How judges, not politicians, could dictate America’s AI rules
The article contrasts the US’s reactive, pro-innovation regulatory stance with the EU’s proactive approach, highlighting the tension between fostering rapid technological development and protecting intellectual property rights. This legal framework allows companies to train AI models on vast datasets while waiting for specific harms to emerge, prioritizing capitalist growth over preemptive restriction. Critical lawsuits against major tech firms aim to address the copyright violations inherent in scraping protected works for generative AI training. These legal challenges seek to establish precedents similar to the music industry’s shift after Napster, moving away from unauthorized use toward structured licensing agreements. The goal is to create a system where artists and authors receive compensation and provide explicit consent for their data’s inclusion in AI models. This development is crucial for open data because it directly impacts the legality of using publicly available information for machine learning. If courts rule against current practices, it could restrict the free flow of data needed for open-source AI development, potentially forcing a transition to curated, licensed datasets rather than openly scraped internet data.
Source: technologyreview.comPublished on 2023-07-18