Today's briefs
Sony Music and Warner Chappell Sue Anthropic Over Copyright Infringement
Sony Music and Warner Chappell Music have filed a lawsuit against Anthropic, alleging that its AI models were trained on copyrighted song lyrics without authorization. The case adds to a growing body of litigation from major content rights holders targeting leading AI labs over training data practices. For developers building on Claude or other Anthropic APIs, this lawsuit signals continued legal uncertainty around the provenance of training data and potential downstream liability. The outcome could influence how AI companies disclose training datasets, seek licensing agreements, or adjust model training pipelines going forward. This follows similar suits against OpenAI and other labs, suggesting a systemic industry reckoning with copyright law.
Anthropic
Code-as-World: Agentic Loop Converts Real Videos Into Executable MuJoCo Physics Programs
Researchers have introduced MIRROS, a system using a Code-as-World approach where an agentic loop analyzes real-world video and rewrites it as executable physics simulation code in MuJoCo. Rather than storing video as pixels or latent vectors, the system represents scenes as structured, runnable programs that encode object geometry, dynamics, and interactions. This approach makes world representations interpretable, editable, and physically grounded — a meaningful step for robotics, simulation, and embodied AI research. Developers working on sim-to-real transfer, synthetic data generation, or physics-based planning pipelines will find this directly relevant. The agentic loop architecture also demonstrates a practical pattern for using LLMs to bridge perception and executable simulation.
MarkTechPost
NVIDIA Earth2Studio Enables Custom Batched Ensemble Weather Forecasting
NVIDIA has published a tutorial and tooling guide for building custom batched ensemble weather forecasting pipelines using Earth2Studio, its open framework for AI-driven Earth system modeling. The update demonstrates how developers can compose multiple forecast models into ensembles and run them in parallel batches, improving both prediction diversity and computational efficiency. Earth2Studio is designed to make it easier to integrate NVIDIA's AI weather models like FourCastNet and Pangu-Weather into production-grade forecasting systems. For engineers working on climate tech, geospatial applications, or scientific AI, this lowers the barrier to building research-grade ensemble systems. The batching capability in particular is relevant for teams who need to run probabilistic forecasts at scale.
NVIDIA
China's Short-Video Ecosystem Is Driving a Competitive Lead in AI Video Generation
A new analysis highlights how China's AI labs are leveraging the country's massive short-video ecosystem — platforms like Douyin and Kuaishou — and significantly lower compute costs to advance AI video generation capabilities faster than Western competitors. Access to enormous volumes of high-quality short-form video data gives Chinese models structural training advantages for this specific modality. Lower GPU and cloud costs also allow for more aggressive iteration cycles compared to US-based labs. For developers evaluating AI video APIs, this suggests Chinese-origin models may offer strong quality-to-cost ratios and are worth benchmarking seriously against models from OpenAI, Google, and Runway. The analysis signals that AI video leadership may be genuinely contested, not a foregone conclusion for Western labs.
Tech - South China Morning Post