Adobe has announced the release of Firefly Image 3, its most advanced image generation model yet, designed to address criticisms of its predecessor, Firefly. The new model promises to generate more realistic imagery, improve lighting and text generation capabilities, and render dense crowds and individuals with detailed features and a variety of moods and expressions.
The new model is now accessible in Photoshop (beta) and the Firefly web app. Adobe has also implemented a program to pay Adobe Stock contributors to the training data set to ward off lawsuits and position itself as a more ethical alternative to generative AI vendors who train on images indiscriminately.
Adobe's design leader, Scott Belsky, has emphasized the company's commitment to developing new policies and safeguards to protect its Generative Match feature from being abused. The feature will prompt users to agree to Adobe’s terms of use and confirm they have rights to use the uploaded image, alongside storing a thumbnail image of the uploaded content on Adobe’s servers to ensure compliance. Generative Match will also remain in beta while the company seeks feedback, and users won’t be permitted to use it for commercial purposes during this time.
However, there are concerns that the system may not be effective in preventing copycat behavior, and may be more about limiting Adobe's liability than preventing copycat behavior in the first place.
Adobe's use of AI-generated images for training has been controversial, with some considering it a form of data laundering. However, the company has taken steps to address these concerns by implementing a program to pay Adobe Stock contributors to the training data set.
The new model is part of Adobe's broader efforts to enhance its generative AI capabilities, which also include the introduction of generative AI in Illustrator and Adobe Express. The company has also introduced new policies and safeguards to protect its Generative Match feature from being abused.
The new model is expected to be well-received by Adobe's users, who have been clamoring for more advanced image generation capabilities. With its improved lighting and text generation capabilities, the new model is expected to be particularly useful for designers and photographers who need to create realistic images for their projects.
In summary, Adobe's new image generation model, Firefly Image 3, is part of the company's broader efforts to enhance its generative AI capabilities and address criticisms of its predecessor. The new model is designed to generate more realistic imagery, improve lighting and text generation capabilities, and render dense crowds and individuals with detailed features and a variety of moods and expressions. The company has also implemented new policies and safeguards to protect its Generative Match feature from being abused, but there are concerns that the system may not be effective in preventing copycat behavior.
