For as long as people have taken photos, the core value has remained the same: preserving a moment exactly as it was. Whether it’s a candid snapshot with friends, a family portrait, or an important professional image, photos are supposed to faithfully capture who we are and what we saw.
Generative AI has opened up incredible new creative possibilities in photography. You can retouch an image, change the background, restyle a shot, or even create an entirely new scene. While today’s foundation models excel at creating synthetic visuals, they consistently fail at capturing the things that make a photo meaningful: identity and context.
The more powerful generative models become, the less usable they are for the moments that matter most — because preserving both individual identity and authentic context remains one of the hardest problems in the field.
That challenge is what drew Cecilia Zhang and Zach Xia – two of Adobe’s leading AI researchers – to start Phota Labs. They achieved new breakthroughs in personalized generative AI models that ensure every photo retains both the likeness of the person and the truth of the moment. The system disentangles who is in the image from what is happening around them, which means users can retake, restyle, or generate entirely new photos without losing fidelity to the person or drifting away from the reality of the scene. The results are AI-generated images that preserve real people in their true context — something generic foundation models can’t achieve without a deep understanding of identity.
Phota Labs has turned this breakthrough into two products. The first is their mobile app, which allows anyone to “reshoot” a photo by correcting issues like closed eyes, awkward expressions, or poor lighting, while keeping identity and context intact. There are no prompts required for generating a new photo, either – the model deeply understands photography itself, from lighting to composition to expression, so it can retake images automatically while preserving what makes the moment real.
Building on the same foundation, Phota Labs is also releasing an API for developers. This API spans both photo retaking and content creation, enabling applications to generate visual content that consistently reflects real people, places, and moments with fidelity. It also supports character swapping for specific individuals, opening the door to a new class of personalized and realistic creative workflows.
Input photo: Low-resolution, blurry screenshot of Yoko Li from a podcast she was on
Generated photos: Phota Lab generated retakes of the previous photo with various styles
Together, these products unlock entirely new experiences. Families can finally trust AI to preserve personal memories without distortion. Creators can generate content with true likeness. And developers can integrate identity-preserving image generation into consumer or professional applications, ranging from social storytelling to fashion to e-commerce. You can be among the first to try Phota Labs’ app and API by joining here.
Cecilia and Zach embody the kind of founders we are proud to back. At Adobe, they were not only prolific researchers but also hands-on builders, tackling some of the most complex problems at the frontier of computational photography and generative imaging. They bring a rare combination of deep technical expertise and product sensibility. In just a short time, they’ve already raised the bar for what personalized generative photography can achieve, and we believe this is only the beginning.
We are thrilled to announce that we are leading Phota Labs’ seed round. Phota makes it possible, for the first time, to create photos and videos that faithfully preserve identity while unlocking new forms of creativity.
This is the future of photography, and we couldn’t be more excited to be a part of it.
Yoko Li is a partner at Andreessen Horowitz, where she focuses on enterprise and infrastructure.
Martin Casado is a general partner at Andreessen Horowitz, where he leads the firm's $1.25 billion infrastructure practice.
Jennifer Li is a General Partner at Andreessen Horowitz, where she focuses on enterprise and infrastructure investments in data systems, developer tools, and AI.