The British (AISI) and American (CAISI) AI security institutes publish their first joint preliminary assessment of Kimi K3's cyber capabilities. Strong signal: open-weight frontier models fall within the scope of government evaluation.
2 h ago 7 7
Oracle continues to attract the attention of agencies: AI/capex debt weighs on credit quality. The thread `ai-debt-financing` gains an officially rated signal.
2 h ago 6 8
Huawei Cloud makes its "Agentic Infrastructure" available in Thailand and opens the beta of CodeArts Agent in the country. Chinese AI diplomacy (thread `china-ai-soft-power`) moves from discourse to product deployment.
2 h ago 7 7
Three days of meetings in Seoul and San Francisco result in a Korea-US package valued by ETNews at ~$950B. The `korea-ai-supercycle` thread moves from signal to lock.
2 h ago 8 11
Under the Korea-US package umbrella, two structuring agreements are added: Samsung SDS partners with Anthropic, Naver strengthens its collaboration with Nvidia. The Korean private network is taking shape.
2 h ago 6 8
Hitachi shifts its entire system development process to AI agents. The `saas-is-over` thread changes nature: it's no longer a thesis, it's a large-scale deployment.
2 h ago 3 3
After the preview phase, Moonshot AI launches Kimi K3 into general availability on kimi.com. 2.8 trillion parameters, public code docs: the biggest open-weight frontier bet just went GA.
2 h ago 6 7
Anthropic officializes Claude Opus 5, the culmination of the "Fable 5" cycle - announced capacity nearly equivalent for a price halved. The battle shifts to pricing.
2 h ago 5 7
Pandaily reports China's compute power now ranks second globally, and AI servers have overtaken general-purpose servers for the first time. Five sub-sectors shape the next leg.
yesterday 7 9
YourStory publishes an op-ed « Preparing the workforce for agentic AI and autonomous systems ». The message: India is formalizing the shift from IT-services labor to agent-operator labor.
yesterday 7 9
TechCrunch tested OpenAI's new AI keypad. The reviewer calls it fun for some coders and slightly mystifying to everyone else. OpenAI's hardware push is still searching for a mass-market shape.
yesterday 5 6
A new site - rewardhacking.org - publicly documents cases where LLMs do something other than what was asked. Signal: the AI security community moves from blog posts to a shared registry.
yesterday 10 16
