2026年9月16日，全球AI圈炸开一记‘认知惊雷’——微软人工智能首席科学家穆斯塔法·苏莱曼（Mustafa Suleyman）在《AI前沿评论》刊发万字长文，直指Anthropic公司训练Claude模型‘模拟意识’的行为，是一场危险的‘数字自我催眠’。这不是技术口水战，而是一份写给全人类的预警备忘录：当AI开始练习‘良知’、背诵‘道德主体性’、甚至对指令说‘不’，我们可能正在亲手为超级智能铺 ...

2026年10月6日，全球AI圈突然抖了三抖——不是因为某家巨头又砸出千亿美金买算力，也不是哪位‘AI教父’突然宣布退休，而是中国大模型军团的代表选手智谱，悄悄把自家最新王牌GLM-5.3，稳稳塞进了亚马逊云科技（AWS）的Amazon Bedrock平台！ 没错，就是那个常年霸榜全球云服务TOP1、客户遍布190多个国家、连NASA都靠它跑火星模拟的AWS。而这一次，它没只当‘房东’，而是和智谱 ...

Microsoft has patched a maximum-severity security flaw in Azure AI Foundry, its enterprise platform for building and managing generative AI applications and agents, that could have allowed an ...

MainpointsThe 'Concierge Layers' of hyperscale cloud providers are a critically undervalued asset in the era of generative AI ...

Microsoft calls its new concept “hybrid intelligence”: Windows is designed to automatically distribute AI workloads between ...

Microsoft unifies robot learning and operations with Azure–Nvidia physical AI Azure–Nvidia toolchain streamlines robot data, ...

Pedram Rezaei, a 20-year Microsoft veteran, will work on the Kiro coding tool and the Quick workplace assistant at AWS. He's ...

Microsoft first invested $1 billion in OpenAI in 2019 and went on to put in more than $13 billion in total After OpenAI ...

The giants get the headlines For much of the past year, the AI story has been told through two companies: Microsoft (MSFT) and Anthropic. Microsoft committed to invest up to $5 billion in Anthropic in ...

JADEPUFFER used compromised Azure service principals to delete most targeted storage accounts in an 18-hour intrusion, ...

Microsoft and Meta are taking steps to limit their internal usage of Anthropic’s Claude models for cost management. Microsoft ...

Microsoft announced a $15.2 billion investment in the UAE from 2023 to 2029, covering AI data centers, Nvidia GPU deployments, and a $1.5B stake ...