By Divyanshi Sinha
In a verdict that has sent shockwaves through Silicon Valley and newsrooms alike, a U.S. federal court ruled today that generative AI companies must establish "direct licensing agreements" for the use of real-time news data. The ruling marks a definitive end to the "wild west" era of AI training, where Large Language Models (LLMs) scraped journalism without compensation.
The court found that while training on historical archives may fall under fair use, the "live-scraping" of breaking news to provide AI-generated summaries constitutes a direct market substitute for journalism. Global media conglomerates, led by The New York Times and News Corp, are hailing the decision as a "survival lifeline" for the industry. Starting next month, a new "Micro-Licensing" protocol will be enforced, requiring AI platforms to pay a fraction of a cent every time an AI summary utilizes a fact sourced from a verified news outlet. This shift is expected to inject billions back into investigative journalism, though smaller tech startups warn it may stifle innovation.
