Can Adobe’s Project Indigo make generative AI photo edits feel like part of taking a picture, rather than a separate trip into an editor?
Adobe is rolling out a limited test today that adds generative AI-powered editing tools to Project Indigo, its experimental camera app, for only a small slice of users, according to 9to5Mac . The new feature set is called AI Playground, and Adobe is treating it as a short-run experiment rather than a full launch.
Who actually gets Project Indigo’s AI Playground today?
Only “a few percent” of Project Indigo users will see the new AI tools, and only “for a few weeks,” Adobe said in the source material cited by 9to5Mac. Users who get access will see a splash screen inviting them to join the test.
Adobe is keeping the trial free and removing the usual account barrier.
“Since Indigo is an experimental app, and this is an experiment inside Indigo, we’re offering the AI Playground for free, with no sign-on required. However, this level of free access will be available only to a few percent of our users, and only for a few weeks.”
That limited design matters. Adobe is not pushing AI Playground as a finished consumer feature. It is testing whether people use generative AI editing when it sits directly inside a camera app.
Project Indigo itself is still an experiment. Adobe announced it a little more than a year ago as a free iPhone camera app with full manual controls and what it described as a more natural, SLR-like approach to computational photography. Last April, Adobe expanded the app to select iPads and the iPhone 17e, while adding a grid view, photo filtering, stability fixes and image-quality improvements.
The AI test is opt-in. Adobe says that for users who join, it will collect only button presses inside the Playground. It says it “won’t look at your prompts, or images, or upload them to [its] servers, or use them to train AI models.” Because there is no sign-on, Adobe says the analytics are anonymous.
Which AI edits is Adobe testing inside the camera app?
The test adds four sections to Project Indigo’s AI Playground: Object Editing, Styles, Photo Guidance and Custom Edit.
| AI Playground section | What Adobe says it does |
|---|---|
| Object Editing | Removes distractors and can defocus the background to create shallow depth of field. Users can toggle categories of distractors, add categories by typing or dictating, and save or delete additions. |
| Styles | Re-renders an Indigo photo using traditional artistic techniques that Adobe says work well on everyday snapshots. |
| Photo Guidance | Uses AI analysis to offer reshooting suggestions and editing suggestions. |
| Custom Edit | Opens text-prompt editing for broader generative changes beyond preset buttons. |
The most revealing feature may be Photo Guidance. Adobe is not only testing AI as a retouching tool. It is testing AI as a coach inside the shooting flow.
Adobe frames this as similar in spirit to AI tools that summarize or critique human-created documents, but aimed at photography. Because Indigo is a camera app, Adobe says the analysis can include both reshooting and editing suggestions.
That turns the camera app into more than a capture interface. In MLXIO analysis, this is the key product question: whether users want AI after the image exists, or while they are still close enough to reshoot.
For adjacent MLXIO coverage on how software features are changing device use, see iOS 27 Beta 3 Lets AirPods Users Dial Out the World and Key Trends Splitting Tomorrow's Winners From Losers.
Why is Adobe using Google’s Nano Banana model here?
Adobe says AI Playground currently uses Google’s Nano Banana model. The tradeoff is image format: Nano Banana only outputs standard dynamic range images, even when a user captures a shot in High Dynamic Range.
Adobe says it built a custom ML model to post-process the standard dynamic range output from Nano Banana and boost it back to HDR. The company also cautions that the result is not perfect and depends heavily on what Nano Banana returns.
That caveat is important. The test is not just about whether people like AI buttons. It is also testing whether Adobe can preserve Indigo’s image-quality goals while sending generative edits through a model with output limits.
Adobe already markets Firefly as a place to “Generate and edit images, video, audio, and designs using top AI models from Adobe, Google, OpenAI, and more,” according to Adobe’s own product page. Project Indigo’s use of Google’s model fits that broader multi-model pattern, but in a more constrained setting: mobile photos captured inside one experimental app.
There is another hard boundary. AI Playground only works on images captured with Project Indigo. If a user imports an image from elsewhere, the sparkle button that opens AI Playground will be grayed out.
That restriction keeps the test cleaner. Adobe can evaluate edits on images created under Indigo’s own capture pipeline instead of trying to support every imported photo source from day one.
What would make this test graduate beyond an experiment?
Adobe has given itself several exits. It may extend the experiment, expand the user pool or run follow-on experiments. If the feature proves popular, Adobe says it will “eventually offer a paid version.”
The next signal is access. A test limited to a few percent of users can show how people behave when the feature appears, but it cannot yet prove how the broader Indigo user base would respond.
The second signal is quality. Object removal, background defocus, style transfer and prompt-based edits will be judged on realism, speed and failure rate. Adobe’s own HDR caveat suggests output quality is still a live technical issue.
The third signal is trust. Adobe’s current privacy claim is unusually specific for a generative AI feature: no prompt review, no image review, no upload to Adobe servers for this test, and no training use. If the feature expands or becomes paid, users will need to see whether those boundaries stay intact.
The practical takeaway is narrow but significant: Project Indigo is becoming a test bed for AI at capture time. The next few weeks will show whether Adobe treats AI Playground as a temporary lab inside Indigo, or as the first step toward making generative editing a standard part of its mobile photography workflow.
Key Takeaways
- Adobe is testing whether generative AI edits work better when built directly into the photo-taking experience.
- The limited rollout shows Adobe is treating AI Playground as an experiment, not a polished product launch.
- The no-sign-on, free trial lowers friction while Adobe measures usage without reviewing prompts or images.
