of my meetings that I have on my calendar. It went through my Gmail sound, some, you know, some people reaching out.

    Jordan Wilson [00:24:28]:
    Great.

    Jordan Wilson [00:24:29]:
    Uh, social media. It went through, you know, here's kind of what's trending on social media. LinkedIn it said top trends include AI powered career coaching, advanced analytics for professionals. It went through some of my LinkedIn DMS. So great. So nothing, you know, nothing that I think is earth shattering. But right here I would go through and obviously improve this prompt a little bit. I wanted something quick that wouldn't take a super long time.

    Jordan Wilson [00:25:01]:
    But I mean, right there, just a huge time saver. All right, let's do another one. In this one we're gonna get a little meta here. All right, so what I'm doing now inside Perplexity comment is I said look in my ChatGPT account for a certain chat and I'm saying carefully comb through that information and find five highlights that I should go more in depth on. Then use Google Gemini to do more research on those five highlights. Here's a huge one that I love.

    Jordan Wilson [00:25:34]:
    Right.

    Jordan Wilson [00:25:35]:
    It can take me a long time to go through my ChatGPT account, my Google Gemini, my Claude, etc. But if I'm logged in in the browser.

    Jordan Wilson [00:25:45]:
    Right.

    Jordan Wilson [00:25:46]:
    And some of these things can obviously take a while.

    Jordan Wilson [00:25:48]:
    Right.

    Jordan Wilson [00:25:49]:
    But that's why I just have perplexity comment just kind of running.

    Jordan Wilson [00:25:53]:
    Right.

    Jordan Wilson [00:25:53]:
    At least the past. Yeah, I've been sick for a couple of days, but last week when I was demoing it a little bit, I just always have it running kind of in the background and then I'm doing my main work in another window or in another instance that's not Comet. So right now I just have Comet kind of running. So in this example. I know. So I'm actually doing a speech keynote next week on Agentic AI, right. So I've been doing a lot of research both, you know, going back and looking at old podcast episodes, smart people I've talked to, making sure I have the most up to date accurate information from July 2025.

    Jordan Wilson [00:26:29]:
    Right.

    Jordan Wilson [00:26:30]:
    So I, I had a, a research chat going inside ChatGPT so I can click open page and it looks like it found that.

    Jordan Wilson [00:26:37]:
    Right.

    Jordan Wilson [00:26:38]:
    Agentic trends in AI in 2025. So I'm having, and this is so meta, right? I'm having perplexity. Go into this chat, inside ChatGPT and you'll see here it pulled up the correct one. That's great. It was a deep research or. No, it wasn't a deep research chat. What, what was this one? Yes, it was originally it was a deep research chat I went through. I was having a lot of conversations with it, so I had perplexity go look at this deep research chat went through, pulled a lot of it and I'm telling it, hey, poke some holes, go find five more things that I should investigate a little further.

    Jordan Wilson [00:27:15]:
    All right, so let's see what it it found. So it went in, it told me, okay, I found all this information and then it said investigating expert insights and recent developments on key agentic AI topics for in depth understanding. So then what happened is it found five different things that it needed to investigate a little more. Agentic AI misalignment examples. That's a good one. Market size projections and funding in Agentic AI. 3. Platform ecosystem competition.

    Jordan Wilson [00:27:43]:
    That's a good one.

    Jordan Wilson [00:27:44]:
    Right.

    Jordan Wilson [00:27:45]:
    4. Policies and government use of Agentic AI in the regulatory climate. Five approaches to building trustworthy human aligned agents in auditability standards. That's great stuff. So not only did I not have to do that, right. It's probably something that I would have gotten to eventually and I'm still going to go through and manually. Yeah, you know, do a lot of this work. But now it's starting to, it found some holes or gaps or opportunities in my original ChatGPT research and then I'm clicking here inside Google Gemini.

    Jordan Wilson [00:28:20]:
    Let's see if it worked. All right, it looks like it's still going. It hasn't started yet, but I, I ran this through multiple times already and it's going to start researching those different topics inside Google Gemini. I mean, mind blown. Mind blown. Let's see. I think y', all, you get the gist here. All right, but, and I showed you a little bit on our AI at work on Wednesdays how I'm using perplexity comment, but I want to leave you all with a couple of things.

    Jordan Wilson [00:29:03]:
    All right.

    Jordan Wilson [00:29:03]:
    And I'm just double checking here. There we go. All right. I told you all that this was actually going to work. I, I just clicked back into, into the Gemini Deep Research.

    Jordan Wilson [00:29:14]:
    Great.

    Jordan Wilson [00:29:15]:
    And it did it. So it pulled out those five different holes or opportunities from my original ChatGPT research, opened up a new window in Google Gemini, and it's running a Google Gemini Deep Research. Y', all, the fact that this is happening right now, this is amazing. This is amazing.

    Jordan Wilson [00:29:36]:
    Right?

    Jordan Wilson [00:29:36]:
    We kind of got sniffs of this with OpenAI operator, with Claude, Claude's computer use agent, you know, a little bit with Project Mariner. But the fact that this runs locally in your browser, right. I, I'm running into so many errors when I'm using more of the computer using agents.

    Jordan Wilson [00:29:58]:
    Right.

    Jordan Wilson [00:29:59]:
    They're a little slower, though, a little less reliable just because the technology is not as solid as, you know, having this hybrid AI approach, having an agentic AI browser. This is huge. Like, I am still as this is happening. I'm like, this opens up so many new possibilities for how we could and should work. And let me just say this, this is coming for you whether you want it or not, right? Google's investing in this. OpenAI is going to be launching their AgentIC AI browser in the coming weeks. Microsoft in their edge browser, they're really expanding their copilot vision. And guess what? Once OpenAI releases their version of this, it's going to.

    Jordan Wilson [00:30:44]:
    Because the ip, Microsoft gets that as well. So Microsoft is going to be rolling this out. Google is going to be rolling it out. This is how we're all going to be interfacing with AI in the future.

    Jordan Wilson [00:30:56]:
    Right?

    Jordan Wilson [00:30:56]:
    So even though as a pot, like on the podcast, and I understand how this might be a little difficult to follow by not seeing everything. So make sure, you know, you go to our website, within a couple of hours of you listening to this, we're going to have the webpage up and you can go watch the video and see exactly what's happening here or on our YouTube channel. But you need to start rethinking how, how you work.

    Jordan Wilson [00:31:20]:
    Right.

    Jordan Wilson [00:31:20]:
    Because even on the AI chatbot side, I understand there's hurdles, right? Oh, my company, you know, we haven't approved this for everyone yet, or we're not using this in all departments or. Okay, we're only using this AI chatbot and it doesn't have access to our data. Sure, I get that. I understand that. This is your browser. This is happening whether you want it or not. So you have to really get in and start using these agentic AI browsers now because this changes how we use the Internet on the rest of that time.

    Jordan Wilson [00:31:55]:
    Right.

    Jordan Wilson [00:31:56]:
    There's so many things that I normally would manually do that now I'm going to be handing off to comments. So we really do have to rethink in the same way when large language models first came out, we had to kind of rethink how we spend our time, how the humans spend our time. This is the same thing, right? I think this is going to a agentic AI in the browser is going to replace a lot of current AI workflows, a lot of these. You know, I, I, I laugh when people call things like this agents, right? These NA, these N8N, you know, agentic workflows, even things in, in zapier and make, I think agentic AI browsers are going to replace so many of these other kind of quote unquote AI powered workflows or agentic workflows. Because when it's happening locally, when it has access to all of your data, that changes what's possible. All right, let's wrap this thing up, y'. All. So here's five different use cases and I showed you some of them, but I'm going to talk about the rest.

    Jordan Wilson [00:33:01]:
    Number one is personalizing multi step, multi platform research.

    Jordan Wilson [00:33:07]:
    Right.

    Jordan Wilson [00:33:07]:
    So I kind of gave you that example, right? My example looking through my, my, my Gmail, my calendar, my social media, etc.

    Jordan Wilson [00:33:13]:
    Right.

    Jordan Wilson [00:33:14]:
    But apply that to what you need it for, right? And I'm going to kind of combine number one and number four. So number four on my list is automated market research and competitive analysis.

    Jordan Wilson [00:33:26]:
    Right.

    Jordan Wilson [00:33:27]:
    You have to give perplexity more context.

    Jordan Wilson [00:33:31]:
    Right.

    Jordan Wilson [00:33:31]:
    Because I'm not trying to spend an hour and a half on this show. I didn't go through and do that.

    Jordan Wilson [00:33:35]:
    Right.

    Jordan Wilson [00:33:35]:
    I was just giving you very topical, very hopefully easy to follow. But the more information you give, perplexity. Hey, here's who I am, here's what I do in my role, here's my competitors, right? Save that in a doc for now and just paste that in when you start a new chat inside the Comet AI assistant. And by the way, that's on the right hand side. I should have explained that for our podcast audience. If you're looking like the assistant is on the right hand side and then on the left hand side, that's where the normal browsing happens. So you need to inform the comment assistant a little bit more information about yourself, but then it can go ahead, take advantage of all the information in your tabs, anywhere that you're logged into, it will go and visit those websites. So number one is personalizing multi step, multi platform research for you.

    Jordan Wilson [00:34:22]:
    Number two, save time on social media.

    Jordan Wilson [00:34:24]:
    Right?

    Jordan Wilson [00:34:24]:
    A lot of people are obviously on social media for work, but social media is a black hole, right? So just having it pull out specific business related insights. Again, personalized for you. Number three, help manage your email and calendar. It can actually create calendar events. It can scour through your email.

    Jordan Wilson [00:34:46]:
    Right.

    Jordan Wilson [00:34:46]:
    If people are trying to, you know, set up meetings with you and just be like, hey, who's setting up meetings with me? Which of these times do I have available? Right. So it is almost like having an executive assistant once you are, you know, comfortable with using it in that matter. Number four, I already mentioned automating market research and competitive analysis, y'. All. This is such a great way, I think if you kind of have these, these kind of prompts or use cases set up, whether you run them weekly, daily, etc, I like the way I'm using this. It's almost like I'm gonna have some of these prompts or some of these use cases ready when I wake up, I'm gonna go put them in, click enter, go grab my Nespresso. Which Nespresso? Why are you not advertising yet on the show? Come on.

    Jordan Wilson [00:35:30]:
    Right?

    Jordan Wilson [00:35:31]:
    And then in four to five minutes I come down and it's like I've done at that point 2, 3, 4 hours of work, right, that I would have been doing manually before. That's huge. It's just such a great productivity boost to start your day. And then number five, which I showed you an example of that is using multiple AI tools at once autonomously. I gave you all an example of that. I was literally using Perplexity's comment browser to go find information from my ChatGPT account. It was analyzing, finding holes and opportunities and then it went out and did deep research research on Google Gemini to help fill in and augment some of my initial research. Y', all, you can have a day's worth of work done before you sit down and sip your first coffee.

    Jordan Wilson [00:36:27]:
    And if that isn't an AI at work Wednesday, I don't know what is.

    Jordan Wilson [00:36:32]:
    All right.

    Jordan Wilson [00:36:33]:
    Hey, look at this. There we go. My Gemini deep research report is done. Number five, having Perplexity comment use multiple AI tools on its own. Amazing. All right, I hope this show was helpful, y'. All. If so, please repost this to your network.

    Jordan Wilson [00:36:52]:
    We spend a lot of time trying to bring you the most up to date reliable and accurate and unbiased information as possible, even when I feel like complete chicken trash. So please, if this is helpful, repost. Tell someone about it. You know, tell your company, email this to them. Have everyone you know then go to your everydayai.com Sign up for our free daily newsletter. So thank you for tuning in. Hope to see you back tomorrow and every day for more Everyday AI. Thanks y'.

    Jordan Wilson [00:37:19]:
    All.

    Gain Extra Insights With Our Newsletter

    Sign up for our newsletter to get more in-depth content on AI