Adobe Launches Firefly, Elevating Photoshop with Advanced AI Image Generation
Adobe has announced the release of its third-generation image generation model, Firefly Image 3, which is now accessible in the beta version of Photoshop and the Adobe Firefly web app. The new model is touted to create more realistic imagery, thanks to its improved understanding of complex prompts and scenes, as well as enhanced lighting and text generation capabilities. Additionally, Firefly Image 3 is more proficient at rendering dense crowds and detailed features of people, which is attributed to a larger and more diverse training dataset.
Alongside Firefly Image 3, the beta version of Photoshop includes an advanced AI-powered Generative Fill feature, which allows users to incorporate a reference image to guide content creation. Other notable additions include an Adjustment Brush for targeted non-destructive edits, Adjustment Presets for rapid visual changes, and an improved font browser with access to over 25,000 cloud-based fonts. Moreover, the updates have been integrated into Adobe's mobile apps, Adobe Express, offering new features and enhancements.
Adobe's move to augment its software with generative AI tools comes amidst growing competition from AI startups and seeks to provide creative professionals with advanced capabilities for image generation and editing. To address ethical concerns and potential copyright infringement issues, Adobe has initiated a program to compensate Adobe Stock contributors and is working on a "do not train" tag as part of its Content Authenticity Initiative. Despite some controversy regarding the training practices of the Firefly models, the company is pushing forward with these AI innovations, with a full public release expected later in the year.
