Core Viewpoint - The article discusses the limitations of traditional AI interactions and introduces MiniCPM-o 4.5, a groundbreaking model that enables real-time, multimodal communication, enhancing human-like interaction capabilities [4][12][40]. Group 1: MiniCPM-o 4.5 Features - MiniCPM-o 4.5 is the first model to achieve full-duplex, multimodal capabilities, allowing it to "see, hear, and speak" simultaneously, thus enabling real-time interaction [4][12]. - The model has a parameter count of 9 billion and has achieved state-of-the-art (SOTA) performance across various benchmarks, scoring 77.6 in the OpenCompass comprehensive evaluation [5][9]. - It outperforms top closed-source models like Gemini 2.5 Flash in key tasks such as visual understanding and document parsing [7]. Group 2: Technical Innovations - MiniCPM-o 4.5 employs a full-duplex architecture that allows continuous input and output without blocking, enabling the model to perceive environmental changes while generating responses [29][36]. - The model features an autonomous interaction mechanism that allows it to determine when to respond based on real-time semantic understanding, eliminating reliance on external tools [33][36]. - It utilizes time alignment and time-division multiplexing to process multimodal streams in real-time, ensuring that input and output are synchronized at a millisecond level [35]. Group 3: User Experience and Comparisons - User experiences with MiniCPM-o 4.5 demonstrate its ability to engage in dynamic interactions, such as providing real-time feedback during drawing games, unlike traditional models that wait for complete inputs [15][16]. - In practical tests, MiniCPM-o 4.5 showed proactive engagement by reminding users about tasks, showcasing its ability to maintain context and provide timely interventions [20][21]. - Comparisons with ChatGPT highlight MiniCPM-o 4.5's superior ability to adapt and respond in real-time, making interactions feel more natural and human-like [16][22]. Group 4: Implications for the Future - The introduction of MiniCPM-o 4.5 signifies a shift towards more human-like AI interactions, where AI can actively participate in conversations rather than merely responding to prompts [41]. - The model's capabilities suggest potential applications in various fields, including smart monitoring, human-computer collaboration, and accessibility support for individuals with disabilities [38]. - The advancements in MiniCPM-o 4.5 reflect a broader trend in the industry towards achieving higher capability density in AI models, moving away from simply increasing parameter counts [40].

Core Viewpoint - Google stock is positioned strongly ahead of its February 4 earnings report, with an 8.94% year-to-date rally to $341.85, despite a general slip in the Dow Jones Industrial Average [1][3] Group 1: Business Performance - Alphabet has shown growth across most divisions, with its Gemini AI platform capturing approximately 20% of the market share from ChatGPT [4] - The core 'Search' business has stabilized, maintaining a dominance close to 90% by February 2026 after previous market share losses [5] - Alphabet has resolved its antitrust issues with Epic Games and secured an $800 million deal related to Unreal Engine, while Waymo successfully closed a $16 billion fundraising round [6] Group 2: Earnings Expectations - Wall Street anticipates Alphabet's earnings per share (EPS) to rise to $2.58 in Q4, a 14.2% increase from the $2.26 forecasted for Q3, reflecting strong analyst optimism [7][8] - A significant surprise in the last quarter was noted at 27% with an EPS of $2.87, indicating that exceeding the consensus by at least 11% is necessary to maintain positive momentum [8][9] - To ensure a rally, Alphabet would need to report an EPS of at least $3.28, matching the previous quarter's surprise [9] Group 3: Industry Risks - The AI sector's overall state poses potential risks, with recent cautious headlines indicating possible instability [12] - Concerns regarding production constraints and energy bottlenecks could leave new data centers underutilized, with sustainability issues arising from high power demands [15] - The rising number of canceled data centers reflects an apparent overreach in promises made by big tech, contributing to the sector's high valuations [16]

Core Viewpoint - The acquisition of xAI by SpaceX, valued at $1.25 trillion, is seen as a strategic move to enhance Musk's portfolio and prepare for an upcoming IPO, despite skepticism regarding the merger's motivations and profitability [1][2][5]. Group 1: Company Strategy - SpaceX is positioning itself as a diversified investment vehicle, appealing to investors who may not be interested in Tesla or X, by offering opportunities in areas like orbital data centers and AI startups [4][5]. - The merger is viewed as a way to capitalize on the current AI investment trend, potentially outpacing competitors like OpenAI and Anthropic by being a first mover in the public market [6][8]. Group 2: Market Context - The timing of the merger is critical, as it aligns with SpaceX's preparations for a historic IPO, which could attract significant retail investor interest [5][6]. - Musk's narrative framing of the AI venture, likening it to his previous successes with Tesla and SpaceX, aims to create a compelling vision for the future of data centers in space [7][8].

With 98.9% success rate and 77 AI agents built autonomously for brands across 8 regions in early access, Nexus by Yellow.ai will move enterprises from “Software as a Service” to “Service as a Software”.SAN MATEO, Calif., Feb. 4, 2026 /PRNewswire/ -- Yellow.ai, a global leader in AI-powered customer and employee experience automation for enterprises, today announced the launch of Nexus, the industry’s first Universal Agentic Interface (UAI). Nexus represents a fundamental shift in how enterprises build, depl ...

Core Insights - Inturai Ventures Corp. has launched new technology advancements in its spatial intelligence platform, enabling health and sleep monitoring using standard wireless signals [1] - The technology allows for health assessments and monitoring of vital signs, movement patterns, and behavioral consistency without the need for cameras or physical contact [1] - The company aims to transform the cost of scalable deployment in various environments, including residential, care, and tactical applications [1] Health Monitoring Capabilities - Inturai has developed a novel approach to generate a general health score using WiFi signals, assessing patterns such as movement consistency and daily activity [1] - This health score provides caregivers and families with an easy-to-understand indicator of overall well-being, facilitating early awareness of potential changes [1] Sleep Pattern Monitoring - The company has proprietary methods for identifying sleep-related patterns using WiFi and mmWave signals, allowing for monitoring of sleep presence and duration without physical contact or wearables [1] - This capability is designed for discreet operation in care settings, emphasizing comfort, privacy, and ease of installation [1] Commercial Expansion - The new capabilities expand the commercial scope of Inturai's spatial intelligence platform, enhancing its value proposition for healthcare, aged care, smart living partners, and defense sectors [1]

闻乐 发自 凹非寺 量子位 | 公众号 QbitAI 较真还得是程序员。 2025已经过去，2026已经到了……但也有较真的科研、程序员，决定站出来整治一下业内的不正之风。 啥事儿呢？ AI学术界的 假开源 问题。 就是发论文的时候说开源，还像模像样挂了GitHub链接，结果后来就没有后来了。 这不，一位匿名AI研究员，直接拿着AI领域最顶的NeurIPS 2024来检验，看看那些被接收的论文，白纸黑字写上了开源的项目，最后究竟有 多少是兑现了的。 不统计不知道，一查锤出大窟窿了—— | 4035 | 98 | 2404 | 1533 | | --- | --- | --- | --- | | 总论文数 | 假/空仓库数 | 真实开源数 | 未提供 GitHub 链接 | 从NeurIPS 2024收录的4035篇论文数据来看，其中真实开源的论文有2404篇； 除了没有提供链接的1533篇之外，有 98 篇论文明确表示了开源，也给出了链接，结果代码仓库点进去不是烂尾就是待建状态。 当然，可能也有错漏，因为这个工作是AI来查的。 但这个"项目"，堪称对事也对人了，有图有真相，有统计有数据， 指名道姓 把单位都放 ...

出品：新浪财经创投Plus 编辑整理：易舍 基于公开数据不完全统计，2026年1月新增登记的私募股权、创业投资基金管理人共8家，同比、环比均下降62.5%；新增备案私募股权投资基金、创业投资 基金分别为140只、408只，合计新增548只，与2025年12月相比减少20.6%，较2025年同期显著增长66.1%。 投资端呈现出相似趋势。1月国内一级市场共发生644起股权投资事件，披露投资金额总和约为529.99亿元，对比2025年同期分别增长2.1%、55.5%，较2025 年年末分别减少16.1%、11.1%。1月平均单笔投资额约为8229.59万元，同比、环比分别增长52.5%、6.0%。 新浪财经创投Plus本月聚焦市场活跃度较高的3家机构，围绕其投资节奏、阶段、行业偏好以及被投项目等方面展开解读。 人工智能、生产制造双线并行 上海、深圳"双核心"布局 结合中国证券投资基金业协会和天眼查公开披露的信息显示，东方富海成立于2006年，投资策略为80%资金投资早期和成长期企业，侧重A轮和B轮，20% 资金投资成熟期企业，主要覆盖信息技术、节能环保、健康医疗、新材料和文化消费赛道。截至目前，机构累计管理规模逾 ...

央广网北京2月4日消息（记者 殷雨婷）1月31日，九源智能计算系统生态联合体2026年度第一次会议在国家会议中心隆重召开。 本次会议以"协同创新·共赢未来"为核心导向，汇聚生态伙伴智慧，共同绘制国产智能计算系统从技术突破到产业落地的跨越式发展蓝图。 聚焦三大核心议题，破解产业协同难题 会议紧扣国产智能计算系统生态建设的关键挑战，设置三大核心研讨方向。 "下一代算力架构专家委员会"聘书颁发。（主办方供图，央广网发） 授聘仪式后，上海人工智能实验室乔宇教授作为专家委员会代表发表了致辞。 议题一：基础软硬件协同创新：围绕芯片、操作系统、中间件及超节点架构的适配优化展开深度探讨，旨在打通软硬件协同调度链路，夯实自主计算生 态根基。 议题二：大规模智能算力调度与集群效能优化：研讨算力资源高效分配、集群协同管理技术方案，助力提升算力利用率与应用落地能力。 议题三：开放创新体系共建：通过开源项目、比赛及课程活动，搭建开放共享的技术平台，推动创新资源整合与产业生态扩容，加速技术成果转化。 权威嘉宾云集，共话产业未来 中国工程院院士、清华大学教授郑纬民出席大会并致开幕辞，随后作题为《"九源"智能计算系统生态联合体2025年工作 ...

（文/观察者网 吕栋 编辑/张广凯） 2026年伊始，国内再度掀起AI大战热潮，而全球人工智能产业似乎也正迎来一个微妙而关键的转折 点。如果说过去三年是"百模大战"的喧嚣期，那当下已走向"价值落地"的深水区。 "卡脖子"无法阻止中国模型 目前业界公认，美国在原生大模型（Foundation Models）的绝对能力上仍领先中国数月。美国有更强的 芯片，模型在"想得深"上更有优势，中国模型则在"用得起"上无出其右。 近日在世界经济论坛上，微软总裁萨提亚·纳德拉直言："得让AI来干点真正有用的事了，要去创造实际 价值，如果这些token不能让医疗更好、教育更棒、政府更高效、企业更有竞争力，那就白搭。" 这句话像是硅谷巨头的反思，但也更像中美AI竞赛下半场的发令枪，比赛的规则已从单纯的"技术参数 比拼"转向了"应用落地"的马拉松。 当我们审视AI带来的这场技术变局时，会发现一个"双寡头"格局正在浮现：美国依然在"天花板"（最强 算力与最强推理）上保持领先，而中国凭借完整的产业链和务实落地的算力服务，正将AI转化为直接 的"工业引擎"，在"地板"（工业渗透与规模化应用）上实现追赶与反超。 再直接点说，美国在原始创新和 ...

以下文章来源于福布斯 ，作者Forbes 福布斯 . "Forbes福布斯"品牌始终秉承企业家精神和创新精神，坚持专业、公正、创新和进取的价值观，中国地区业务涵盖广告、咨询、调研、策 划、会议及展览服务等。 来源丨福布斯（ID： forbes_china ） 作者丨Richard Niev 翻译丨 Lemin 去年 9 月，风头正劲的 氛围编程 初创公司 Cognition 宣布完成 4 亿美元融资，公司估值一举跃升至 102 亿美元。 伴随公司 估值 上涨 的还有 三位年轻创始人 的身家 ——他们已 成为生成式人工智能热潮中最新的造富赢家。 三人中最年轻的 Yan 曾入选 Neo Scholar 项目 ，该计划由 Facebook 早期投资人阿里 · 帕托维 （ Ali Partovi ） 发起，旨 在挖掘在 校大学生中的 潜力科技人才。与 蒂尔奖学金（ Thiel Fellows ） 计划不同， 参加 Neo 项目 无需退学创业，但 Yan 仍选择从哈佛大学 退学 ，全 身心投入 Cognition 的 创建 。 目前尚无信息披露 Hao 的持股比例为何高于 Yan 和 Wu ，三位创始人未回应《福布斯》 ...