In a sharp retort to OpenAI and Microsoft, Adobe today announced Firefly, a family of responsible generative AI products and services. That is, instead of training its learning models on copyrighted imagery online, leading to potential legal issues, Firefly is trained on images that Adobe owns. So anything it creates is safe to use.

“Generative AI also opens the door to new questions about ethics and responsibility in the digital age,” Adobe’s Dana Rao writes. “As Adobe and others harness the power of this cutting-edge technology, we must come together across industries to develop, implement and respect a set of guardrails that will guide its responsible development and use.”

To that end, Adobe’s family of Firefly generative AI models is trained on Adobe Stock images, openly licensed content, and public domain content where the copyrights have expired. This gives Adobe’s customers a “competitive edge,” the firm says, because any commercially created content is both legally safe to use and ethically sound.

As for Firefly itself, its capabilities are being built into any Creative Cloud, Document Cloud, Experience Cloud, and Adobe Express workflows where content is created and modified. Anyone interested in trying it out today can sign-up for the beta, though there is a waiting list. When it does come to market more broadly, it will do so first in Adobe Express, Adobe Experience Manager, Adobe Photoshop, and Adobe Illustrator.

And in the future, Firefly will expand to leverage new assets, technology, and training data from Adobe and others. But it will never generate content based on other people’s or brands’ intellectual property.