10 Photoshop Skills AI Just Made Obsolete

And let's click on Generative Expand. But you would notice that all of those details that were there right here, here in the expanded version, are just garbage trash. What is that art right there? That's not art. That's AI mumble rap. This is crazy bad right there. What even is happening? I'm not going to sugarcoat it. I'm going to be straight with you. In this video, we're going to be talking about 10 Photoshop skills that took people a long time to build, sometimes even years, that seems to be completely made obsolete by AI. Whether you choose to use AI or not, that is your artistic call. But it's the difference between spending 4 hours and 4 minutes. And many of these processes are thankfully not creative. They used to take manual work, hours and hours of skills, just to remove something distracting or do a thing that may not have creative influence. Let's take a look at these processes and see whether AI has actually taken over. Or there's a chance for our skills still. The first obvious one being removing or replacing items. In this case, removing this man in the background would be a There are strings on this tennis racket. And the color of the top of the man in the background is the same as the strings. If you were to select the rectangular marquee tool here and just make a selection like this, obviously click on generate a fill, type something like remove the man in the background from the model drop down, pick whatever model you want, and you would click on generate and gone. I actually generated a few times. Here was the first one. Here was the second one. But this is not over. That's where your real skills come in. Because if you were to zoom in, you would notice that it may not be the same person. Here's the before. See the textures are changing all around. Here's the after. It's changing the eye, the lips a little bit, the skin texture. So what do we do? We only apply it in areas where we actually need it. So I recommend just dragging and dropping the mask to the trash can, then holding the Alt key or the Option key and then clicking on the mask button. This creates a negative mask. Remember, black hides. So you would take the brush with white as the foreground color and with a soft round brush, you would only paint on areas where you actually need it. And this does a wonderful job that way. Before, after. It also helps you remove things accurately that were kind of an impossibility just a few years ago. Have a look at the glasses here. It's not only covering the face, but there's shadow of the glass, there's some smudging. Again, I've made a square selection, generate a fill, type in something like remove the glasses, and you are good to go. Now these AI models are not perfect. First time I generated, it made the face smaller. I generated again, this time it was good, but I just wanted to try one more time. So I did, but it still moved the mouth. So again, we use the same trick that we used before. We delete the mask, hold the Alt key or the Option key, click on the mask button to create a negative mask, take the brush and only paint over the areas where you need it. And have a look at the result. It is the same eye, the same skin texture, the same everything. Many, many years ago, I made a video about how to expand your photos in Photoshop. And the way we did that in one of the earliest videos, say if you wanted to expand it in a way that it looks good on your website, we just took a sliver of it by using the Rectangular Marquee tool, we made a sliver, we either stretched it or made a copy on a separate layer, we moved it like so, and then we inverted that by right clicking, Flip Horizontal. And it looked weird, but maybe it was manageable. Maybe if we used Patch tool here, it would be hard to tell, but this was all we had. And we would painstakingly do all of that and correct that later as well. You kids have it easier. All you got to do is to press C for the Crop tool, 16 by 9, or whatever aspect ratio you want, and expand it like so. Click on Generate and Expand, and you should be good to go. By the way, you can type in a prompt, you can choose your model. I'm going to choose the newest model, Firefly Fill and Expand, and click on Generate. Now, even though it has a resolution limit, because for most backgrounds, the backgrounds are blurred, it shouldn't affect it much. Let's take a look. And whoa, that's pretty good. There are three options you get. First, second, third, you can go with either one of those. And before, after. Even if you zoom in, since the background is blurred, it is hard to tell. But what if the background was more complicated than that? So if you press C, and if we choose Square right there, let's expand it like so. And let's click on Generate or Expand. Just generate it. It does something, but you would notice that all of those details that were there right here, here in the expanded version are just garbage trash. What is that art right there? That's not art. That's AI mumble rap. This is crazy bad right there. What even is happening? So what is the solution to this? You see, Nano Banana is a better model. But Adobe doesn't let you use that with Generate or Expand. And it's not really made for expanding, and it doesn't see those transparent pixels. So here's a smart thing we do. I'm adding a black background by clicking on the Adjustment Layer icon and choosing Solid Color. Picking black right over there, hit OK, and placing it under the layer. And then simply export the file to whatever platform you use, whether it's Magnific, Adobe Firefly, up to you. I'm going to leave a discounted link to Magnific if you want it. The only reason we are doing it here is because this supports 4K. Inside of Photoshop, it doesn't as of now. I've uploaded this photo and just type in something like Remove the black bars on both sides, 1 is to 1 aspect ratio, 4K, you can generate four of these, and click on Generate with Nano Banana 2 or Nano Banana Pro. While the rest is processing, if I were to get this in Photoshop, by the way, this looks good. This is the result with Nano Banana and obviously it has changed things. If you were to zoom in, here's the before, here's the original person. What are those? But our focus is the edge. And the edge, it does way better than Mumblewrap AI. Have a look at it. All the decorations, mostly good and definitely way better than what it was before. And you know the drill, just to keep the edge, hold the Alt key or the Option key, click on the mask to create a negative mask, take a brush, take a soft brush with white as the foreground color, just bring back the edge areas and merge them nicely. That's it. That's all you got to do. That way, you keep the subject exactly who he was. Just have a look at it. The quality has gotten so much better. Keep in mind, all of this is generated, the bricks and everything. Now, changing the lighting was a big, big challenge and quite impossible at that. And many years ago, we made a video where we zoomed in, did high end dodging and burning one stroke at a time. Took hours to remove a single shadow. Here's an extremely complex example. I selected the entire image, clicked on Generate a Fill and typed in Remove all shadow from the face and use the Firefly model. Have a look at it. Firefly image 5 from Adobe. And this is what happened. Here is the after. If I were to ask the AI to directly create something like this, it would probably ban me. But if it wants to do that, it does. So again, I exported the image to Magnific and use Nano Banana 2. You can use any platform you like actually. And typed in Remove all shadows from the subject, only because it could do 4K. And here we have four results and all of them removed it beautifully. Got that back inside of Photoshop and have a look at the results. Done. Here's the before. Here's the after. I don't know what to say. You can actually change the entire scene. For example, here's a lot of harsh sunlight. And to remove the shadows, we would have to change the entire scene and not just the subject. So this time I typed in a prompt like Remove all shadows and change the lighting to Overcast and Soft. Got it back into Photoshop and this is the result. We're going to zoom in. Here's the before. And actually it did not change the subject as much. This is insane. Now where AI actually needs improvement is when you want to change the lighting in a specific style. And this leads into the creative aspect of it. Here I asked it to create a three-point flash lighting setup with a rim light. And this is what it did. Now, so far it looks good. But as soon as you zoom in, you realize it may not even be the same person. Here's the before. Here's the after. It's very similar. But the eye changes. Before. After. Have a look at the teeth and the braces. They change as well. The skin texture changes as well. Another thing that would take so much time before to do in Photoshop would be the reflections. You would either have to create Curves Adjustment layer and do High End Dodging and Burning to remove all the reflections or use the Clone Stamp tool to clean that up later. But right now there's a feature for that. Let's make a copy of the Background layer by selecting it and pressing Ctrl or Command J. I recommend going to Filter, Convert for Smart Filters and then Filter, Camera Raw Filter. Inside of that, go to the Remove section, scroll down and you will find Reflections. Just click on Apply. You can even control how much of the reflection you want. And if you take it all the way to the left, you'll see just the reflection. For higher quality, change the quality to Best. And there you have it, cleanly and beautifully removed. Now, although this Reflection Removal feature is actually for situations where you have an entire glass throughout the image, but it doesn't remove glare from glasses as of now. But if you were smart about it, you made a selection, went into Generative Fill, used Nano Banana Flux, completely up to you. You could easily remove all the glare by keeping the eye exactly the same. Here's the before. Here's the after. I actually used Firefly Image 5 and used the prompt Remove All Reflections from Glasses. Now, here's one of the major real applications of AI that completely changed the game. Remember earlier, if we had to select an object like this, it would make us cry. We would have to zoom in, select the Pen tool, slowly and gradually make a path around it. Now, although nothing comes closer to the accuracy of the Pen tool, and you might need it for huge billboards, but for most of the work, you can right now click on Remove Background, and it does an incredible job, even for complex objects like this. AI will help you create those accurate selections fast. But you can creatively put this in whatever background you want. And this is something that I manually created, as you can see right here, and brought back in the original shadow. Here's the cutout for the simple white background, but we can put it in something like this. Have a look how cool this looks. And you can change the background to whatever you want. And this is all manually created in super high resolution, even though it has made all of those selections incredibly easy, you need to know the tricks. For example, in this case, if we click on Remove Background, if you have a look at the hair, the mask is okay. But as soon as you put it in something like a red background, the same color as her lipstick, let me do this, place it under the subject, the hair starts to look weird. And if you know the concept of blend modes, you can easily fix this. Just duplicate this layer by pressing Ctrl or Command J, turn the first one off, with the second one, change the blend mode to Multiply. Have a look, the edges are fixed, because it's only darkening it. But we do need to get the subject area back. For that, turn this layer on. And from this layer, subtract the edges, select the mask, take the brush, black as the foreground color, just erase around the edges. That's all you got to do. See how easy this is to fix. Now, you can take your time to do it more cleanly, but you get my point. You would only know these tricks if you understand the concepts of Photoshop and not just a prompt. Little compositing edits have become so much easier. May not be accurate, but have a look. Here's a jewelry, and I want this jewelry on this subject. And if you were to copy this and paste it right here, you would have to dodge and burn to first match the lighting, do some kind of warp to match that up. And on top of that, paint a lot of shadows to make it possible. The whole process would require years of skills and practice, and it would give you accuracy. But as much as I love to do it, I just have to share with you the reality. If you were to simply make a selection of this area and with Generate a Fill, you chose something like Nano Banana 2, there is an option for a reference image. You click there and you added the image of the same jewelry and typed in something like, add that jewelry. More likely than not, since it's a smaller area, the resolution shouldn't be an issue. And this is what it generates. Now, you can go back and compare the design of this. There might be some discrepancies here and there, but this is where we are going. I got to be honest with you, I made that video about removing sweat patches, even though that could be the most accurate thing. It took me a while to perfectly target that area, clean up the edges. But these days, somebody could just make a selection. For example, in this case, here was the before. Cleaning up braces would be a nightmare. You could just type in something like, remove braces, and it is done. Keep in mind, AI is not perfect. It took me three tries for it to happen. Whether you're trying to get an accessory from a reference or just typing in something like, add aviator sunglasses, it does that in a way that matches the lighting. And again, I would recommend masking just that area in. Now, you got to know that this technology is not perfect. It messes things up. So, I took a picture of this watch on the table and I fed that as a reference image and tried to change the watch of this person from this to this watch right there. Here's the after. And you would instantly figure out the watch is being worn at completely wrong direction. First of all, this knob should be on that side. And on top of that, the text is terribly all over the place. And I would personally never buy a product based on an AI image of the product. Another incredibly massive change that has come about is removing all kinds of distractions. Earlier, you would have to take the spot healing brush tool and go over all these wires and cables one by one and still come back to correct all of these. AI has made this manual work incredibly fast and our kudos on that. Create a new layer and simply with the remove tool, find distractions, wires and cables. Click on remove and it should do the job. Just make sure sample all layers should be checked and it's done. Here's the before, here's the after. Apparently, I didn't have to create a new layer but you get the point. Similarly with people, this is not a new feature. Find distractions, people and it automatically finds the people including my shadow taking this picture and it gets all of them removed. Solo picture of my parents. Here's the before, here's the after. It's not just limited to people, wires and cables. If there's dirt on the ground, if there's all kinds of distractions, find distractions, general distractions. It actually looks at the image and recognizes all kinds of distractions. For example, poles, spots and dust, text, everything, visual clutter, urban elements. Have a look at the dust and the debris. It just selected all of those areas. Even in retouching, there are certain manual things. I'm not talking about frequency separation, high-end dodging and burning because those are artistic in nature. But there are annoying things like removing wrinkles from backdrops, fixing little things. That takes up all your time. Simply press Ctrl or Command J and a series of plugins that I use and have been using for five years now and highly recommend are Retouch For Me plugins. Go to Filter, Retouch For Me. You need to have this plugin for it to work. It's an AI plugin. There are many plugins that touch upon the various aspects of retouching like removing blemishes, dodging and burning, cleaning up the veins in the eyes, removing shine. But the one we are using today is Clean Backdrop and it does it all automatically. Let us choose Bigger Dirt, which is the bigger ups and downs and Fixed, Done. You can choose how much of it do you want to remove, make mask and hit Apply. And you have it all done in a brand new layer. I made a popular video a few years ago about extending the backdrop. Again, it's the most accurate thing. But these days I'm surprised. You can just select that area, type something like extend the backdrop all the way to the ground and generate something like this. I know it is not perfect. Then you can select the mask, take the brush, black as the foreground color, just fade it in. That's it. That's really it. And the details are so good. You can either crop it, stretch it or generate it and you'd be good to go. Many a times we have missed focus when taking a photo and sometimes the depth of field was so shallow, it didn't get the whole thing. Now earlier we would use something like Filter, Sharpen, Unsharp Mask and try to fake all the details by sharpening it too much, increasing the radius and it would never get it right. And now you can try just typing in something like fix and sharpen the blurry nose and the lips and this is what you get. Crazy. Before, see all those blurred areas. It's hard to believe they maintained the position before, after and the edges as well. Here's some real world examples of one of the most coolest things in upscaling that has ever happened. Decent resolution, not that high. If you zoom in too much, it does start to pixelate. If you go to Magnific, inside of that if you were to open the Image Upscaler, keep in mind this is very expensive. I have uploaded the same photo and chosen the Precision model. Here we can make it larger by 4x. I don't want extra sharpness or grain and click on Upscale. Let's compare it in Photoshop and if this doesn't blow your mind, I don't know what will. Here's the before and here is the after. Didn't see the difference? Again, here is the before. See the leaf and what do I even say? What do I even say? Even if you take a look at the texture of the wood before, no texture, real texture. Upscaling with AI has entirely changed the game. I've made a video about it. You can check it out right here or check the links in the description for the guide. So what do you think of where everything is going with AI? If you were to ask me, you're always leaps and bounds ahead of people who just know AI if you have the core fundamental Photoshop skills or core fundamental design or photography skills. Someone who doesn't understand it may create something with AI, but they won't have taste. And when AI goes wrong, they would not know what to do. I hope you enjoyed this video and if you did, make sure to give us a like and also don't forget to subscribe and not just subscribe. Ring the bell so that you, my friend, don't miss any other future tips, tricks or tutorials. Thank you so much for watching. I'll see you in my next one. Until then, stay tuned and make sure that you keep creating.