How judges, not politicians, could dictate America’s AI rules

The article contrasts the reactive US approach to AI regulation with the EU’s proactive stance, arguing that this permissive environment fosters innovation. However, emerging copyright lawsuits against major tech firms challenge this open-data framework. These legal battles aim to clarify how "black box" AI systems operate and whether scraping copyrighted material for training constitutes fair use, potentially reshaping the legal boundaries of open-source data utilization. Legal representatives hope these cases will establish new licensing models similar to those in the music industry. Instead of relying on ambiguous fair use doctrines, the proposed solution requires explicit permission and royalty payments for using copyrighted content. This shift would fundamentally alter the open_data ecosystem by introducing contractual obligations and financial considerations into the collection and usage of publicly available digital assets, moving away from unrestricted access toward a rights-managed system. This development is crucial for open_data because it threatens the current norm of freely accessing and utilizing public web data for machine learning. If courts side with copyright holders, the cost and complexity of gathering training data will increase, potentially hindering open-source AI development. The outcome will determine whether open_data practices remain legally viable or if they must adapt to strict compliance and licensing requirements, significantly impacting how researchers and developers can legally access and share information.

Source: technologyreview.com
Published on 2023-10-25

Expand FOIA laws right now »

Open Data

Related news