In the ongoing discussion surrounding AI and copyright, legal battles might surprisingly pave the way for progress. These lawsuits could create much-needed ’safespaces‘, clarifying what’s permissible and what’s not in AI development. Such clarity might even accelerate the mass adoption of AI production pipelines, as companies gain confidence in a defined judicial landscape.
Personally, I’m not overly concerned about these issues being dragged into the spotlight. In fact, it’s an opportunity for technical realities to come to light. Discussions around technical details will soon reveal that there’s no such thing as ’stolen art‘ in the traditional sense. Instead, we’re dealing with endless lines of codified concepts, compressed from terabytes of image data into a mere few gigabytes. Furthermore, much of the art was likely scraped legally from websites like ArtStation or DeviantArt, whose Terms of Service either explicitly covered such use (via opt-in or opt-out clauses) or never fully guaranteed that user content wouldn’t be scraped by other systems, including AI.
Let’s discuss:
- What’s your point of view on this?
- How do you foresee these lawsuits concluding?
- What steps should all parties take to work towards a resolution?