Yadullah Abidi is a Computer Science graduate from the University of Delhi and holds a postgraduate degree in Journalism from the Asian College of Journalism, Chennai. With over a decade of experience in Windows and Linux systems, programming, PC hardware, cybersecurity, malware analysis, and gaming, he combines deep technical knowledge with strong editorial instincts.
Yadullah currently writes for MakeUseOf as a Staff Writer, covering cybersecurity, gaming, and consumer tech. He formerly worked as Associate Editor at Candid.Technology and as News Editor at The Mac Observer, where he reported on everything from raging cyberattacks to the latest in Apple tech.
In addition to his journalism work, Yadullah is a full-stack developer with experience in JavaScript/TypeScript, Next.js, the MERN stack, Python, C/C++, and AI/ML. Whether he's analyzing malware, reviewing hardware, or building tools on GitHub, he brings a hands-on, developer's perspective to tech journalism.
You're trying to take a photo on your phone, the preview looks perfect -- natural light, accurate colors, genuine texture. You snap the photo, swipe to the gallery, and everything changes. The subject's skin looks plasticky, the colors have changed entirely, and the photo barely resembles the real scene. What happened?
Your phone's AI happened. Specifically, the computational photography pipeline buried in your device's ISP (Image Signal Processor) is automatically applying aggressive post-processing before you even see the result. Understanding what your phone camera's settings do can help, but the real culprit is a system-level setting that's working against you the entire time.
Your smartphone is enhancing photos you never asked it to Computational photography prioritizes pop over realism
Modern smartphones don't simply capture light and create JPEGs. They perform multi-stage AI analysis, scene detection, texture manipulation, and algorithmic sharpening -- all without asking permission. When you take a photo on a Google Pixel, Samsung Galaxy, iPhone, OnePlus, or any other device, the built-in camera app makes real-time decisions about what looks good based on training data instead of what the scene actually looks like.
The result? Photos that look overprocessed before you open them in any editing app. The core problem is how AI models interpret sharpness and clarity. These algorithms were trained predominantly on high-contrast, studio-lit portraits. They don't understand what natural skin looks like under diffused window light. So they make conservative assumptions and when in doubt, sharpen aggressively. For example, compare this photo of my bike taken with manual settings versus one that my Pixel 9a decided to capture.
This works differently on every device, and manufacturers are often coming up with new solutions to fix these problems. Despite that, the problems exist. Even if there are no human subjects in the photo, they can look drastically different from the real scene. That's not to say all AI photo features are fluff. Some AI photo features can be genuinely useful, but these often work once you've already taken the photo.
Generally speaking, the AI image processing pipeline works in stages. First, the AI uses edge detection to identify boundaries, but often misses subtle gradients like freckles as noise that should be removed. Then, to reduce grain in low-light shots, the phone applies aggressive denoising before sharpening, which flattens fine texture. The subsequent sharpening also artificially brings in edge contrast, creating that plastic skin effect you see.
Related 10 Smartphone Photography Mistakes That Held Me Back for Years
These mistakes kept me stuck -- don't let them stop you.
Posts By Danny Maiorca
Your phone's hardware can also compound the problem. Variable aperture lenses anticipate shallow depth of field bokeh, so the AI preemptively oversharpens foreground skins even when the lens is optically sharp. This design is also intentional: social media platforms reward high-contrast, saturated images that catch a viewer's attention, so your phone's AI learns to optimize for that outcome by default.
All of this brings together an image that your phone thinks looks good, not what the scene you're trying to capture actually looks like. With AI systems getting better and people generally getting used to seeing a certain kind of photo, these changes often go under the radar, and phone companies pass them off as better photos. But as a photographer who's used to seeing optically correct photos, these changes can be really off-putting and often get in the way of creative decisions.
Settings that make your photos look fake Aggressive HDR and scene optimization are often the settings to blame
The specific setting that you need to tune will vary based on your phone's manufacturer and the camera app you're using. Even if you can find and disable any such settings, you still won't get rid of AI post-processing entirely.
Most phones' built-in camera apps will have some sort of scene optimizing setting that's the primary offender for these changes. This setting might be labelled something else on your phone, so you're going to have to dig around in your camera settings to find and disable any AI optimizations.
On Google Pixel phones, this setting is called Scene Detection. It uses real-time AI to analyze what's in the frame and apply multi-frame stacking, exposure fusion, and sharing during capture. Unfortunately, Google doesn't allow much control over what AI features get to meddle with your photos, but you can disable the Ultra HDR setting to tone down some of that aggressive post-processing.
Close Open the Google Camera app and tap the camera settings icon in the bottom left. Tap the overflow menu icon in the top right of the settings menu. Disable the Ultra HDR slider to reduce post-processing.
On Samsung phones, the Scene Optimizer and Super HDR settings tend to do the most processing. iPhones also have the Smart HDR and Photographic Style settings to blame.
If you can't find any toggles to disable AI settings, you can try capturing RAW photos if your phone supports it. You might think RAW on phones is pointless, but it produces a much more real image of the scene you're trying to capture. Not to mention the editing headroom you get.
Alternatively, using third-party camera apps that give you manual control is also a good way of avoiding AI processing. Apps like Open Camera on Android and Halide Mark II on iOS disable auto-sharpening and AI enhancement entirely. Phone OLED displays also tend to exaggerate contrast and edge definition, making AI sharpening appear more severe than it actually is.
AI image processing isn't the villain Smart enhancements often actually improve a photo
The AI processing on your phone isn't necessarily bad. It's what lets everyone with a smartphone tap a button and come back with a good-looking photo every time. It's what allows low-light photography with small lenses and camera sensors and is eventually responsible for the high-quality photos we expect our smartphones to produce.
Subscribe for practical tips on taming phone AI Get the newsletter for clear, device-specific guidance on turning off AI overprocessing, choosing RAW or manual apps, and restoring realistic textures -- actionable steps that help you control how your phone renders photos. Subscribe Subscribe for practical tips on taming phone AI Get the newsletter for clear, device-specific guidance on turning off AI overprocessing, choosing RAW or manual apps, and restoring realistic textures -- actionable steps that help you control how your phone renders photos. Subscribe
By subscribing, you agree to receive newsletter and marketing emails, and accept Valnet's Terms of Use and Privacy Policy. You can unsubscribe anytime.
Different AI image processing algorithms also give different phones a distinct look that's preferred by their users. If you capture the same scene with a Samsung, iPhone, Pixel, or any other device, you'll get slightly different results, which may or may not be your preference. It's what allows phone manufacturers to differentiate their cameras from each other.
Related Why I Don't Trust AI to Do Most Photo Edits for Me
When it comes to photo editing, I'd rather take control than rely on AI that doesn't always get it right.
Posts By Danny Maiorca
But for serious photographers, or anyone looking to make creative decisions about how their photos look, this AI processing gets in the way. If you like the photos your phone takes, there's nothing you need to do. But be aware that after you press the shutter button, your phone decides what the photo should look like, not you.
Make your photos look realistic again Dialing back processing without switching phones
Your phone's AI isn't trying to capture what you see; it's optimizing for engagement scores and platform-friendly contrast. These goals don't align with the human perception of realism.
Every time you disable AI features, step into better light, or make manual adjustments, you're working with your phone on your own terms -- reminding the AI that realism isn't noise to suppress, and texture isn't imperfection to erase.