Open-Source "PhotoCraft" Image Editor Emerges as AI-Generated Clean-Room Clone of Adobe Photoshop

It was only a matter of time before developers began pointing advanced AI coding tools directly at industry-standard software like Adobe Photoshop to see if they could construct a functional alternative. That concept has now materialized in the form of PhotoCraft, a brand-new, open-source, cross-platform image editor written entirely in the Rust programming language. The application forms a key part of an evolving suite of creative utilities spearheaded by the ArtCraft project, a venture led by Brandon Thomas, who previously served as the CEO of storyteller.ai.

Rather than relying on traditional reverse engineering of proprietary code—a practice fraught with legal ambiguities and potential infringement risks—the team behind ArtCraft is billing PhotoCraft as a "clean-room reimplementation" of Adobe Photoshop. Because the developers have carefully avoided reverse-engineering any undocumented aspects of Adobe’s software, the project successfully maneuvers around major copyright hurdles by relying exclusively on publicly available specifications and observed behaviors. This legal and architectural distinction is precisely the kind of development that free and open-source software (FOSS) enthusiasts are likely to celebrate as an impressive technical achievement.

Despite currently residing in its early alpha stage, PhotoCraft arrives remarkably well-stocked with core imaging features. The application natively supports layers, layer masks, complex layer styles, smart objects, and a fully functional brush engine. Furthermore, it boasts an impressive array of 34 distinct tools, 27 blend modes, and a wide variety of image filters, alongside robust support for opening and saving files directly in Adobe’s proprietary .psd format.

To lower the barrier to entry for prospective users, the developers have meticulously mirrored the interface layout of the industry giant. The placement of top-level menus, individual keyboard shortcuts, and specific tool icons have all been closely adapted from the real deal. The underlying philosophy driving this design choice is simple: individuals who are already deeply familiar with Photoshop’s notoriously steep learning curve should be able to transition smoothly to PhotoCraft and continue their daily workflows without missing a beat.

Performance is another area where the project punches above its weight class. GPU-acceleration is baked directly into the architecture, leveraging Vulkan on Linux systems, Metal on macOS, and DirectX 12 on Windows, alongside a WebGPU implementation available for self-hosted WebAssembly builds. Interestingly, the application also incorporates an extensive network of agentic hooks for virtually every feature, a forward-looking design choice intended to allow external AI agents to programmatically interact with and "use" the application alongside human operators.

When put to the test, the application starts up with remarkable speed. Even when running on modest hardware—such as a legacy Chuwi workhorse laptop featuring an entry-level m3-8100y CPU, integrated graphics, and minimal computing overhead—the interface feels consistently responsive. True to the legacy of its commercial inspiration, it is capable of quickly swelling to consume several gigabytes of RAM shortly after loading a complex, multi-layered PSD file, providing an almost uncanny authenticity to the resource-heavy Photoshop experience.

The motivations behind building such an ambitious project stem from widespread frustrations with the status quo of proprietary software licensing. Based on public statements shared across developer communities like Reddit, Brandon Thomas was initially motivated to clone Adobe’s flagship wares due to the mounting subscription fees, administrative hurdles, and general friction associated with attempting to cancel an ongoing Adobe subscription. For a software engineer with a deep background in artificial intelligence development, building a functional, high-performance clone presented an irresistible technical challenge.

It Won’t Replace Photoshop for Professionals

For users who possess an intimate, daily familiarity with the genuine article, approaching a newly minted clone often invites a healthy dose of skepticism. Having spent considerable time testing PhotoCraft, it certainly comes closer to replicating the actual user interface and user experience than previous community efforts to simply skin GIMP to look like Photoshop.

However, despite these visual similarities, PhotoCraft remains a superficial facsimile with some pronounced structural flaws. In this regard, the application evokes memories of Photopea, the remarkably capable free online image editor created by a human developer years before the current generative AI boom. PhotoCraft feels far more indebted to that lineage than to traditional desktop alternatives.

Having similar features positioned in familiar locations with matching icons and names is fundamentally different from those features behaving or rendering outputs in an identical manner. While PhotoCraft achieves an uncanny visual resemblance to Photoshop, it falls short of actually feeling like Photoshop under the hood. Over many decades, Adobe has rigorously fine-tuned its complex blending modes, filters, content-aware fill tools, advanced object detection algorithms, and spot-removal routines. While an independent open-source clone can study public specifications and attempt to mimic observed behaviors, it cannot instantly replicate decades of proprietary optimization and exact mathematical output.

Whether this discrepancy matters ultimately depends entirely on an individual’s level of expertise and reliance on the software. Casual users performing basic cropping, color correction, or light retouching may not notice any difference at all—until they attempt to follow along with an intricate, step-by-step Photoshop tutorial on YouTube. For that specific cohort of users, a wide range of capable image editors already exists, and alternatives like Affinity are even known to run quite smoothly on Linux environments via Wine.

For professional creators, however, the gap between the two applications becomes immediately apparent. PhotoCraft’s missing features, occasional bugs, divergent rendering outputs, and the cumulative friction of all those invisible operational differences add up quickly in a high-pressure, day-to-day production workflow. Whether PhotoCraft represents a genuine, long-term attempt to democratize professional image editing or serves as a sophisticated development flex directed at a major software conglomerate remains to be seen.

Try PhotoCraft on Ubuntu

PhotoCraft is distributed as free, open-source software with official support spanning Windows, macOS, and Linux platforms. The project’s official GitHub repository offers pre-compiled alpha builds for all major operating systems, including a native DEB installer tailored for Ubuntu systems running on Intel, AMD, or ARM64 architectures. Alternatively, users can easily opt for a portable AppImage format.

While the core software itself remains entirely free to use, the broader ArtCraft website outlines tiered monthly and yearly subscription models. Financial contributors gain access to specialized AI credits intended to power generative artificial intelligence features within the broader ecosystem of apps, though those generative tools are not currently bundled directly into this specific image editor.

In common with many modern software projects accelerated by AI coding assistants, the pace of development has been breakneck. Version 0.1.0 debuted shortly before public attention surged, and the repository has seen a rapid succession of minor and major updates released in quick succession. As the software continues to evolve through its early alpha lifecycle, development milestones are crossing the finish line almost daily.

Given widespread public animosity toward subscription-based software models and Adobe’s dominant market position, alongside PhotoCraft’s gratis price tag, it comes as little surprise that the project has rapidly attracted immense developer interest, community goodwill, and thousands of stars on GitHub as its user base continues to grow.