China’s GLM-5.3 Is Closing the Cyber-AI Gap—but Trusted Access Changes the Story
Z.ai says GLM-5.3 nears a leading closed model on vulnerability detection while reserving advanced capabilities for verified users.
Tag: Open-Weight AI
US Senator Urges Trump to Back Open-Weight AI Against China—The AGI Race Is Becoming National Policy
A US senator wants incentives for American open-weight AI to counter China. The debate now sits between innovation, national security and distributed AGI.
“Open” AI Is Getting a Price Tag: Alibaba Wants a Cut When Big Companies Profit From Qwen
Alibaba is reportedly preparing revenue-sharing terms for large commercial users of Qwen, challenging the assumption that open-weight AI means free forever.
The US Says It Will Not Safety-Test Open-Weight AI Models—Leaving a Major Gap
US officials reportedly told leading AI companies that government safety testing will exclude open-weight models, raising concerns about a major oversight gap.
China’s GLM-5.2 Is Closing the AI Capability Gap—but Researchers Say Its Safety Gap Is Wide Open
Researchers say China’s open-weight GLM-5.2 approaches frontier performance but refused none of the offensive cyber or biological tasks tested.
DeepSeek’s New AI Model Is Reportedly 100 Times Cheaper to Run Than Claude Fable 5
DeepSeek’s V4 Flash costs more than 100 times less to run than Claude Fable 5 in benchmark tests, renewing the AI price war.
Alibaba Unveils a 2.4-Trillion-Parameter AI Model That Challenges America’s Frontier Labs
Alibaba’s 2.4-trillion-parameter Qwen 3.8 Max has climbed global AI leaderboards, intensifying China’s challenge to leading US models.
NVIDIA Warns Washington: Do Not Kill Open-Weight AI Models
NVIDIA, Microsoft, Meta, OpenAI, IBM and other technology organisations have urged Washington not to impose sweeping restrictions on open-weight AI. Here is what the letter says and why Anthropic declined to sign.