AI News

Hi, you are here because you said that AI image editing is the same as Photoshop

By The AI Sckool
Hi, you are here because you said that AI image editing is the same as Photoshop

Share

“We’ve had Photoshop for 35 years” is a common response to concerns about generative AI. And you ended up here because you made that argument in a comment thread or on social media.

There are countless reasons to worry about how AI image editing and generation tools will affect the trust we place in photographs, and how that trust (or lack thereof) could be used to manipulate us. badand we I know it’s already happening. So, to save us time and energy, and to not tire our fingers out by constantly responding to the same arguments, we’re just listing them all in this post.

But sharing it will be much more effective — just like AI! Isn’t that amazing!

It’s effortless to make that argument if you’ve never gone through the process of manually editing a photo in an app like Adobe Photoshop, but it’s a frustratingly simplistic comparison. Let’s say some nefarious scoundrel wants to manipulate a photo to make it look like someone has a drug problem—here are a few things they’d have to do:

  • Have access to (potentially costly) computer software. Sure, there are mobile editing apps, but they’re not great for anything beyond minor tweaks like skin smoothing and color adjustments. So you’ll need a computer for this task—an costly investment in internet junk. And while some desktop editing apps are free (Gimp, Photopea, etc.), most professional tools aren’t. Adobe’s Innovative Cloud apps are among the most popular, and recurring subscriptions ($263.88 per year for Photoshop are extremely tough to cancel.
  • Find the right drug apply equipment pictures. Even if you have them on hand, you can’t just throw in any elderly photo and hope it looks good. You have to consider the lighting and positioning of the photo you’re adding them to, so everything has to match. For example, any reflections on bottles should be from the same angle, and objects photographed at eye level will look obviously phony if they’re dropped into a photo taken from a higher angle.
  • Discover and apply a full range of sophisticated editing tools. Any inserts need to be cut out of any background they were on, then blended seamlessly into their up-to-date surroundings. This may require adjusting the color balance, tone, and exposure levels, smoothing edges, or adding up-to-date shadows or reflections. This takes time and experience to ensure the results look even satisfactorynot to mention natural.

Photoshop has some really handy AI tools that make this easier, like auto-select and background removal. But even if you apply them, it still takes a lot of time and energy to manipulate a single image. By comparison, that’s what Edge editor Chris Welch had to do it To get the same results using “Reimagine” on your Google Pixel 9:

  • Launch the Google Photos app on your smartphone. Click on the area and tell it to add a “medical syringe filled with red liquid”, some “fine lines of crushed chalk” next to the wine and the rubber hose.

That’s it. A similarly effortless process exists in Samsung’s latest phones. The skill and time barrier is not only reduced – it’s lost. Google’s tool is also incredibly good at combining any generated footage with images: lighting, shadows, opacity, and even focal points are taken into account. Photoshop itself now has an AI image generator built in, and the results from it are often not nearly as convincing as what this free Android app from Google can spit out.

Image manipulation and other forgery techniques have been around for nearly 200 years—almost as long as photography itself. (19th-century ghost photography and the Cottingley Fairy are examples.) But the skill and time required to make these changes are why we don’t think to check every photo we see. Manipulations were infrequent and unsuspected for most of photography’s history. But the simplicity and scale of AI in smartphones means that any idiot can produce manipulative images on a frequency and scale never before experienced. It should be obvious why this is alarming.

That’s why You having a respectable ability to detect when an image is phony doesn’t mean everyone can. Not everyone hangs around tech forums (we love you all, fellow hangers-on), so typical AI indicators that seem obvious to us can be effortless to miss for those who don’t know what signs to look for — if they’re even there. AI is quickly getting better at creating natural-looking images that don’t have seven fingers or Cronenberg-style distortions.

In a world where anything can be phony, it’s much harder to prove something is true

It may have been effortless to spot when the occasional deepfake was thrown into our feeds, but the scale of production has changed seismically in the past two years. It’s incredibly effortless to do, so now it’s fucking everywhereWe are getting dangerously close to living in a world where we must be careful not to be fooled by every image that is presented to us.

And when everything could be phony, it’s much harder to prove something is real. That doubt is easily exploited, opening the door for people like former President Donald Trump to make false accusations that Kamala Harris is manipulating crowd sizes at rallies.

It’s true: even if AI is much easier to apply than Photoshop, the latter was still a technological revolution that forced people to face a whole up-to-date world of deception. But Photoshop and other editing tools predated AI he did create social problems that persist to this day and continue to cause significant harm. The ability to digitally retouch images in magazines and on billboards promoted impossible standards of beauty for both men and women, with the latter being disproportionately affected. For example, in 2003, then 27-year-old Kate Winslet was unknowingly slimmed down for the cover of GQ – And This was justified by the editor of the British magazine Dylan Jones saying her look had been changed “no more than that of any other cover star.”

Such edits were common and rarely disclosed, despite the stern scandals that surrounded early blogs such as Jezebel published unretouched celebrity photos on the covers of fashion magazines. (France even passed the law (requiring disclosure of retouched content). As easier-to-use tools like Facetune have emerged on emerging social media platforms, they have become even more insidious.

One study conducted in 2020 found that 71 percent of Instagram users would edit their selfies with Facetune before they were published, and another stated that the images in the media caused the same decline in body image in women and girls with or without a label stating that they have been digitally altered. There is a direct channel connecting social media with real plastic surgerysometimes striving for physically impossible results. And men are not immune — social media has a real and measurable impact on boys and their self-image.

Impossible beauty standards are not the only problem. Staged photos and photo editing can mislead viewers, undermine trust in photojournalismand even highlight racist narratives — as in the 1994 illustrated photograph that darkened OJ Simpson’s face in a police photo.

AI generative image editing not only amplifies these problems, lowering the barriers even further—sometimes it does so without any clear direction. AI tools and applications have been accused of causing women to develop breast enlargement AND revealing clothes without instructions to do this. Forget about viewers not being able to trust that what they see is real – now photographers can’t trust their own tools!

First of all, we need to develop good speech laws – and let’s be clear, these are likely I would do be speech laws — is incredibly tough. Managing how people can produce and publish edited images will require separating uses that are overwhelmingly harmful from those that many people find valuable, I like art, commentary and parody. Lawmakers and regulators will have to take this into account applicable provisions on freedom of speech and access to informationin this First Amendment in the USA.

Tech giants have entered the era of artificial intelligence at full speed, apparently without considering the possibility of regulation

Tech giants have also charged headlong into the AI ​​era, seemingly without even considering regulation. Global governments are still struggling to pass laws that can stop those who abuse generative AI technology (including the companies that create it), and development of systems to identify real photos from phony ones has proven sluggish and woefully inadequate.

Meanwhile, uncomplicated AI tools have already been used to manipulate voters, digitally undress photos of children, and grotesquely deepfake celebrities like Taylor Swift. That’s just in the past year, and the technology is only going to get better.

In an ideal world, adequate protective barriers would already be in place before a free, idiot-proof tool that can add bombs, car crashes, and other nasty things to photos in seconds makes it into our pockets. Maybe we If fucked up. Optimism and willful ignorance won’t fix this, and it’s not clear what will, or even Power at this stage.

The AI Sckool
The AI Sckool

Latest Posts