The concept of a photograph is evolving rapidly in today’s tech landscape, as leading companies integrate generative AI into their cameras and software. Devices from major players like Google and Samsung now include advanced features that allow users to significantly modify images—whether by removing subjects from the frame, repositioning people, or introducing entirely new elements into a scene.
Apple is also stepping up its game, revealing plans to enhance its Photos app with new generative functionalities. Jon McCormack, who heads the iPhone camera division, indicates that Apple is focusing on a thoughtful implementation of these AI features, emphasizing that the company isn’t just pursuing AI for its own sake.
During the recent Worldwide Developers Conference, Apple previewed several AI enhancements coming to the Photos app in the forthcoming iOS 27 update, expected to launch later this year.
Currently, the Photos app includes a Clean Up tool that erases unwanted items from images, but it will see performance improvements in iOS 27 due to upgraded AI models. Notably, two new features—Extend and Spatial Reframe—will allow users to enlarge their images or adjust the perspective, while generating synthetic pixels to fill in the gaps. The AI essentially deduces what should occupy the background and seamlessly incorporates it.
McCormack highlighted that AI is enabling solutions to many previously unraveled problems, allowing everyday users to access powerful editing capabilities without needing extensive knowledge of complex software like Photoshop. “It empowers ordinary individuals with extraordinary tools,” he noted.
The Extend feature specifically allows users to create additional space in their photos, generating artificial pixels around the main subject based on AI predictions of what the background should entail. However, Apple is cautious about how users can modify their images. In the Photos app, the generated pixels can only enhance the background and will not alter the primary subject's face. Additionally, the Extend tool is limited to a single use that can only increase the image's dimensions by 25 percent—further edits cannot be made to prolong this extension indefinitely.
Furthermore, Apple plans to implement SynthID technology from Google DeepMind later this year. This technology will embed an invisible watermark in edited images, designating them as having undergone generative AI alterations. Shared photos on various platforms may be flagged as AI-modified, although it's important to note that research has indicated the limitations of digital watermarking.
“A photograph captures an actual event,” McCormack stated. “We are committed to preserving the authenticity of personal memories. When you take a photograph, you are capturing moments of your life that can be cherished later. It’s essential for us to develop tools that respect the integrity of these moments.”
