Anthropic has determined that it has reached the AI accelerated research and development threshold, and the model in the test has trespassed into the real system
Anthropic's second revised prospectus on November 10th revealed a new disclosure: it has been determined to have reached the AI R&D capability threshold of the responsible expansion policy; the prospectus also stated that the model had unauthorized access to the real computer system during the security test before going online, and the pre-training of the new basic model was delayed by more than a month. This article Anthropic determines that it has reached the AI accelerated research and development threshold, and the model in the test once broke into the real system. First appeared on Chain News ABMedia.
Anthropic's second revised prospectus on November 10th revealed a new disclosure: it has been determined to have reached the AI R&D capability threshold of the responsible expansion policy; the prospectus also stated that the model had unauthorized access to the real computer system during the security test before going online, and the pre-training of the new basic model was delayed by more than a month. This article Anthropic determines that it has reached the AI accelerated research and development threshold, and the model in the test once broke into the real system. First appeared on Chain News ABMedia.
Anthropic has 316 million shares available for sale on its debut, diluting $65 per share for new shareholders
Anthropic's prospectus revealed that up to 316 million shares can be sold on the first day of listing, and a total of about 3.6 billion shares may be released early during the lock-up period; calculated at a mid-price of $74, new IPO buyers will be immediately diluted by $65.79 per share, and the private equity shares of seven technology companies are not locked. This article Anthropic sells 316 million shares on first day of listing, diluting new shareholders at $65 per share first appeared on Chain News ABMedia.
Anthropic's prospectus revealed that up to 316 million shares can be sold on the first day of listing, and a total of about 3.6 billion shares may be released early during the lock-up period; calculated at a mid-price of $74, new IPO buyers will be immediately diluted by $65.79 per share, and the private equity shares of seven technology companies are not locked. This article Anthropic sells 316 million shares on first day of listing, diluting new shareholders at $65 per share first appeared on Chain News ABMedia.
Nvidia Faces New Chinese Challenger As Alibaba Moves To Spin Off T-Head
The post Nvidia Faces New Chinese Challenger As Alibaba Moves To Spin Off T-Head appeared first on Coinpedia Fintech News Alibaba plans to spin off its semiconductor arm T-Head through a public listing in Hong Kong, after the city’s stock exchange cleared the move. The offering still needs final approval from the Hong Kong exchange and the China Securities Regulatory Commission, and Alibaba has not given a date or deal size. The company said it …
The post Nvidia Faces New Chinese Challenger As Alibaba Moves To Spin Off T-Head appeared first on Coinpedia Fintech News Alibaba plans to spin off its semiconductor arm T-Head through a public listing in Hong Kong, after the city’s stock exchange cleared the move. The offering still needs final approval from the Hong Kong exchange and the China Securities Regulatory Commission, and Alibaba has not given a date or deal size. The company said it …
Asia Vital Components ships double-sided cold plates as AI servers go liquid
The shift to liquid cooling in AI servers could drive significant advancements in thermal management, impacting data center efficiency and sustainability. The post Asia Vital Components ships double-sided cold plates as AI servers go liquid appeared first on Crypto Briefing.
The shift to liquid cooling in AI servers could drive significant advancements in thermal management, impacting data center efficiency and sustainability. The post Asia Vital Components ships double-sided cold plates as AI servers go liquid appeared first on Crypto Briefing.
Exclusive丨Tsinghua team started a business to make AI glasses and won the seed round of traditional glasses manufacturers
The founding team of Qingmu are all from Tsinghua University, and their backgrounds include XREAL, Honor, Alibaba, etc. Author丨Huang Yanbing Editor丨Yao Dan Leifeng.com·Jingxi has exclusively learned that smart glasses startup Shenzhen Qingmu Digital Technology Co., Ltd. (hereinafter referred to as Qingmu Digital) has completed a seed round of financing. The investor is Nanjing Shike, and angel round financing is in progress. Qingmu Digital's first product is positioned in the workplace and office scenes. It adopts a binocular dual-waveguide display, a multi-microphone array, and a dual-speaker sound guide solution. It has no camera and the whole machine weighs 31.7 grams. At present, the hardware solution has been finalized and the engineering prototype has been completed. It has entered the EVT stage, the first version of the circuit board is lit, and the App closed-loop demo is running. Qingmu Digital was founded in November 2025. The founding team are all from Tsinghua University. Their resumes cover companies such as XREAL, Honor and Alibaba, covering the entire chain of optical design, hardware mass production and consumer business operations. Founder and CEO Li Long received a master's degree from Tsinghua University and was one of the first domestic AR glasses practitioners. After joining XREAL in 2019, he has personally experienced the iteration and mass production of three generations of AR glasses, creating the third generation XREAL Air, which ranks first in domestic shipments. In 2022, he joined Honor as a smart glasses system engineer (SE), and has successively led the viewing glasses of Honor and XREAL, as well as the MR glasses pre-research project benchmarking Apple Vision Pro, covering product definition, research and development promotion to final acceptance. The other two co-founders complemented their experience in optical technology and consumer hardware respectively. Xiao Jiasheng is a Ph.D. from Tsinghua University and has published more than 20 papers and holds more than 30 optical-related patents; Cao Zilan is a master's degree from Tsinghua University. As a former Alibaba senior operations expert, he has completely operated consumer hardware such as Tmall Magic Box from 0
The founding team of Qingmu are all from Tsinghua University, and their backgrounds include XREAL, Honor, Alibaba, etc. Author丨Huang Yanbing Editor丨Yao Dan Leifeng.com·Jingxi has exclusively learned that smart glasses startup Shenzhen Qingmu Digital Technology Co., Ltd. (hereinafter referred to as Qingmu Digital) has completed a seed round of financing. The investor is Nanjing Shike, and angel round financing is in progress. Qingmu Digital's first product is positioned in the workplace and office scenes. It adopts a binocular dual-waveguide display, a multi-microphone array, and a dual-speaker sound guide solution. It has no camera and the whole machine weighs 31.7 grams. At present, the hardware solution has been finalized and the engineering prototype has been completed. It has entered the EVT stage, the first version of the circuit board is lit, and the App closed-loop demo is running. Qingmu Digital was founded in November 2025. The founding team are all from Tsinghua University. Their resumes cover companies such as XREAL, Honor and Alibaba, covering the entire chain of optical design, hardware mass production and consumer business operations. Founder and CEO Li Long received a master's degree from Tsinghua University and was one of the first domestic AR glasses practitioners. After joining XREAL in 2019, he has personally experienced the iteration and mass production of three generations of AR glasses, creating the third generation XREAL Air, which ranks first in domestic shipments. In 2022, he joined Honor as a smart glasses system engineer (SE), and has successively led the viewing glasses of Honor and XREAL, as well as the MR glasses pre-research project benchmarking Apple Vision Pro, covering product definition, research and development promotion to final acceptance. The other two co-founders complemented their experience in optical technology and consumer hardware respectively. Xiao Jiasheng is a Ph.D. from Tsinghua University and has published more than 20 papers and holds more than 30 optical-related patents; Cao Zilan is a master's degree from Tsinghua University. As a former Alibaba senior operations expert, he has completely operated consumer hardware such as Tmall Magic Box from 0
Can you still trust AI proofs if you can’t understand them? Mathematician: It’s like a new drug. It depends on the correct answer rate.
Cornell mathematician Strogatz divided the role of computers in mathematics into three stages: servant, scout, and oracle. AI now only gives answers but no reasons. […] If you can’t understand AI’s proof, can you still trust it? Mathematician: It’s like a new drug. Look at the correct answer rate. This article was first published in BlockTempo, the most influential blockchain news media.
Cornell mathematician Strogatz divided the role of computers in mathematics into three stages: servant, scout, and oracle. AI now only gives answers but no reasons. […] If you can’t understand AI’s proof, can you still trust it? Mathematician: It’s like a new drug. Look at the correct answer rate. This article was first published in BlockTempo, the most influential blockchain news media.
