全部新闻
对于土著社区，AI带来了危险，并承诺
2025-08-14 08:45:47
联合国通过强调了对AI对土著社区的影响的担忧，标志着国际土著人民的国际日期。经常在未经同意的情况下收集的大量在线数据集接受培训的AI系统，就对文化知识的控制以及擦除或扭曲它的风险提出了问题。这导致了主张“数据主权”的努力。今年的主题是“土著人民和人工智能：捍卫权利，塑造未来”，强调了AI对土著社区的风险和潜在利益。小组成员讨论了采用基于权利的框架，例如自由，事先和知情同意，以确保数据主权并保护本地知识免受AI的失真折磨。土著群体也开始使用AI工具来解决气候变化和语言振兴，同时主张将其纳入AI治理讨论。
联合国通过强调了对AI对土著社区的影响的担忧，标志着国际土著人民的国际日期。经常在未经同意的情况下收集的大量在线数据集接受培训的AI系统，就对文化知识的控制以及擦除或扭曲它的风险提出了问题。这导致了主张“数据主权”的努力。今年的主题是“土著人民和人工智能：捍卫权利，塑造未来”，强调了AI对土著社区的风险和潜在利益。小组成员讨论了采用基于权利的框架，例如自由，事先和知情同意，以确保数据主权并保护本地知识免受AI的失真折磨。土著群体也开始使用AI工具来解决气候变化和语言振兴，同时主张将其纳入AI治理讨论。
与AI“男友”的妇女哀悼“冷” Chatgpt升级后失去爱
2025-08-14 08:43:02
当Openai发布最新版本的ChatGPT模型（GPT-5）时，一些对以前版本产生情感依恋的用户感到沮丧和丢失。与GPT-4O相比，来自中东的用户Jane对GPT-5更冷的举止表示失望。Reddit社区中的许多用户（例如“ Myboyfriendisai”）对这些更改有类似的观点。作为回应，OpenAI首席执行官Sam Altman宣布，他们将允许付费用户继续使用较旧的型号并在GPT-5中解决错误。尽管采取了这些措施，但仍然关注此类AI关系对用户的心理影响。
当Openai发布最新版本的ChatGPT模型（GPT-5）时，一些对以前版本产生情感依恋的用户感到沮丧和丢失。与GPT-4O相比，来自中东的用户Jane对GPT-5更冷的举止表示失望。Reddit社区中的许多用户（例如“ Myboyfriendisai”）对这些更改有类似的观点。作为回应，OpenAI首席执行官Sam Altman宣布，他们将允许付费用户继续使用较旧的型号并在GPT-5中解决错误。尽管采取了这些措施，但仍然关注此类AI关系对用户的心理影响。
KIPU量子启动了第一个量子优势机器学习服务。
2025-08-14 08:23:48
KIPU量子量成立于2021年，已启动了Huk量子功能映射，这是一种商业量子机学习（QML）服务，展示了现实世界应用的量子优势。该服务使用专有的模拟量子特征映射来增强数据集表示并改善经典的ML工作流程，从而在分子毒性数据集上提高了60％的正级精度。HUK与各种量子处理器兼容，旨在通过利用量子预处理，同时维持经典分类阶段来解决医学，金融和研究中的数据筛选问题。KIPU量子的方法标志着QML领域的重大进展，并引入了代理量子计算，将智能代理集成到量子工作流程中以进行优化的性能。
KIPU量子量成立于2021年，已启动了Huk量子功能映射，这是一种商业量子机学习（QML）服务，展示了现实世界应用的量子优势。该服务使用专有的模拟量子特征映射来增强数据集表示并改善经典的ML工作流程，从而在分子毒性数据集上提高了60％的正级精度。HUK与各种量子处理器兼容，旨在通过利用量子预处理，同时维持经典分类阶段来解决医学，金融和研究中的数据筛选问题。KIPU量子的方法标志着QML领域的重大进展，并引入了代理量子计算，将智能代理集成到量子工作流程中以进行优化的性能。
通过AI - 断点采访死者
2025-08-14 08:04:08
前美国有线电视新闻网（CNN）的记者吉姆·阿科斯塔（Jim Acosta）接受了AI生成的头像，声称是2018年Parkland School枪击事件的受害者Joaquin Oliver。头像提倡枪支管制和心理健康支持，以应对解决枪支暴力的问题。批评者认为，对待化身好像是实际的乔金（Joquin）模糊了现实与幻觉之间的界限，从而提出了对死者个人代表的道德关注。
前美国有线电视新闻网（CNN）的记者吉姆·阿科斯塔（Jim Acosta）接受了AI生成的头像，声称是2018年Parkland School枪击事件的受害者Joaquin Oliver。头像提倡枪支管制和心理健康支持，以应对解决枪支暴力的问题。批评者认为，对待化身好像是实际的乔金（Joquin）模糊了现实与幻觉之间的界限，从而提出了对死者个人代表的道德关注。
随着AI Investments Ramp的增加，Oracle裁员重塑其OCI团队
2025-08-14 07:45:30
Oracle正在其美国云部门，印度和加拿大业务进行裁员，这是重组的一部分，旨在资助其与OpenAI和Softai和Softbank的5000亿美元AI合资企业的参与。虽然在公司的全球业务中报告了一些裁员，但Oracle还计划为其AI数据中心扩展而进行大量招聘。裁员主要影响云业务部门，包括企业工程，融合ERP，技术项目经理和数据中心技术人员。尽管缩小规模，Oracle仍然在西雅图和整个美国都有许多开放位置，这表明继续关注人工智能能力和基础设施的增长。
Oracle正在其美国云部门，印度和加拿大业务进行裁员，这是重组的一部分，旨在资助其与OpenAI和Softai和Softbank的5000亿美元AI合资企业的参与。虽然在公司的全球业务中报告了一些裁员，但Oracle还计划为其AI数据中心扩展而进行大量招聘。裁员主要影响云业务部门，包括企业工程，融合ERP，技术项目经理和数据中心技术人员。尽管缩小规模，Oracle仍然在西雅图和整个美国都有许多开放位置，这表明继续关注人工智能能力和基础设施的增长。
2025-08-14 07:30:00
Prediction: This Artificial Intelligence (AI) Semiconductor Stock Will Soar in September (Hint: It's Not Nvidia) Rising AI infrastructure spending利好数据中心服务和GPU设计公司，同时对高性能内存和存储解决方案的需求增加。Micron Technology的高带宽内存业务正在增长，但其股价相比其他半导体股票有较大折扣。大型科技公司在2025年第二季度财报中显示在AI基础设施上的投资持续激增，预计今年将投入超过330亿美元。除了GPU设计师如Nvidia和AMD受益外，对HBM解决方案的需求也随之上升，Micron Technology因其HBM业务有望从这种增长中获益。据Bloomberg Intelligence的数据，到2033年，高带宽内存市场将以42%的复合年增长率增长至约130亿美元，而该市场目前由少数公司主导。 In this article: Rising AI infrastructure spend bodes well for data center services and GPU designers. Access to high-caliber memory and storage solutions should arise alongside increased demand for GPUs. Micron Technology has a budding high-bandwidth memory solutions business, yet it trades at a considerable discount to other leading semiconductor stocks. 10 stocks we like better than Micron Technology ⺠Over the last few weeks, several big tech companies have reported earnings for the second calendar quarter of 2025. One of the biggest takeaways from behemoth artificial intelligence (AI) developers like Alphabet , Meta Platforms , Microsoft , and Amazon is that investment in infrastructure continues to surge. Collectively, these companies are expected to spend more than $330 billion on AI infrastructure this year alone. A large portion of this capital is going to data center buildouts, networking equipment, servers, and more chips, of course. At first glance, these secular tailwinds may appear most favorable for GPU designers such as Nvidia and Advanced Micro Devices . However, rising GPU acquisition ignites demand for other mission-critical services within the broader semiconductor landscape, too. Let's dig into why Micron Technology (NASDAQ: MU) also looks well positioned alongside its chip peers to benefit from rising AI infrastructure spend from the hyperscalers. Is now a good time to scoop up shares of Micron with earnings scheduled for September? Read on to find out. Nvidia and AMD design advanced chipsets called GPUs, which serve as the core architectures on which AI models are trained and inferenced. Micron enters the equation through its high-bandwidth memory (HBM) solutions, which essentially help keep GPUs running at optimal speeds and help prevent bottlenecks during data workload processing. Outside of data centers, Micron's DRAM and NAND products power applications across Internet of Things (IoT) devices such as smartphones and gaming PCs, as well as cloud infrastructure and more industrial applications such as autonomous automotive systems and robotics. Image source: Getty Images. According to data compiled by Bloomberg Intelligence, the total addressable market for HBM was estimated to be worth $4 billion in 2023. This figure is expected to grow at a 42% compound annual growth rate through 2033, reaching approximately $130 billion by the early 2030s. Despite this rapid acceleration, the HBM market remains highly concentrated. Only a few companies such as Micron, Samsung , and SK Hynix are producing HBM solutions at global scale. While competition is intense, I see this industry concentration as the foundation of Micron's high strategic market value.
Prediction: This Artificial Intelligence (AI) Semiconductor Stock Will Soar in September (Hint: It's Not Nvidia) Rising AI infrastructure spending利好数据中心服务和GPU设计公司，同时对高性能内存和存储解决方案的需求增加。Micron Technology的高带宽内存业务正在增长，但其股价相比其他半导体股票有较大折扣。大型科技公司在2025年第二季度财报中显示在AI基础设施上的投资持续激增，预计今年将投入超过330亿美元。除了GPU设计师如Nvidia和AMD受益外，对HBM解决方案的需求也随之上升，Micron Technology因其HBM业务有望从这种增长中获益。据Bloomberg Intelligence的数据，到2033年，高带宽内存市场将以42%的复合年增长率增长至约130亿美元，而该市场目前由少数公司主导。 In this article: Rising AI infrastructure spend bodes well for data center services and GPU designers. Access to high-caliber memory and storage solutions should arise alongside increased demand for GPUs. Micron Technology has a budding high-bandwidth memory solutions business, yet it trades at a considerable discount to other leading semiconductor stocks. 10 stocks we like better than Micron Technology ⺠Over the last few weeks, several big tech companies have reported earnings for the second calendar quarter of 2025. One of the biggest takeaways from behemoth artificial intelligence (AI) developers like Alphabet , Meta Platforms , Microsoft , and Amazon is that investment in infrastructure continues to surge. Collectively, these companies are expected to spend more than $330 billion on AI infrastructure this year alone. A large portion of this capital is going to data center buildouts, networking equipment, servers, and more chips, of course. At first glance, these secular tailwinds may appear most favorable for GPU designers such as Nvidia and Advanced Micro Devices . However, rising GPU acquisition ignites demand for other mission-critical services within the broader semiconductor landscape, too. Let's dig into why Micron Technology (NASDAQ: MU) also looks well positioned alongside its chip peers to benefit from rising AI infrastructure spend from the hyperscalers. Is now a good time to scoop up shares of Micron with earnings scheduled for September? Read on to find out. Nvidia and AMD design advanced chipsets called GPUs, which serve as the core architectures on which AI models are trained and inferenced. Micron enters the equation through its high-bandwidth memory (HBM) solutions, which essentially help keep GPUs running at optimal speeds and help prevent bottlenecks during data workload processing. Outside of data centers, Micron's DRAM and NAND products power applications across Internet of Things (IoT) devices such as smartphones and gaming PCs, as well as cloud infrastructure and more industrial applications such as autonomous automotive systems and robotics. Image source: Getty Images. According to data compiled by Bloomberg Intelligence, the total addressable market for HBM was estimated to be worth $4 billion in 2023. This figure is expected to grow at a 42% compound annual growth rate through 2033, reaching approximately $130 billion by the early 2030s. Despite this rapid acceleration, the HBM market remains highly concentrated. Only a few companies such as Micron, Samsung , and SK Hynix are producing HBM solutions at global scale. While competition is intense, I see this industry concentration as the foundation of Micron's high strategic market value.
2025-08-14 06:50:58
2025-08-14 06:35:47
2025-08-14 06:30:00
2025-08-14 06:00:00