Adobe has just launched the beta of its new product, Adobe Firefly. A kind of Adobe Photoshop on steroids that uses generative AI art to create or modify images in seconds. Generative art by AI is an art form that uses intelligent algorithms to create original and varied works, from data, rules, or parameters defined by the artist or the AI itself. Adobe Firefly is a new AI imaging tool, part of Adobe Sensei's generative template family.
Adobe Firefly is currently in beta and uses images that have been licensed for AI training. A problem that has often been mentioned by several artists who believe that several other models such as MidJourney, Stable Diffusion and Dalle 2 literally steal their works.
Adobe Firefly allows users to generate high-quality images and text effects, using natural language. For example, one can write “a black cat with glasses” or “a blue logo with the word Firefly” and get a corresponding image. You can also modify the generated image by changing the words or by using graphic tools. Adobe Firefly integrates directly with workflows from Adobe products, such as Creative Cloud, Document Cloud, Experience Cloud, and Adobe Express.
It will soon be very easy for a photographer to modify an annoying detail in a photo or for a designer to create a logo or even the interface of a website. No need to tell you that many jobs such as "stock photo" photographers are greatly in danger. This will also lead to the creation of jobs for new "prompt" experts who will be able to fully understand the needs of a client and to match a style that corresponds well to the request.
This brings us to the question of the distinction between creativity and the technical ability of artists to make art. AI can produce works that exceed human limits in terms of complexity, diversity, or speed. But can AI be considered as an artist in its own right, or as a simple tool at the service of human vision?
Generative art by AI can also offer new opportunities for creative people who do not have talents or skills in painting, graphic design, or other artistic fields. Thanks to AI, these people can express their ideas, their emotions, or their messages through automatically or semi-automatically generated works. It is no longer necessary to understand how to draw, or do photography in order to be a creator, it will only be necessary to have an idea and let the AI do the work.
AI can also help them learn and improve their artistic techniques, providing examples and allowing to mix styles and see the result in seconds. We are already seeing the emergence of artists who use AI to inspire themselves and then produce the work in the traditional way.
Generative art AIs may therefore lead to an explosion of creative content in the future, by democratizing access to and participation in art. However, it will also pose challenges and risks, such as:
The protection of copyright and intellectual property of works generated by or with AI (One of the big problems is that it is difficult for an artist creating a work by AI to know if he is in the process of to copy a work without knowing it)
The preservation of the diversity and originality of artistic expressions in the face of the standardization or trivialization of works generated by or with AI
Recognizing and valuing the role and human contribution in the creative process involving AI
The ethical and social responsibility of artists and AI users in the production and dissemination of generative works. Some images may create confusion between real and fake.
These challenges and risks require critical reflection and constructive dialogue between the actors involved in generative art by AI, such as artists, researchers, developers, critics, viewers, and regulators. Governments should already be consulting on the subject in order to properly regulate these new gray areas.
This is undoubtedly one of the greatest revolutions in the artistic field and the impacts that this revolution will have should not be taken lightly because it is unlikely that we will go back.
