Gemini's New File Feature Tested - What Actually Works

Google quietly added a feature to Gemini this week that lets it create files for you. PDFs, Word Docs, Excel Sheets, Google Docs, Google Sheets, Google Slides. You ask for it, Gemini makes it. On the surface, that sounds like a small thing. The kind of update you kind of nod at and move on from there. But 6 days before that rolled out, Google announced something at Cloud Next that puts this whole update into a different light. I'll come back to that at the end of the video because once you understand what's actually happening here, it changes how you should think about Gemini going forward. First, let me show you what it does, where it works well, and where it falls apart.

This update went live on April 27th, and it's already rolled out worldwide. The important thing to know up front is there's no special plan required. Free Gemini users get this, paid users get this, everyone gets this. When you log into gemini. google. com, you don't see a new button or a new menu. The interface looks identical. You just type what you want and Gemini figures out what kind of file to make. The supported file types are Google Docs, Sheets, and Slides on the work side, workspace side. On the downloadable side, you've got PDF, Word, Excel, CSV, plain text, rich text, markdown, and LaTeX. There's one notable absence from that list, and I'm going to come back to that in a few minutes because it actually matters more than what you'd expect. All right

let's start with the simplest version of this. I'm going to paste in some rough notes from a staff meeting. Just bullet points, nothing formatted, the kind of mess most of us end up with after, you know, we're taking notes in a hurry. So, I'm going to take these into Gemini and just give them here. And then I'm going to actually add this to the end here. So, all I'm going to say is uh take these notes and turn them into a one-page parent newsletter as a PDF. Use a friendly tone, clear section, headings, and no jargon. So, notice what comes back right here. I'm going to open this up and notice this PDF is already formatted. Headings, spacing, paragraph breaks, the kind of thing that I normally spend 10 minutes cleaning up on Word. I can preview it right inside Gemini by just clicking on that PDF card and I can download it as well just by clicking up here. Now, here's the part that most people miss. If I want the same content but in a different format

I don't need to start over again. If I just go back to this and say, "Now, give me the exact same content as a Word document so I can edit it later. " Okay? And look what we get. We get the same content in a different container. That's really the whole point of this feature. You generate work once, then move it into whatever format you actually need. And you'll notice the difference between here is just the editable one where I can download this as a Microsoft uh Docs document here and then edit it later versus the PDF that was already formatted in kind of a nice clean uh graphical way. Now, in this next demo, it's where things get more interesting because

this is the piece that genuinely separates Gemini from Chat GPT or Claude for most everyday users. So, I have this spreadsheet right here, and it's just in my Google Drive. It's tracking performance data for videos I published last quarter. This is just a fictional sheet, but just to show you how it works. So, if I want AI to summarize it, I'd have to open the AI spreadsheet, copy the data, paste it in the chat, and then ask my questions. I'm not going to do that. This is what I'm going to do instead. I'm going to give this prompt. Find my Q1 tutorial performance sheet in my drive. That's what I called it. Give me a one-page PDF executive summary with all the three highest performing videos, three lowest, and a chart comparing watch time across them all. Let's go ahead and submit. Okay, it gives me a little bit of summary here. It's even telling me uh asking me if I want to upload this to my drive. But let's go ahead and open this here and take a look at this. Okay, so it pulled the data. Let's see. It built the chart. I did only ask for one page, but that would have been kind of squishy, I guess. uh it formatted it and it gave me this downloadable PDF. You know, I never opened the spreadsheet. I never copied a single number. Gemini read the file directly out of my drive and turned it into something completely different. Now, that cross app behavior is where Gemini is reaching into your Google Drive and works with what's already there. It's actually a significant part of the update. Now, hold on to that idea because it connects to the bigger picture and I'll get to that at the end.

Now, here's a feature people have been waiting for a long time is just the ability to make full Google slideshows right inside Gemini. And so now, if I just give it the prompt, and I'll just stick within this one right here. Now, take the same Q1 video data and turn it into Google Slide presentation. Six slides, clean, modern design with a chart on slide three. So, let's go ahead and submit this. Okay, let's open it up here. And you'll notice it's not in Google Slide format yet, but it's easy to get there. Now, I can go through and take a look at it as I go through. I might go through and ask it to do some fixes. I can continue the conversation on right here. But if I wanted to Google Slides, all I have to do is export to Google Slides. And I wait a second. And then you'll notice down here it will just say open in Google Slides. So, I just click on it. And you'll see it will keep the same format here. And as I go through, I have this Google slides that it created. And even though I asked for six, it gave me seven because it gave me a source slide at the end. But what about PowerPoint? Okay, now I want to show you where the feature breaks. Uh

one of the limitations that the announcement post does not make obvious. So this is the part you need to know before you go try it yourself. So watch what happens when I try to ask Gemini to create a PowerPoint. So I'm going to give this the prompt here. Make me a PowerPoint presentation about the top three videos from the report include images and clean modern design. Let's submit. You're going to see one of two things happen. Either Gemini gives me a Google Slides deck instead of a PowerPoint or it dumps out code and tells me to build it myself. Neither of those is what I asked for. Okay, let's hit open. So, this I can export to slides. It made me a presentation. Nice present looking presentation. But this isn't in PowerPoint. If I was going to download it, it's going to download it as a PDF, not a PowerPoint. Here's why. PowerPoint, the actual PPPTX file format is not on the supported list. Gemini can't directly create a PPTX file. The supported list includes Google Slides. But if you specifically need a PowerPoint file, the workflow is two steps. generate a Google Slides deck. Then go to Google Slides and use the file download and pick Microsoft PowerPoint from the menu. If you've already tried this and gotten frustrated wondering why your PowerPoint prompts aren't coming back, this is why. It's not your prompt. It's the feature that isn't there yet. I expect Google would add a direct, you know, PPPX export some point, but as of today, it's not around. All right, one more limitation

and this one's bigger than it looks. Once Gemini creates a file and saves it to your Google Drive, you'd reasonably expect it to be you'd be able to ask Gemini to edit it, but that's not how it works. And let's go back to this file right here. So, uh, this is the newsletter one. It's in my drive. I actually downloaded it and put it in my drive, but I'm going to go into the same thing. So, if I ask it to take the parent newsletter and shorten the headline, remove the third paragraph, what I get is a brand new version of it. I didn't ask it to create a new PDF that time, but if I did, that would create another version of it. So, anytime I would make an edit, if I'm going from that file, that PDF from my drive, it's going to create a new version. So, Gemini can read existing files in your drive, and it can create new files, but it can't yet edit a file that already exists. Now, that sounds like a small annoyance, but hold on to that because this is the thing that actually is going to matter most when I get to the bigger picture next. Okay, so this is the part I promised

at the start of the video. I'm being straight with you. Gemini can make a PDF. It's not a feature that just justifies a 10-minute video on its own. Claude's been doing that since September of last year. Chat GPT has been doing it even longer. Google is catching up here, not breaking any new ground. So, why am I making a video about it? Well, because six days before this update went live, Google held its Cloud Next 2026 conference and at the event, they announced something called Workspace Intelligence. The short version is this. Workspace intelligence is a layer that sits underneath all of Google Workspace, Gmail, Drive, Calendar, Chat, Docs, Sheets, Slides, and gives Gemini a real-time understanding of what you're working on across all those apps at the same time. The pitch is this. You stop having to feed Gemini context every time you ask a question. It just knows what you've been working on, who you've been emailing, what's on your calendar, and what's in your files. The file generation update I just walked you through is the first piece of that strategy that regular users actually get to touch. Remember that second demo where Gemini pulled data out of a drive spreadsheet without me explaining what was in it? So, that's workspace intelligence starting to show up in the consumer Gemini app. and that version two duplication problem I showed you in the last demo. Now, that's the gap Google still has to close. Right now, Gemini can read your drive and create new files in your drive, but it can't yet act on those files that are already there. It can't truly work where your work lives. Tools like Claude Co-work and OpenAI's codecs are already starting to do that in the respective platforms. Whichever AI closes the gap first, doing the actual work inside the place where you keep your stuff, not a separate sandbox you have to copy things into, becomes much harder to walk away from. So, the real story here is not a new

feature. The real story is that Google's finally moving Gemini from chatbot you talk to toward an assistant that works inside your Google Drive. This update is the first visible step. There's going to be more coming. If you live inside Google Drive day-to-day if your docs, your sheets, your slides, your work is all in there, this update is worth trying today. The Driveaware prompts are genuinely useful, especially for anyone who deals with spreadsheets, reports, or shared documents. If you don't use Google Drive that much, this is more of a watch this space moment. The foundation matters more than this individual feature. I've got a follow-up video coming that walks through workspace intelligence specifically, including the features that are only in the available on the paid workspace plans. If you want to catch that one when it comes up, just hit that subscribe button. Thanks for watching this time. I'll see you in the next one.