Google Just Changed Image Editing FOREVER [Gemini 2.0 Experimental Demo]

Google just changed the game. You can now be like a Photoshop master with just typing in what you want the image to look like, and it's all for free. Hey, everyone, I'm here with some exciting news.

Google has kind of under the radar, rolled out some of the most valuable and interesting AI updates of the year, and the biggest one is pretty mind blowing. Google now for free, will let you edit any image with just text. It's pretty wild.

And today I am going to show you exactly how you can do it and show you some of the use cases that you might not have thought of yourself. Okay, so first, I want to talk about what the heck is actually happening. And you'll see friend of the pod, Logan Kilpatrick.

He's going to be joining us on the show soon. Google wrote out deep research is now free to all Gemini users. It's powered by a new model, 2.

0 flash thinking, which is reasoning model, which makes it kind of on par with ChatGPT deep research, but that show is coming soon. The best news, the best release, is now with Google's new flash model. It's bimodal, so that means it's text and images.

You can now edit images with just text. You used to have to go to Photoshop or Canva and do that manually, and now AI is able to edit your images for you. And here's some example.

Make the stake in this picture vegetarian, and it swapped the steak out to broccoli, tomatoes, carrots, pretty good, right? Make the bridge go away in this landscape, you just see the Marin Headlands here without the Golden Gate Bridge. Pretty wild, like you're like, wait a second, that would take me a while to both learn how to do and then do an app like Photoshop or Canva.

And now I can do it with just typing in a request, and I can do it for free. Pretty incredible. So one of the most immediate use cases people came out, which I thought was pretty interesting.

We've all if you've tried to get your passport before, you know one of the worst parts is about going to like Walgreens, CVS, triple A Walmart, one of these places to get your perfectly square, formatted passport portrait. And so what somebody did in Gemini was say, hey, take the blonde haired woman in this picture on the right with a neutral face expression on a white background, and make it square, which are basically the requirements of a passport photo, and boom, right there. Like that saves everybody a trip to Walmart photo.

But more importantly, it kind of demonstrates the power that if you're a business owner, you're a marketer, and you kind of don't have that perfect image, and you've been trying to maybe generate an AI image, but that's not going great either. You can now take an image that might be close and do some basic editing with the new Google Gemini, 2. 0 experimental, and I'm going to show you how to do that right now.

Look on this show, we have preached the good word about how AI is completely changing the game for content, for creators and for marketers, but we've never given you the exact steps you need on how you can change your content creation workflow. That is until now. We just dropped a step by step guide on which AI tools you can use today to start 10x ing your content creation workflow.

Learning AI for content creation has never been easier. You can steal our entire process right now, link in the description below. Now back to today's show.

You can do this in Google gym night, but I find that it's better and easier to do this in Google's AI studio. And so you're gonna want to go to AI studio. google.

com, or simply just search Google AI studio, and it's going to bring you here. And this is, this is their more kind of technical developer experience. But you don't need to be technical.

You don't need it to be a developer for this. When you land, you'll get a little pop up that says, Do you have an API key, or do you want to use Google Gemini? And you just say, I want to use Google Gemini.

Now I want to walk you through exactly how to do this. So the most important thing is that you need to go over here and pick the right model. You need to pick the new Gemini 2.

0 flash preview, and then you need to make sure the output is set to images and text. Once you've done that, you'll see you kind of have a prompting area. And one of the things I'll show you just with their stock image so it tells you, hey, you can do image editing.

And so it's like, it says, add some chocolate drizzle to these croissants. Oh, did that now I can say, replace that drizzle with strawberry icing. Now maybe we're at a strawberry bin and.

Let's see how it does. I love the little counter. It's telling you how long it's taking.

Boom, see, and I said strawberry icing. So it did icing and drizzle. Interesting pickup for those croissants.

Notice it did the same croissants, because I said, replace. Pretty amazing. It took 5.

6 seconds. Now let's say, let's say you're a bakery and you want to have a have an image where somebody's reaching and grabbing a croissant. So let's say add a hand to the picture grabbing the croissant.

Pretty wild, right? So I took that same picture, and now it's got a hand, and the hand is grabbing their croissant. And so like this might seem simple or rudimentary, but you can do a lot in terms of, imagine you have product images you're trying to edit.

You could take your product and juxtapose it in different ways, really, really quickly and iterate through those images in a really awesome way. Okay, now that we've played around with some images of croissants, what could it do with something that wasn't a physical good I'll give you a real, practical example. Here I am.

I hosted a podcast YouTube show. One of the things you always have to do is thumbnails. So I have a recent thumbnail from a show, and it's awesome.

It was a show we did last week, and now I think I can going to be able to manipulate parts of this thumbnail, like for example, can you replace the Manish logo with the word Google? Boom, it did it. It replaced it a little high.

Can you lower the Google text 20% Oh, it lowered it. It's better, but it lowered it. Can you put the word Google in middle?

It's getting a little confused, right? Once you get in a few edits in it does get a little bit confused. Let me refresh this start start it over again.

I think one of the things about image editing is you do need to know exactly what you're talking about. So let's see if it can do multi step directions. Can you replace the Manish logo with the word Google man center that word in middle of the box.

Additionally, and can you give the man? This is me in the picture, 1970s mustache. Let's see.

Okay, it got Google. It didn't get the mustache. Let's see.

So one of the things clearly is that it's still early days, and it's not good at complex prompts or multi steps directions. Let's see if I just make it a single step direction. Can it give me a mustache?

Oh, it did. It did. And if you're not watching on YouTube, you definitely should produce So Darren is in the background, losing it.

It isn't a mustache version. It is a 1970s mustache. Clearly go, clearly a winning thumbnail for this episode.

But in all seriousness, what this lets you do, if you're a marketer, is you can have basic templates and just do really quick changes. And like, I'm talking to producer Darren off air, and he's like, this is gonna save me a ton of time. Gonna save all of us a ton of time.

It's amazing to do that. And then you can download the new image. You can open up that image, because I opened that up so I knew you wanted to see sweet, see sweet, sweet stash and but you can even do basic thing.

Can you make the man's shirt blue? You know, one of the very common things for thumbnails is to change an element about it and look, I instantly have a sweet blue shirt. I actually think the blue looks better, better on me.

Candidly, uh, it's pretty mind blowing. And I want to, I want to kind of close this out with some other cool examples that have come out. Let's look at some other examples of how you might use this new image image editing, and the fact that this new model is multi modal.

Our friend Linus, who's been on the show before, has a few interesting examples. One of the things you can do is do visual story telling, so you can basically say, hey, the original Prop was generate a story of a white baby goat going on an adventure in a farm in a 3d cartoon animation style. For each scene, generate an image.

And so what it's doing is it's generating the story and the description so you can write it a book right away, like. Let's say you wanted to do an awesome children's book for your kid. You could do a children's book for your child every night, like for like, why do you need to go to Amazon anymore?

You can make bespoke children's book for your children every night. And quite frankly, I'm shocked nobody has took put a little wrapper on this API yet to charge like $5 a month, to just have a perfect output that plays like automatically and narrates it with 11 labs, like on your phone. Would kill.

Would kill, would kill. So when we talk about this multimodal API, it's even more and bigger than that image editing itself. So you can do this visual story, storytelling which is really cool.

You can just do the all the image editing, work, work, what you talked about, you can create Gif animations with this do tool, which is pretty awesome. And another really cool use case that I think if you're a marketer, you will use a lot is, let's say you really like the style of something, and you want to transform it, transfer it to another image. So it's like, Hey, I like the style of this.

Use the same style, but generate an image of a dog so you don't have to edit existing you can upload something and use that as an inspiration to have aI generate a net new image. It's pretty mind blowing. What is possible now from an image based world in this new experimental API.

It's still early days, but the one thing I would for sure do is I would never use Photoshop or Canva for really basic image editing, and I would push the limits of what's in your day to day work that you can do with this. And then also, if you have, like, a presentation or something you're working on, I would do a v1 like to try to get the flow right. I would put a script in this, and I'd have it generate the slot, the the like slide, image and story, and go through it in a really visual story way.

There's some really powerful things you can do right away. So again, you could do this now for free at Google, AI studio. And so AI studio.

google. com, and if you really liked what we shared today, please hit like, hit, subscribe. Check everything out.

We're going to come out with more videos like the this. It's been really awesome.