Harvard Magazine's headline signals a growing academic skepticism about copyright law's fit for the AI-training fight, echoing arguments that infringement doctrine wasn't built for large-scale statistical learning on text and images. For data licensing markets, that framing matters: if copyright proves an awkward fit, pressure may shift toward contract-based licensing regimes, sui generis data rights, or regulatory intervention instead of court rulings alone.
The piece appears to weigh in on a debate central to pending AI copyright litigation, though the full argument isn't available here. Data companies watching the copyright suits should note that even legal scholars are questioning whether courtroom wins will settle the underlying policy question.
Is Copyright Law the Wrong Weapon Against AI?
