If there’s one recurring theme at this year’s OFFF festival in Barcelona, it’s that NFTs are so last year – and this year’s hot topic is generative AI. And nobody is pushing the topic on stage more than Adobe, which this week made some pretty huge generative AI announcements of its own. “Generative AI is here. There is no discussing that. It is really part of the next phase of human driven creativity,” said Rufus Deuchler, Adobe’s director of worldwide Creative Cloud evangelism.
Adobe Firefly, a family of generative AI models, was announced with great fanfare this week, and the company was keen to talk about it at the design and creativity festival. In particular, Adobe is at pains to stress its commitment to ethics and transparency – and, perhaps more importantly, the importance of user feedback at this critical early stage.
Adobe Firefly, available to try in beta right now, arrives at a time when generative AI is plagued with controversies around copyright and ethics, with countless examples of the tech having been shown to be trained on the intellectual property of artists. But with Adobe Firefly only trained on Adobe Stock images, it seems the company is trying to buck the trend. “If I ask Firefly to draw Mickey Mouse,” said Deuchler, “it won’t understand what I mean”. While Firefly would likely generate an image of a mouse, it won’t be able to reproduce Disney’s intellectual property.
But Deuchler placed particular emphasis on the importance of user feedback during the beta stage. “It is really important that as many people as possibly try it out,” he said. ““I want to build this together. Tell us what’s good and what’s wrong. This is truly a community effort. We need your feedback to help us and guide us so that we do the right thing all along.”
You can sign up to try the Adobe Firefly beta – and perhaps more importantly feed back to Adobe about the experience – right now (opens in new tab). And don’t forget to check out the best Adobe Creative Cloud deals (opens in new tab) too.
