model capability claims

Sep 19

World Model Startups Hide Development Behind Closed Doors

The world models boom is creating a new opacity problem in AI development: companies are raising massive capital on vague promises while declining to share research, benchmarks, or even basic technical details about their systems. This mirrors the pre-GPT-3 secrecy playbook but with higher stakes. If these models are truly foundational infrastructure (as investors believe), the lack of independent verification makes it impossible to assess whether the technology delivers or if it's venture-scale capital masquerading as proof.

theme-ai model capability claims security & safety geopolitical competition

Jun 28

Chinese AI Model Matches US Rivals on Security Testing, Exposing Export Control Gap

Source: The Wall Street Journal (paywall)

Zhipu AI's GLM-5.2 matches frontier US models (GPT-4, Claude) on vulnerability detection—a task previously assumed to require closed Western systems—and runs as open-source software available globally. The finding challenges the national security case for restricting Chinese AI exports. The US restricts closed Chinese models on dual-use risk grounds, yet allows open-source models trained on similar architectures and data to circulate freely. Either the export controls are insufficient, or the US must extend them to open-source releases—a move European regulators are approaching differently, and one that raises technical and political obstacles.