Adobe Firefly, an AI image generator, has been marketed as ethical, primarily trained on licensed stock images. However, a report suggests that some images in its training data may have been sourced from competitor Midjourney without proper licensing. Adobe claims only 5% of the images were from this questionable source and went through a moderation process. Despite this, concerns arise about the model's copyright safety, especially considering its use in creative projects.
Jeremy Beckler
Posted on
Adobe Firefly used thousands of Midjourney images in training its 'ethical AI' model
For further actions, you may consider blocking this person and/or reporting abuse
Discussion (1)
Wow, this is just plain wrong. It's a slap in the face to the original creators whose work might have been exploited without consent (obviously). But what about the customers who rely on the AI's output? They deserve better than this kind of sketchy behavior. Adobe can't just brush this off by saying it's only a small percentage of images. This kind of thing undermines trust in AI systems and the companies behind them.