Ep 647: The New Secret Google Gemini Feature that Quietly Kills Powerpoint slides
31 Dec 2025 · 19 min
Jordan Wilson explores Google Gemini's Canvas mode, a feature that lets you build presentations without PowerPoint. It works faster than Microsoft Copilot and handles design automatically, though it has limits on what it can create.
Chapters
This is the Everyday AI Show, the everyday podcast where we simplify AI and bring its power to your fingertips. Listen daily for practical advice to boost your career, business, and everyday life.
One of the most important elements of winning back time with generative AI is not going out of your way. You need to match the best AI tool or mode with the work that you're already needing to do. And let's be honest, presentations are still one of those things that we're spending so much time on. And there's great AI tools to make slides, but they also have their limitations. Like Gamma is great, but I find it to be unpredictable at times, especially with buggy PDF and PowerPoint exports. And Microsoft Copilot obviously has some great AI features inside of PowerPoint, but so few people actually have access to most of Copilot's features because of data privacy and security. So will a new AI slide champion enter? Well, Google Gemini, it might be you, uh, because there's a new secret Google Gemini feature that might quietly kill PowerPoint slides. But it's technically in Gemini, but you gotta technically be in Canvas mode, and you technically have to use the right words, but Don't worry, I'm gonna tell you exactly what you need to know. It's gonna be a fast one, and we're gonna put AI to work for us on this Wednesday. What's going on, y'all? Welcome to Everyday AI. My name's Jordan Wilson, and this is your daily livestream podcast and free daily newsletter helping everyday business leaders like you and me not just keep up with AI, but how we can make sense of all these changes and advancements, use them to get ahead, to grow our companies and our careers. That's what you're trying to do. Starts here with the unedited, unscripted livestream podcast, but to take it to the next level, if you wanna be the smartest person in AI at your company, go to our website, youreverydayai.com. There we obviously break down the highlights from each and every, uh, podcast episode of that day, but also we keep you up to date with all of the other latest AI news that you need to know. Let's get into it. I'm not gonna make you wait any longer. There's 3 words that you're going to need. Create a presentation. All right. But there's a lot more that you need to know. So I'm going to show you how to use this new feature inside of Google Gemini. And at the end, I'm going to give you 3 tips on how to best use it. But on today's show, we're going to learn how and why to use Gemini Canvas for all types of presentations. We're going to under— we're going to uncover and explore the that 3-word phrase, create a presentation that can kill off maybe some of your PowerPoint designing. And we're going to learn the pros and the cons of using the new slides feature in Gemini Canvas that's kind of hidden but extremely powerful. Let's get into it. Let's learn live. Livestream audience, love to see you. Good to see you. Thanks for joining. Bronson, thanks for joining on YouTube. Rolando, Jay, Juliette, everyone else, good to see you. Robert, let's get straight into it. This is gonna be one of those ones, not necessarily a visual show, but if you do want to follow along, make sure to go to our website, youreverydayai.com. All right, so we always have the video version there. Live stream audience, do me a favor. I'm going to be jumping around a little bit, um, on our, uh, live stream here. So if you could let me know if you see my screen. All right, uh, before we get started, so here's what we're going to do. I'm going to give you a primer on how Canvas mode in Google Gemini works, and that's where you are going to use that 3-word phrase, create a presentation. Yeah, there is no toggle or button that you click, which can be a good thing. Maybe it's a bad thing at first, but if you want to take advantage of this amazing feature, that's how you do it. All right, so here's exactly what we're going to do, and I'm going to do this live. On Wednesdays, we do this thing called Putting AI to Work. On Wednesdays, we do a usually live demo of a newer AI feature. Usually from one of the big four players, right? Microsoft Copilot, Anthropic Claude, Google Gemini, or ChatGPT. So, you know, kind of demoing new features. And this is one, honestly, this is worthy of a keynote, but Google literally not even on their main Twitter account, on their Google Gemini app account, they put out one tweet and that is it. And this is one of those small little features that I think everyone needs to be using. All right. All right. Hopefully, hopefully there's a delay. I didn't see anyone say that they see my screen, but I'm just going to hope and assume that y'all can. So here's what I have. I have yesterday's podcast episode. All right. It went a little long. So today's episode, we're going to be a little short. It's going to be a short and sweet one today. So this was how can a former nonprofit losing $12 billion a quarter Go public at $1 trillion. So this was a story, uh, about kind of our Hot Take Tuesday episode about OpenAI. So on our website, if you didn't know, you can always go listen to the, um, the podcast episode on our website. You can also watch the video. All right. And then we, uh, break down some of the topics, have a little write-up, uh, and then an entire transcript of the actual episode. So I'm going to just copy and paste this entire thing. Okay, now I am inside Google Gemini. All right, thank you, Joe, Marie, and Jay, for letting me know you can see the screen. So in Google Gemini, you're going to click the new— or not the new, there it is, an updated layout, but this is not a new button. So you're going to click on Tools and then go to Canvas Mode. So I'm going to just make this very easy. All right, I'm going to just say, create using Canvas, highlight the main points of this podcast and create a presentation. All right, and then I'm going to copy and paste it. So that's all I'm saying. I'm saying using Canvas, um, highlight the main points of this podcast. Love typing live. And our 3-word phrase, create a presentation. All right, I've tried, uh, other combination of words. Some work, some don't. You know, uh, you know, create a Google Slides file, create a, uh, slideshow. Sometimes it works, sometimes it doesn't, because Canvas mode— and I'll explore it, uh, just a little bit with you all here— it can actually create a lot of different things. So, uh, I'm letting this cook right now. It should just take 2 minutes, but that's literally all I did. I just said, using Canvas, highlight the main points of this podcast and create a presentation. And then I pasted the entire content of the podcast, um, that's on our episode page. That's it. All right, so our livestream audience is gonna see this happening live. If you're on the podcast, I will walk you through, uh, what's happening, but, uh, it's creating it now. Okay, so I can see it kind of building the slides. It split my screen on the left-hand side. I could continue to chat with Google Canvas mode, but on the right side, you will see now it says creating slides. And if you've listened to the show at all, you know one of my, what I think is probably the most underrated feature or mode out there is Gemini Canvas. So yes, it has some similarities with OpenAI's Canvas mode and some similarities with Claude's Artifact. Mode. The difference is I think it is great at building interactive websites visually. That's what I use it for. I use canvas mode all the time to put together just visual presentations, right? But up until now, you haven't been able to actually export those as slides. All right. And you'll see now it's done. That was quick, right? We can end the show at 8 minutes. We're not, because I'm also going to give you 3 tips on how to best use this feature, but you'll see it's already done. Okay. 12 slides completely done. So it, it named it the $1 Trillion Paradox. Maybe that's what I should have named the actual episode. And then it says how OpenAI, losing $12 billion a quarter, preparing for a $1 trillion IPO. All right. So now on the right side, I can toggle through these slides. And you'll see for our livestream audience, pretty good, right? They're designed fairly well. It has a nice little heading for each of our slides using some bright colors going in a dark mode that I really like. It broke out— love this, didn't even tell it to— it broke out essentially a cap table on OpenAI's recapitalization because in the context of the episode, I specifically said Microsoft 27%, OpenAI Foundation 26%. It took a little leeway with putting the other 47% because there's actually, uh, a little bit more of a breakdown. But if I would've named off the other percentages, it would've put that together. But it accurately kind of created this nice little cap table on the new ownership structure of the public benefits corporation, uh, of OpenAI transitioning from a strict nonprofit to a public benefits corporation. It created a table or a chart there. All right, then we have a slide, very nice looking slide here, FYI, uh, with 3 little breakout boxes with kind of subheadings, and each of them has a, a little icon. Looks good, right? So it says, uh, the user-first strategy. This is one of the things that I broke down. It says 800 million weekly users, 70% of their revenue comes from consumers and ecosystem stickiness. So not only did it do a good job of creating very usable slides, FYI, it did a great job of consolidating and organizing all of the information because, yeah, yesterday's podcast was an accidentally kind of long one. So it cut through a lot of my fluff. Yeah, I know there's fluff in the show, y'all, and it created a pretty good looking pretty good-looking presentation here. Okay. It also, here we go on slide 9 here, it put together a kind of a timeline on some of the other startups that have gone many years without being profitable, right? OpenAI is technically on year 4 of not being profitable, but on yesterday's show we talked about Amazon, Spotify, Tesla, you know, taking between 9 to 17 years, uh, to become profitable. So it created a nice little timeline here visually. So, uh, pretty good. It did put some placeholders in here. And again, let me say this, generative AI is generative. I ran this exact same, uh, prompt verbatim like 2 or 3 times last night, 2 or 3 times today. Sometimes it will actually create images for you using Google Gemini, right? Sometimes it will give you placeholders with a little more prompting and a little more iterating. It, you know, you can obviously update it accordingly. All right. That's the gist, y'all. I'm going to get into some advanced tips here and 3 best practices, I think. But that's it. Go into Google Gemini, uh, you don't even need a, uh, paid plan to do this. Google Gemini, click tools, click canvas, use those 3 magic words, and then you have the options to export to slides. All right. And then that will open that presentation and I'll go ahead and do that so you can see what it looks like. Uh, you can export it to Google Slides or you can instantly download it as a PDF. And then obviously, you know, if you do, right, like if you are a PowerPoint organization, you can obviously open this in Google Slides and then export it or convert it over to a PowerPoint file if you ultimately need to make some updates inside of PowerPoint, right? If you are sharing that, you know, in your team, in your organization. So let's see. There we go. So, little UI/UX thing. All right, if you're like, wait, where's my slides file? It's a very little thing on the lower left-hand corner of the screen if you're looking for it. The first time I clicked the export to slides, I'm like, okay, well, where is it? Okay, it's in the bottom left-hand corner. Click that little open slides and there you go. All right, so now it is create, it's created that file, it's exported it to Google Slides. I can make any, uh, changes here. All right, now you might be wondering, why not just do this in Google Slides? Why do it in Gemini? Well, let me tell you, or, you know, why not, uh, yeah, why do it in, uh, Google Gemini's Canvas mode? And why not just do it in, uh, Google Slides and use the Gemini option? Well, let me show you. I'm going to put that exact, exact same prompt that I used. Okay, let me just go ahead and copy and paste it and you'll see what I mean. All right, so I'm now in Google Slides. I'm using the Gemini integration and you'll see here in a few seconds and I'll spoil it for you. It's not good. It's not good at all, right? Google, to their credit, has really improved the Gemini integration across their workspace products, but some are great, some aren't. Okay. The Gemini integration in Gmail, hit or miss. Gemini integration in Microsoft Word, pretty good. Gemini integration in Google Slides, Really good. Gemini integration in Google Vids, amazing. Gemini integration in Google Slides, useless, right? So if you're wondering, oh, well, I'm just gonna use the Google Slides, don't. It's not good, right? In this case, it, you know, didn't even work. Last time, it at least gave me one or two ugly-looking slides. This time, it didn't. All right. So let's get into the 3 tips that you need to know to make this better. Okay. So kind of some 3 expert tips and a livestream audience. If you guys do have any questions, go ahead and get them in. Now we are literally going to make this a fast and furious episode. So number 1, start with a screenshot to cut down on design time. I'm going to, actually, let me give you the 3 tips. Number 1, start with the screenshots, cut down on design time. Number 2, understand the capabilities and limitations, right? So as an example, you can't upload photos to Include, even though Gemini will tell you it can. You can't. All right. You can't include image source links. Uh, you can do that in Canvas. You can't do it. In the Slides integration in Canvas, but Canvas in Google Slides, or, or sorry, the, uh, Google Slides feature inside Canvas can create great looking charts, graphs, tables, breakout boxes, et cetera. Tip 1, start with a screenshot. Tip 2, understand capabilities and limitations. Tip 3, iterate in normal Canvas mode first. With the select and ask feature. All right, so I'm going to go ahead and kind of reverse engineer this and show you exactly what I mean. So I'm going to do something a little different here. All right, I'm going to open a new chat here in Gemini. I'm going to go to canvas mode. I'm going to copy and paste this, and I'm going to tell you what I'm doing differently. Now, here's why you need the 3 magic words. I said, using Canvas, highlight the main points of this podcast episode and create dramatic visuals. All right. I'm doing this for a reason. What's going to happen now? Well, it's going to create probably an interactive, like, website. It is not going to create a presentation. Okay. So I'm going to give it a minute and I paste it in the exact same content. And I'm going to show you now. All right. It's going to take a minute, but on my screen, Canvas is doing what it's really good at, which is writing and rendering code. All right. That's what Canvas mode is really good at. So now I'm going to actually just iterate on this real quick.
AI moves too fast to follow. but you're expected to keep up. Otherwise, your career or company might lag behind while AI-native competitors leap ahead, but you don't have 10 hours a day to understand it all. That's what I do for you. But after 700+ episodes of Everyday AI, the most common questions I get is, where do I start? That's why we created the Start Here series, an ongoing podcast series of more than a dozen episodes you can listen to in order. It covers the AI basics for beginners and sharpens the skills of AI champions pushing their companies forward. In the ongoing series, we explain complex trends in simple language that you can turn into action. There's 3 ways to jump in. Number 1, go scroll back to the first one in episode 691. Number 2, tap the link in your show notes at any time for the Start Here series, or you can just go to starthereseries.com, which also gives you free access to our inner circle community where you can connect with other business leaders doing the same. The Start Here series will slow down the pace of AI so you can get ahead.
All right.
Didn't give me quite enough info, so I'm just saying, please include more info. Please try again. But what it did is it created almost like an interactive website with the same content that I pasted in using canvas mode. I wanted it to have a little bit more information. Usually when I do a simple prompt like that, it will include a lot more. So it's going to rebuild this. So an example, you can do the exact same thing. This is the iteration or iterating in natural language with your presentation. So you can just say, make this more visual, include more information. Hey, I pasted in a lot, but you didn't include A, B, and C. Please include it. So iterate. With your slides once they are done. Say in slide 6, change this. But a little tip, here we go. All right, so first I just did a normal kind of website presentation. So use this feature. There is a Select and Ask feature in the bottom right-hand corner when you are using normal canvas mode in Google Gemini. Okay, this is not going to show up once you are kind of in slides mode, but the trick is to use it in the normal canvas mode first because it's an extremely powerful, uh, little tool. And then you can create the canvas mode— or sorry, the, uh, the slides mode once you've used it. So as an example, uh, I'm going to use this Select and Ask feature. It broke out, uh, one section and kind of created 3 separate boxes. All right. Product velocity, revenue levers, and infrastructure deals. All right. From the podcast episode yesterday. So I can highlight that and then a little box pops up that says, ask Gemini. All right. And I'm going to say it made 4 bullet points and it's not the best layout. So I'm going to say, please add icons to the top of the boxes. And limit to 3 bullet points only. Okay, so this is a very powerful feature. So if you've ever had to iterate on anything in any large language model, the downside is generative AI is generative AI. So you can try the best with your words and say, hey, on this paragraph that says A, B, and C, make X, Y, Z changes. But sometimes it's going to change the top, the bottom, and everything in between. So with this select and ask feature, you literally drag a box over just the section that you want to update, give it changes in natural language, and then most of the time it is only going to update those things. So let's scroll down. Bam, we got it. It worked. Yay! A live generative AI demo worked great. Okay, so that's kind of one of the three tips there. Technically two, because I know the capabilities and the limitations as well. So I know As an example, one thing that Gemini Canvas can do really well is it creates nice looking breakout boxes with icons, right? Which translate really well to slides. All right, so technically we did two of our advanced tips. All right, we used some of the normal Canvas features before kind of quote unquote going into slides mode. We understood the capabilities and limitations, and then I'm going to iterate with a screenshot. So I have a screenshot here. Of kind of our normal slides layout. Okay? And I'm just going to drag that in. Sorry, livestream audience. Lot of jumping around here in my different tabs. Okay? So I'm just going to drop that into the chat window on the left-hand side. And all I'm going to say is, this is our normal style. Please update the presentation accordingly to reflect this style. So what's going to hopefully happen here, um, it's going to kind of take into account, um, kind of the design choices. My livestream slides are ugly, it's fine. Um, you can tell me, y'all, this is for, you know, to help keep me on track. Um, but it's going to take in the colors, the layout, and even the— it's going to try to match up fonts. So you can't upload a slide deck to start with, uh, and then say, you know, convert this to a Google Slide, right? So that's not what it is. But you can upload multiple examples of how you would normally build slides, right? Whether there's a certain layout. My, I guess my brand on slides is ugly, uh, right? So it's probably technically gonna make this uglier cuz, uh, the version that it did was actually pretty good. And so it's it's probably gonna make a worse visual. But if I want consistent, uh, visuals, if I want consistent branding, uh, right across, uh, the slides that I ask Canvas to create, uh, this is the way to do it. So it is iterating. And FYI, y'all, as with anything else, there's, there's going to be this, uh, as you wait, even though it's not a ton of time, it gives you the chain of thought. So click on that and always read what's happening. Because as an example, sometimes large language models, especially new features, are buggy. They don't always know what they're talking about. They don't know their capabilities, right? Because as an example, you can upload images in normal canvas mode. You can't do that when you go into slides mode. Okay, so bam, we're done. It created— it did a really good job of instead of, you know, making the really good-looking slides that it had created on its own with Gemini Canvas's style, it just made them uglier now. But it matched, right? But it matched exactly what I uploaded, right? It put kind of the bluish heading, the reddish kind of subhead, and then the big white bullet points and the call to action at the bottom. So it did a pretty good job of just changing based on the information that I uploaded. All right. So, yeah. Okay. Juliet says the colors are muted. I wouldn't call them ugly. I don't know. All right. I'd say it's a little ugly. All right. Jean says, can Gemini Slides make a presentation from a YouTube clip URL like one of your episode episodes, for example? Kind of. Kind of, right? So it depends on what version of Google Gemini you have. So as an example, not all Workspace accounts, accounts have the essential YouTube app integration inside Gemini Canvas. Like for that, I would have to use my personal account. I have multiple paid accounts, but technically, yes, you can. Right. So inside Canvas, right? Like the same with any large language model, you can take advantage of the context window. So yes, Google Gemini in general does a great job of integrating with your Gmail if you enable those apps and extensions, right? So if you want to know how to access that, let me just go ahead and show you. You would go into your settings, you would go into apps, And then you can toggle these apps on and for— yeah, okay, good. I was right. I'm like, man, I hope I don't get this. I hope I don't get this wrong. So yeah, for whatever reason, Google Workspace accounts don't have the YouTube app integrated inside Google Gemini. I don't understand why, but they do have Asana and MailChimp, right? But they also have Gmail, Google Docs, Google Keep, Google Tasks, Google Drive. Google Calendar. Right. But, Jean, a great question. Use— take advantage of all of the features that a large language model has, specifically the context window. Right. So you could, as an example, pull in information from all of your different Google Drive files, your Google Calendar, your Gmail. Right. If you're putting something together for a client, as an example, it can go look at your calendar, say, oh, you have 3 meetings. Here's what the meeting's covering. Let me see. Oh, It looks like you're, you know, going over Q4 KPIs. I see that in your calendar notes. Let me go look through your Gmail. Oh, looks like there's some documents I need to reference. Yes, you can do that inside Google Gemini. Keep that in the context window and then use Canvas and that 3, the magic 3-word phrase, create a presentation. And then, yes, it will create slides. All right. Uh, yeah, Jackie, Jackie here echoing what I said. Uh, she said, 100%, Jordan, when I wanted to add a slide in Google Slides, it sucked. It's not that good, right? It's not. But it's actually legit really, really good inside Google Gemini. So now you all know the cheat code. That's it, y'all. Actually, no, I want to ask you, what should we cover next week, right? As we wrap, uh, On Mondays, we do the AI News That Matters. We go over usually the top 10 AI news stories of the week that impact your company or career. Tuesdays, we kind of do Hot Take Tuesdays, you know, kind of my opinion on something that's happening. I did my opinion on the whole OpenAI restructure and $1 trillion IPO. And then on Wednesdays, we do this AI at Work on Wednesday. So livestream audience, let me know, what should we tackle next week? Also, podcast audience, if you're listening on Spotify, go scroll down in the show notes. There's gonna be a poll today. I'm gonna be including more polls, uh, inside of Spotify. So what should we tackle next week for Putting AI to Work on Wednesdays? Claude skills, really great new feature people aren't talking about. Uh, NotebookLM updates. There's actually been quite a few, uh, NotebookLM updates, different video styles, context windows, a ton of things. Or going over, again, these are 3 low-key updates that are pretty big. The 3rd one, ChatGPT's company knowledge. So this is only available if you have a business kind of team plan or an enterprise plan, but there's actually new connectors and some new features and functionalities if you do have a team plan. So go vote, livestream audience, let me know. Just say skills, NotebookLM, company knowledge, and that's it. I hope today's show was helpful, but here's the reality. If you want to save time with generative AI, you need to use the right tool at the right time for the right purpose. And I think regardless of what you work on, so many people have to create presentations. And like I said, yes, there's great AI tools that create, are dedicated to creating AI, you know, slides, presentations. They can be buggy, and a lot of them require a separate subscription. PowerPoint has completely dominated the presentation scene for decades, yet I know probably less than 5% of people that I talk to who have Microsoft 365 Copilot license actually have access to all of Microsoft Copilot's features, including creating, uh, PowerPoints with AI, with Copilot. So I think In this case, Google Gemini can be your go-to tool. Take advantage of the power of Gemini 2.5 Pro. Take advantage of the context window. Take advantage of everything that Google Gemini can access in terms of your apps and all of its capabilities. And then use that magic 3-word phrase, create a presentation. I hope this was helpful. If so, go to youreverydayai.com. If you're listening on the podcast, I'd appreciate it if you subscribe to the podcast, follow us, leave us a rating. Thanks for tuning in. Hope to see you back tomorrow and every day for more Everyday AI. Thanks, y'all.
And that's a wrap for today's edition of Everyday AI. Thanks for joining us. If you enjoyed this episode, please subscribe and leave us a rating. It helps keep us going. For a little more AI magic, visit youreverydayai.com and sign up to our daily newsletter so you don't get left behind. Go break some barriers and we'll see you next time.
More from GHION.AI
20VC: Is Chamath Right: Is DPI The Only Thing That Matters | Does OpenAI Even Matter | Mary Meekers AI Report: The Analysis| IPO Breakdown: Chime, Circle & Thoma Bravo's New Fund