Adobe Adds AI Critics to Mobile Cameras

Mobile photography is moving beyond simple filters and automated exposure adjustments into the realm of active, real-time creative guidance. Adobe is now pushing its experimental iOS camera platform, Project Indigo, launched last year, into a more interactive role by using large language models to actively critique user images.

Automating Creative Photographic Feedback

The latest update to Project Indigo introduces features designed to offer a professional-style assessment of a captured frame. Rather than relying on standard prompt-based interactions, the application utilizes buttons to trigger specific, deterministic outcomes. These tools evaluate elements like framing, lighting, and color, providing users with actionable feedback on how to improve their shots or adjust their composition before or after capture.

Guided Editing and Object Removal

The app introduces a capture and edit system that provides suggestions for re-shooting a scene by adjusting exposure or viewfinder composition. Complementing these diagnostic features are advanced object removal tools. Unlike traditional methods that require manual circling of an item, the app offers specific toggles to isolate and remove background elements such as fences, wires, vehicles, or trash.

Most generative AI tools provide prompt-based editing and, often, finding the perfect prompt to tweak a photo or get the right result can be a tricky endeavour. That’s why most of the new experimental AI features are buttons that can generate more deterministic outputs.

— Marc Levoy, the person heading Adobe’s project

Infrastructure and Model Flexibility

Adobe is currently utilizing Google’s Gemini-based Nano Banana to power these photographic critiques, though the company indicates it may integrate other technologies, such as the Adobe Firefly model, in the future. These capabilities are currently confined to an AI playground tab within the app, which remains in a testing phase restricted to select users.

• Project Indigo launched as an experimental iOS camera app last year.
• The current AI playground features are limited to a select group of testing participants.
• The app now includes depth-of-field simulation and style transfer options, drawing comparisons to the early days of the Prisma app.

For professional photographers and casual users alike, the shift toward AI-assisted critique represents a significant evolution in how software interacts with visual capture. While these tools aim to educate users on the nuances of composition, the reliance on automated models could lead to a standardization of aesthetic choices. As Adobe continues to test whether these features reach a broader audience, users should consider how much reliance they place on algorithmic feedback versus their own creative intuition when refining their digital portfolios.