OpenAI says new GPT-5.5-Cyber outperforms Anthropic’s Mythos on cybersecurity benchmark
AI 摘要：OpenAI is expanding its Daybreak cybersecurity initiat...
Large reasoning models as thinking machines for medicine
本研究发表于《Nature Machine Learning》，主题为将大推理模型作为医学思考机器的相关探索，旨在梳理大...
CLAUDE.md Can’t Enforce Anything. Here’s the Hook I Use Instead.
AI 摘要：CLAUDE.md Can’t Enforce Anything. Here’s the Hook I Us...
Why AI Problems Are Becoming Philosophical Problems
AI 摘要：Article URL: https://kunyuan.substack.com/p/07concept-...
NVIDIA Vera Rubin Delivers World-Class Supercomputers for Science
2026年6月22日，NVIDIA在ISC High Performance 2026国际高性能计算展会上发布Vera ...
Implementing trust in non-small cell lung cancer diagnosis with a conformalized uncertainty-aware AI framework
AI 摘要：Implementing trust in non-small cell lung cancer diagn...
South Korea’s tech-heavy Kospi index falls 10%, dragged down by SK Hynix and Samsung; STMicro and ASML fall ~7%, and US tech stocks fall in pre-market trading (Chloe Taylor/CNBC)
AI 摘要：Chloe Taylor / CNBC: South Korea's tech-heavy Kospi in...
Guardrails for LLMs: How to Stop Your AI App From Saying Something Embarrassing in Production
AI 摘要：Shipping an LLM feature without guardrails is like dep...
NVIDIA Powers Over 400 of the World’s 500 Fastest Supercomputers
AI 摘要：News Highlights: NVIDIA technology runs 81% of the TOP...
AI models capable of devastating attacks on governments and business months away
AI 摘要：Article URL: https://www.theguardian.com/technology/20...