US IP litigation filings surge to 19,000 as AI copyright cases mount
A report by Cornerstone Research highlights a massive increase in US intellectual property litigation, which reached nearly 19,000 filings by 2025. The study points to a highly diversified landscape of patent, copyright, trademark, and trade secret disputes, driven by global technology competition, generative AI training data lawsuits, and multi-jurisdictional enforcement strategies.
Key Takeaways
- Annual U.S. IP litigation filings grew from fewer than 3,000 in the mid-2000s to approximately 19,000 by 2025.
- Generative AI has triggered nearly 100 copyright lawsuits focused on training data and fair use applications.
- U.S. patent damages remained at an elevated level of $3 billion annually through 2024 and 2025, driven by individual verdicts exceeding $300 million.
- Strategic forum selection has shifted toward the U.K. for global FRAND rate-setting and Europe for rapid injunctions, while the U.S. remains the primary venue for monetary damages.
- India has emerged as the fastest-growing major jurisdiction for patent litigation, reflecting a broader shift in innovation toward Asia.
Why It Matters
The explosion in IP litigation signals that content and data are now the primary battlegrounds for streaming and tech giants. As generative AI models require massive video and text datasets, the influx of nearly 100 copyright cases suggests that 'fair use' boundaries will soon be redrawn by the courts, potentially forcing a pivot from scraping to high-value licensing. For the streaming ecosystem, this trend increases the premium on proprietary libraries and underscores the risk of multi-million dollar 'blockbuster' verdicts. Stakeholders should monitor upcoming 2026 rulings on AI training data, which will likely dictate the cost of content ingestion for future video-generation tools.
Additional Context
The surge in intellectual property litigation coincides with a landmark shift in judicial handling of AI training. Per a March 2026 overview from Morrison Foerster, a 'judicial consensus' is emerging that training general-purpose AI models is 'highly transformative,' a key factor that often favors fair use. However, specialized cases continue to challenge this trend. In February 2026, YouTube content creators filed a wave of class-action lawsuits against Meta, Snap, and Runway AI, alleging that these companies circumvented technological protection measures to scrape video data, per the Copyright Alliance. Streaming-specific enforcement has also intensified globally. In June 2026, a California federal judge advanced a copyright suit brought by adult film producers against Meta, ruling they plausibly alleged the company utilized BitTorrent to download thousands of films for training its 'Movie Gen' AI model, per Westlaw Today. Concurrently, traditional media giants are targeting generative platforms directly. In early 2026, Disney, Universal, and other major studios filed a lawsuit against Midjourney, characterizing the platform as a 'bottomless pit of plagiarism' for replicating droids and other protected characters, per Ludwig APC. Beyond the U.S., the intellectual property landscape is rapidly maturing in Asia to support local innovation. WIPO reported in January 2026 that India recorded a record 110,375 patent applications in the 2024–25 fiscal year, representing nearly 20% year-on-year growth. This expansion mirrors the rise of ‘Schedule A’ litigation, which allows brand owners to sue hundreds of anonymous online sellers simultaneously. According to IPWatchdog, U.S. trademark cases filed in district courts surged 25% in 2025, reaching over 4,200 filings as e-commerce platforms became conduits for counterfeit digital and physical goods.
Read full article at fixedincome.fidelity.com