Rapid advancements in global AI infrastructure and model development are occurring alongside intensifying geopolitical competition and scrutiny of AI's societal impact.

Chinese AI firm Zhipu AI has launched its GLM-5.3-Flash model, which the company claims was trained entirely on a cluster of 100,000 domestically produced chips, highlighting significant progress in China's AI self-sufficiency.

Nvidia has resumed shipments of its H200 data-center processors to China under a new US licensing scheme, though the company expects these sales to represent a negligible share of its total revenue.

Anthropic has introduced a new hardware standard that enables AI agents to exert control over the physical world, marking a shift toward more interactive and embodied artificial intelligence systems.

Alibaba Group is expanding its global AI footprint by launching its first data centers in Brazil, aiming to capture emerging market demand for AI infrastructure.

Google has expanded its AI product suite with the release of Gemini Omni 1.1 Flash, providing developers with enhanced creative controls and new generative capabilities for building interactive applications.

A recent academic paper has ignited debate regarding the future of healthcare, suggesting that AI models are already performing better than human doctors in specific medical tasks.

Meta has implemented stricter limitations on its AI glasses, curbing nonconsensual recording features to address growing privacy concerns among the public.

A federal judge has ruled that the Trump administration's blacklisting of the AI safety company Anthropic was illegal, marking a significant legal development in tech-government relations.

The platform Cara, known for protecting artists' work from AI training scrapers, reports that it is currently being targeted by trolls attempting to seize and publish its protected data.