算法爱好者 . 算法是程序员的内功！「算法爱好者」专注分享算法相关文章、工具资源和算法题，帮程序员修炼内 功。 以下文章来源于算法爱好者 ，作者小蒜 深夜爆料：前高管群内怒指"几十亿假账" 12 月 15 日深夜，互联网圈突然炸开锅。 前 360 相关高管玉红在多个微信群组和朋友圈公开发文，直指 360 集团创始人周鸿祎存在财务造假问 题。 他言辞激烈，不仅声称"帮周鸿祎做假账就至少几十亿，我都有数据"，还具体点出是通过游戏业务"自充 值"虚增流水，甚至放话要"全网公布"证据。 除了核心指控，玉红还宣泄强烈不满，直言"认识周鸿祎是我一生的耻辱"，暗示自己因此被"坑惨了"，蒙 受巨大经济损失。 有群友证实了这些言论的真实性，不少人起初怀疑他账号被盗，玉红还特意在群内澄清账号安全。 火速回应：360 否认指控，周鸿祎转发声明 爆料发酵不到一天，12 月 16 日下午 18:00 左右，360 集团迅速发布严正声明，周鸿祎本人也同步转 发。 声明首先厘清了玉红的身份：2014 年因旗下公司被 360 收购入职，2015 年就因个人原因离职，在职期 间从未担任集团核心管理层职务，直接反驳了外界对其"前核心高管"的认知 ...

以下文章来源于伯乐在线 ，作者伯小乐 伯乐在线 . "前脚募 80 亿，后脚 75 亿拿去理财"的消息一传开，瞬间在投资者社区和网友圈炸了锅，风波就此引 爆。 网友吵翻了：是合规操作还是圈钱摆烂？ 争议的核心的是"预期差"，网友和投资者分成了鲜明的两派。 反对声里满是失望，不少人吐槽摩尔线程是"上市理财概念股"，质疑它根本没打算好好搞研发。 有网友直言："我们投钱是盼着它突破 GPU 技术壁垒，不是让它拿着股民的钱稳赚利息的"，还有人拿同 赛道的寒武纪对比，觉得它"明显是奔着融资上市来的，不如踏实做事的企业"。 伯乐在线分享IT互联网职场和精选干货文章（原域名已不再维护）。组织维护10万+star的开源技术资源 库，包括：Python, Java, C/C++, Go, JS, CSS, Node.js, PHP, .NET 等 上市刚一周就官宣理财，风波如何引爆？ 被称为"国产 GPU 第一股"的摩尔线程，12 月 5 日刚登陆科创板，IPO 募集资金总额接近 80 亿元，扣 除发行费用后，实际到手的净额约 75.76 亿元。 按照招股书规划，这笔钱本该用在三大芯片研发项目和补充流动资金上，市场也正是冲着它 ...

Core Viewpoint - OpenAI's GPT-5.2 has received significant criticism and is perceived as underperforming compared to Google's Gemini 3 Pro, which is seen as the true leader in AI capabilities [1][2][8]. Performance Comparison - GPT-5.2 has an ability index (ECI) score of 152, ranking just below Gemini 3 Pro [3]. - In various benchmark tests, GPT-5.2 did not dominate as expected, particularly in the FrontierMath challenge where it only excelled in lower tiers [4]. - In the Chess Puzzles test, GPT-5.2 achieved first place, but it performed worse than GPT-5.1 in the SimpleQA Verified test, indicating a decline in reliability [5][8]. Benchmark Results - Multiple third-party evaluations show that GPT-5.2 is not meeting expectations and is outperformed by Gemini 3 in several tests [8][10]. - In a detailed performance table, GPT-5.2 scored 73.61 in global average reasoning, lower than competitors like Claude 4.5 Opus and Gemini 3 Pro [12]. - Users have expressed dissatisfaction with GPT-5.2, suggesting it has regressed in user experience and natural language processing [13][56]. OpenAI's Response - OpenAI has prioritized improvements to ChatGPT and halted AGI development, indicating a serious internal response to the competitive pressure from Google [14][15]. - The company is reportedly focusing on optimizing ChatGPT and preparing for a larger release in early 2026 [44][45]. Market Position - OpenAI's market share remains significant, with over 800 million weekly users and a 76% market share in consumer AI [51]. - However, the emergence of Gemini 3 Pro poses a substantial threat to OpenAI's business, as it has developed technologies that can rival or surpass OpenAI's models [52]. User Feedback - Users have noted that GPT-5.2 lacks warmth and has become less user-friendly, with some describing it as "cold" and "robotic" [56]. - In visual reasoning and 3D model generation, GPT-5.2 has been reported to perform slower and at a higher cost compared to Gemini 3 [59]. Future Developments - OpenAI is exploring new projects, including advertising and e-commerce integrations, while continuing to enhance its AI technologies for enterprise applications [48][49]. - Upcoming releases may include new AI image models, but initial tests suggest these may not meet expectations either [90].

往期趣图 （点击下方图片可跳转阅读） ↓↓↓ 写垃圾代码的你 vs 读垃圾代码的你 ...

7、 外包转正请客，大家激动点了4k多的菜，结账时他只付了2k，余下大家A。同事：早知道就不吃了 字节一校招生感叹：字节真的有钱，年包 180万，先发了一波每月几万期权，最近又补了一笔豆包期 权… - EOF - 推荐阅读 点击标题可跳转 0、 "讨厌X里文化！公司空降X里系高管后差点黄了。干 1 年大家都裁了，包括他们自己。" 1、 "做一个APP接入微信支付，主要功能仿摇一摇，只要摇一下，他的微信零钱都… " 2、 吵翻了！CTO 不搞管理反而沉迷写代码改 Bug，还靠这保住百万订单 3、 外包牛马傻眼！公司错发邮件曝真相：甲方付 3.3 万，到手才 2 万！ 4、 "夫妻俩都在某为工作的，咋搞娃？" "外包出去让合作方带，只需拉通/对齐/再…" 5、 "（女生）和大厂程序员接触，他白天不主动发信息，晚上下班了才联系，这正常吗？" 6、 "男友写代码，一个月 1.7 万，他辞职去跑外卖，这是什么逻辑啊！！" ...

Core Viewpoint - JetBrains has decided to discontinue its Fleet IDE, which has been in public preview since its launch in 2021, and will focus on a new intelligent development environment called Air, indicating a shift in strategy to adapt to market trends and developer needs [3][4][8]. Group 1: Fleet IDE Discontinuation - Fleet IDE was intended to be a lightweight, collaborative tool that could adapt to remote projects, developed in response to the competitive pressure from Microsoft’s Visual Studio Code [3]. - Despite some initial interest, the majority of developers preferred the established IntelliJ tools due to its robust plugin ecosystem and the uncertainty surrounding Fleet's prolonged preview status [4]. - JetBrains announced that Fleet will no longer be available for download after December 22, 2025, citing confusion among users and a diversion of company resources as reasons for this decision [4][5]. Group 2: Transition to Air - The new tool, Air, is built on the Fleet platform and aims to introduce a novel workflow where developers guide an intelligent assistant to generate code, which they then review [5][8]. - Air is currently in public preview and requires a subscription to Anthropic services for access, with plans to support additional operating systems and features in the future [7]. - The decision to pivot to Air reflects JetBrains' ongoing strategy to keep pace with the evolving landscape of AI development tools, similar to offerings from AWS, Google, and Microsoft [8]. Group 3: Developer Reactions - Some developers expressed disappointment over the discontinuation of Fleet, feeling that it had the potential to compete effectively with VS Code [8]. - There are concerns among developers that instead of creating new tools, JetBrains should integrate intelligent features into its existing, mature IDEs, which are already well-regarded [9].

Core Viewpoint - Meta's commitment to the "metaverse" is shifting as the company plans to cut jobs in its Reality Labs division and redirect resources towards the unexpectedly successful AI smart glasses [3][4][9]. Group 1: Job Cuts and Strategic Shift - Meta is preparing to lay off 10% to 30% of employees in its Reality Labs, which focuses on VR headsets and VR-based social networks [6][7][28]. - The layoffs signal a strategic resource reallocation rather than a complete abandonment of the metaverse vision, with funds being redirected to the more successful AR glasses and wearable devices [9][17]. - The Reality Labs division has incurred over $70 billion in losses over the past four years, prompting a reassessment of investment priorities [18][23]. Group 2: Success of AI Smart Glasses - Meta's collaboration with Ray-Ban in 2021 resulted in smart glasses that include a camera and microphone, which have now evolved into AI-enabled devices allowing voice interaction [12][13]. - The smart glasses have significantly outperformed internal sales expectations, contrasting with the slow consumer adoption of VR headsets [14][15]. - Meta's spokesperson confirmed the shift in investment focus towards AI glasses and wearables, emphasizing that this does not indicate broader organizational changes [17]. Group 3: Design and Innovation - Meta is enhancing the design and fashion aspects of its smart glasses, with the appointment of former Apple designer Alan Dye to lead a new creative studio within Reality Labs [30][31]. - The company aims to integrate technology with thoughtful design to improve user interaction and connectivity [32][33]. - This strategic pivot suggests that the path to future innovation may not solely rely on creating a new virtual world but rather on enhancing existing technologies [34].

Core Viewpoint - NVIDIA has developed a chip location verification technology aimed at preventing its AI chips from being smuggled to restricted countries, responding to U.S. government concerns about chip control [3][10]. Group 1: Technology Overview - The chip location verification technology is currently in the internal demonstration phase and will be offered as a software add-on for customers to install [5]. - This software leverages the confidential computing capabilities of NVIDIA's GPUs and will help customers monitor the overall performance of their chips [5]. - The software will utilize latency data from the chips communicating with NVIDIA servers to determine their location, achieving accuracy comparable to other internet location services [5]. Group 2: Software Features - NVIDIA plans to adopt an open-source model for this software, allowing external security researchers to conduct thorough reviews [7]. - The software will not allow NVIDIA to remotely control or intervene in registered systems, with telemetry data being in "read-only mode" [9]. - There is no "kill switch" feature in NVIDIA GPUs that would enable the company or external parties to disable the chips remotely [9]. Group 3: Regulatory Response - If the technology is officially launched, it will address the demands from the U.S. White House and bipartisan lawmakers to prevent AI chips from being smuggled to certain countries [10]. - The U.S. Department of Justice has previously filed criminal lawsuits against smuggling rings attempting to export over $160 million worth of NVIDIA chips to a specific country, increasing calls for chip control [10]. Group 4: Concerns and Reactions - The push for chip location verification technology has raised concerns among foreign regulators about potential "backdoor" access, allowing the U.S. to bypass chip security measures [11]. - Despite these concerns, NVIDIA has denied the existence of any backdoor and asserted its capability to implement the location verification feature without compromising product security [12].

东软集团表示公司生产经营活动不会因此受到影响。 东软集团在公告中评价其"始终恪尽职守，勤勉尽责，开拓创新，为公司发展倾注了全部心血，并做出了重要 贡献"。 2025 年 12 月 11 日东软集团发布公告称，公司高级副总裁陈宏印因突发疾病于 2025 年 12 月 10 日辞世，享 年 55 岁。 陈宏印个人背景与任职经历 陈宏印出生于 1970 年，拥有大连理工大学应用数学专业本科、计算机工程专业硕士学历。他于 1996 年 7 月 加入东软集团，在公司任职近三十年，期间担任过多个重要职务： PS： 2025 年 6 月，东软集团完成权力交接，70 岁的创始人刘积仁卸任法定代表人、董事长，荣新节接任，但刘 积仁仍以荣誉董事长等身份参与公司战略决策。 1997 年起担任大连分公司副总经理； 1999 年起历任公司电力事业部总经理、能源事业部总经理、能源与民航事业部总经理等职； 2002 年始任公司副总裁； 2023 年 2 月始任公司副 CTO； 2023 年 7 月始任高级副总裁兼首席解决方案官。 东软集团成立于 1991 年，是中国第一家上市的软件公司，也是行业领先的全球化信息技术、产品和解决方案 公司。 ...

以下文章来源于算法爱好者 ，作者小蒜 算法爱好者 . 算法是程序员的内功！「算法爱好者」专注分享算法相关文章、工具资源和算法题，帮程序员修炼内功。 12 月 12 日凌晨，恰逢 OpenAI 成立十周年，新一代大模型 GPT-5.2 突然上线，瞬间引爆行业。 这次更新来得又快又猛——距离上一代 GPT-5.1 发布还不到一个月，背后其实是 OpenAI 的"紧急应战"。 此前谷歌推出 Gemini 3 模型，给 OpenAI 带来不小压力，CEO 奥特曼甚至启动了"红色警报"，要求团队全力提速迭代。 这次 GPT-5.2 算是正式反击，性能直接拉满：在涵盖 44 个职业的任务测试里，70.9%的表现超过或持平人类专家，完成速度还比专家快 11 倍，成本却 不到 1%。 奥特曼说，预计明年 1 月就能解除"红色警报"，下周还会有"圣诞小礼物"。看来 AI 行业的神仙打架，还远没结束。 （参考：腾讯新闻、智东西、每经，本文经由 AI 大模型优化） 不管是写代码、做表格，还是分析百页长文档、解读图表，都比上一代靠谱得多，幻觉率也降了 30%。 这次推出的 3 个版本特别接地气： Instant 版适合日常查资料、写 ...