最危险的盲区不在于"犯错"，而在于无法识别自己正在犯错。 以下是他的故事： 老于的儿子2025年6月参加高考，考试之前要体检，报告显示合格，有几项指标略显异常，他一时手欠，把报告发给了DeepSeek，DeepSeek R1在2025年 初爆火，性能比肩OpenAI o1正式版的同时，实现了超低训练成本，并且全面开源，给全球AI界带来了一场"地震"，老于对它高度信任。DeepSeek倒也尽 职，每一项分析得都很详细，只是有个用黑体标注的部分让老于倒吸一口冷气—— 虽然体检表格里没有直接写"乙肝表面抗原阳性"，但ALT（丙氨酸氨基转移酶）如此大幅度升高，强烈提示考生存在肝脏疾病（很可能是乙型肝炎或其他 肝炎）。根据规定，体检医院有责任提示该考生属于"二-5"条款限制的范围。因此，他不能报考学前教育、航海技术、飞行技术、食品类、烹饪类等相关 专业。 高考前查出有肝炎，如果真的有问题，受限专业恰好都是孩子准备报考的，老于这可慌了。他连夜发动所有关系，请专家看体检报告，同时又从教育部官 网下载《普通高等学校招生体检工作指导意见》，发现"二-5"条款和DeepSeek说的完全不一样。他上传了原文，并指出了大模型的错误。 ...

炒股就看金麒麟分析师研报，权威，专业，及时，全面，助您挖掘潜力主题机会！ 春节大战，全员在岗。 作者/冯雨晨 报道/投资界PEdaily 一批人开始复工，一批人则陆续放假。 回到春节前夕，大厂同事圈中一则帖子流传：某宝作为重点项目团队全员留守，原则上不批假，工资按 照工资三倍支付。 一位参与加班的腾讯元宝人士告诉投资界：春节加班是要支持元宝的春节活动，另一方面，则是等待 Deepseek V4模型的更新。 没有人敢掉以轻心。时间回到1月底，腾讯掌门马化腾在公司年会上宣布，旗下AI应用腾讯元宝将启动 春节分10亿元现金活动。"希望重现当年微信红包摇一摇绑定数亿用户的盛况"，马化腾当时如此定调。 外界可能不知道，去年DeepSeek爆火之际，元宝宣布接入满血版DeepSeek R1，从原本只依赖自研模型 转向支持多模型的产品策略。这一策略延续至今天，而在春节前夕，市场频频传出DeepSeek要在春节 期间更新上线V4版本。 如此一来，超级流量与氛围狂欢中，叠加可能到来的DeepSeek更新，腾讯无疑想把技术保障做好。于 是，打工人们来到加班"作战"状态。 同时，字节有些外地团队奔赴北京留守，将在春节期间支持豆包。 ...

我来为您详细解读这份腾讯研究院发布的《塑造自己的下一个版本：2026前沿科技趋势》报告。这份报告以"用户视 角"出发，眺望2030年的自己，围绕五个维度展开前沿科技趋势分析。 --- 一、生命力2030：从"活得久"到"活得好" 核心观点：人类生命正经历"第三次转型" 报告开篇指出一个关键转折：过去一百年人类寿命翻了一倍，但从1900年到2000年的快速增长后，预期寿命增速已大 幅放缓。2024年《自然·衰老》期刊的研究表明，通过消除早夭和中年疾病来延长寿命的"容易摘的果实"已被采摘殆 尽。 - Alnylam Pharmaceuticals开发的RNA干扰技术仅需每六个月一次皮下注射，即可控制高血压 - 斯坦福大学开发的mRNA CAR-T技术在小鼠淋巴瘤模型中实现75%的长期无瘤生存 新范式诞生：全球正在从追求单纯的"寿命"（Lifespan）转向追求"健康寿命"（Healthspan）——即在没有严重慢性病、 残疾或认知功能衰退的情况下维持良好生活质量的年限。据世界经济论坛报告，若将人类健康寿命延长1年，产生的全 球经济价值将高达38万亿美元。 三大技术支柱 基因疗法进入"生命代码优化"时代 - CRI ...

不知道大家有没有这种感觉：和AI机器人短时间聊天的话还行，时间一长，就感觉对话开始变的前言不搭后语、逻辑不通。 其实这种感觉并不是错觉。 研究人员对包括 GPT-4.1、Gemini 2.5 Pro、Claude 3.7 Sonnet、o3、DeepSeek R1 和 Llama 4 在内的 15 款顶尖模型进行了超过 20 万次模拟对话 分析，揭示出一个被称为"迷失会话"的系统性缺陷。 数据显示，这些模型在单次提示任务中的成功率可达 90%，但当同样的任务被拆解成多轮自然对话后，成功率骤降至约 65%。 研究指出，模型的核心能力仅降低约 15%，但"不可靠性"却飙升 112%。 最近，微软发表的一项研究证实，即使是目前最先进的大语言模型，在多轮对话中的可靠性也会急剧下降。 研究人员指出，现有的基准测试主要基于理想的单轮场景，忽略了模型在真实世界中的行为。 因此，对于那些依赖 AI 构建复杂对话流程或智能体的开发者而言，这一结论意味着未来将要接受严峻挑战。 再来看看其他消息。 也就是说，AI 大模型仍然具备解决问题的能力，但在多轮对话中变得高度不稳定，难以持续跟踪上下文。 | Short Form | Nam ...

Core Viewpoint - The article discusses the competitive landscape of AI large models in China, highlighting the emergence of major players and the strategic moves of DeepSeek, led by Liang Wenfeng, amidst intense competition in the AI sector [4][19]. Group 1: Competitive Landscape - Major internet giants are aggressively competing to establish their AI large models as the primary traffic entry point, with significant cash incentives being offered to users [7][22]. - Companies like Tencent, Baidu, and Alibaba are investing heavily in user acquisition through cash giveaways, indicating a fierce battle for market share in AI applications [7][22]. - The release of new models by ByteDance and Alibaba demonstrates a coordinated competitive response, while DeepSeek appears to be taking a more subdued approach [9][10]. Group 2: DeepSeek's Position - DeepSeek, founded by Liang Wenfeng, gained recognition for its cost-effective AI model R1, which competes with top global models at a fraction of the cost [4][17]. - Despite speculation about the release of a new flagship model (V4), DeepSeek has maintained silence, opting for a quiet update that significantly increased its context window from 128K to 1M tokens [10][11]. - The company continues to recruit talent, indicating ongoing development and a commitment to innovation in AI technology [11][20]. Group 3: User Engagement and Market Strategy - DeepSeek is shifting focus towards understanding and addressing user needs, as evidenced by its recruitment for product management roles aimed at enhancing user experience [20][21]. - The competitive landscape is shifting towards meeting real user demands, with companies that can effectively solve user problems poised to dominate the AI market [24]. - The article emphasizes that the next decade of internet order will be defined by which companies can successfully engage users and leverage AI capabilities [24].

Core Insights - The article discusses the competitive landscape of AI large models in China, highlighting the ambitions of major internet companies to dominate this space and the notable presence of DeepSeek, led by Liang Wenfeng, who previously made a significant impact with the release of their R1 model [2][4][12]. Group 1: Company Developments - DeepSeek, founded by Liang Wenfeng, gained recognition for its R1 model, which achieved top-tier performance at a fraction of the cost compared to competitors [12]. - Despite the competitive environment, DeepSeek has remained relatively quiet, with speculation about the release of their new model, V4, which is aimed at coding AI [6][12]. - On February 11, DeepSeek updated its model's context window from 128K tokens to 1M tokens, indicating ongoing development [6]. Group 2: Competitive Landscape - Major players like Tencent, Baidu, and Alibaba are aggressively promoting their AI products with substantial cash incentives, indicating a fierce competition for user engagement [4][15]. - ByteDance's new model, Doubao 2.0, and Alibaba's Qwen-Image 2.0 were launched around the same time, showcasing the rapid advancements in AI model capabilities [5][14]. - The competition is shifting towards understanding and addressing user needs, with companies focusing on enhancing user experience and engagement [14][17]. Group 3: Market Trends - The article suggests that the demand for AI applications in consumer markets is on the rise, with companies needing to address real user problems to establish themselves as key players in the AI era [16][17]. - The strategies employed by major companies, such as cash giveaways and user engagement initiatives, reflect a broader trend of cultivating user familiarity with AI technologies [15].

Core Insights - The article discusses the intense competition among internet giants in the AI large model sector, highlighting the ambitions of companies to establish their AI applications as the primary traffic entry point [4][23] - DeepSeek, founded by Liang Wenfeng, emerged as a significant player in the AI landscape with its R1 model, which was launched at a surprisingly low cost, challenging the perception of high investment requirements for top-tier models [14][31] - Despite the competitive environment, DeepSeek has maintained a low profile, with recent updates suggesting a potential new model release, V4, but with no official confirmation [26][27] Industry Competition - Major companies are aggressively distributing cash incentives to attract users, with Tencent offering 1 billion yuan, Baidu 500 million yuan, and Alibaba 3 billion yuan, indicating a fierce battle for user engagement [25] - The launch of new models by ByteDance and Alibaba, including the 2.0 versions of their respective models, reflects a rapid evolution in AI capabilities and competition [8][25] - The article notes a peculiar competitive dynamic where companies are responding to each other's moves, creating a sense of mutual awareness in the market [8][25] DeepSeek's Position - DeepSeek's recent updates include an increase in context window length from 128K tokens to 1 million tokens, suggesting advancements in their technology [26] - The company continues to recruit talent despite a slowdown in hiring across the industry, indicating its commitment to innovation and development [27] - Liang Wenfeng's vision for DeepSeek is to lead in AI research and development, aiming to create a general-purpose AI that goes beyond existing models [31][32] User Engagement and Market Dynamics - The article emphasizes the importance of addressing user needs in the AI sector, with companies like DeepSeek beginning to focus on consumer-facing products [33] - The competition is framed as a quest to meet real user demands, which will determine the leading players in the AI landscape [36] - The article concludes that the current battle among internet giants is crucial for defining the next decade of internet order, highlighting the strategic significance of user engagement in AI applications [36]

Core Viewpoint - The article discusses the intense competition among major internet companies in China to dominate the AI model application space, highlighting the strategic positioning of Deep Seek and its founder Liang Wenfeng as a significant player in this evolving landscape [2][4]. Group 1: AI Competition Landscape - Major internet giants are aggressively investing in user incentives, with Tencent distributing 1 billion yuan in cash red envelopes, Baidu offering 500 million yuan for promoting its Wenxin assistant, and Alibaba launching a 3 billion yuan campaign [4]. - The competition is characterized by rapid product releases, with ByteDance announcing its Doubao model 2.0 and Alibaba introducing its Qwen-Image 2.0 model, indicating a synchronized response among competitors [5][6]. Group 2: Deep Seek's Positioning - Deep Seek, founded by Liang Wenfeng, has maintained a low profile despite its significant achievements, including the release of the R1 model in early 2025, which matched top global models at a fraction of the cost [2][9]. - The company is rumored to be preparing to launch its next-generation model, V4, aimed at coding AI, but has remained silent on the exact timeline [6][10]. - Deep Seek's recent updates have increased its context window from 128K tokens to 1 million tokens, suggesting ongoing advancements in its technology [6]. Group 3: Liang Wenfeng's Background - Liang Wenfeng, born in 1985 in Guangdong, has a strong academic background in computer science and has been involved in AI and quantitative trading since his university days [7][8]. - He co-founded Hangzhou Huafang Technology, which became a significant player in quantitative trading, and later established Deep Seek to pursue general artificial intelligence [9]. Group 4: User-Centric Approach - Deep Seek is shifting its focus towards user experience and product innovation, as evidenced by its recent job postings aimed at enhancing C-end product functionality [10][11]. - The article emphasizes the importance of addressing real user needs in the AI sector, suggesting that the ability to solve genuine problems will determine the success of AI applications [11].

Group 1 - ByteDance is set to officially launch the Doubao Model 2.0, marking a significant development in China's AI large model sector, which has seen a surge in new product releases from various tech companies since the beginning of the year [1] - The Doubao Model 2.0 features enhanced multimodal understanding capabilities, excelling in areas such as multimodal perception, chart comprehension, and long video understanding [1] - Other companies have also introduced advanced models, including Zhipu's GLM-5 model focusing on complex tasks and video generation, and Kuaishou's 3.0 series models that cover image and video generation, smart editing, and post-processing [1] Group 2 - The development of large models is driven by market demand and the improvement of the industry ecosystem, with the user base for generative AI in China reaching 602 million by December 2025, indicating a growing market space for technology implementation [5] - In 2025, over 80% of the 205 new AI applications launched in the second half of the year are concentrated in specific scenarios such as image processing, office tasks, and education [3] - The industry is becoming increasingly regulated, with an average of more than one generative AI service being registered daily with the National Cyberspace Administration, totaling 748 services registered, which lays a solid foundation for the widespread adoption of large model technology [5]

Core Insights - The article discusses the significant impact of Feng Ji, the founder of Game Science and producer of "Black Myth: Wukong," on the narrative of China's AI development through his social media posts [4][34] - Feng Ji's evaluations of DeepSeek R1 and ByteDance's Seedance 2.0 mark pivotal moments in the evolution of China's AI landscape, highlighting the transition from mere technological competition to a more mature stage of innovation and application [4][34] Group 1: Key Events and Evaluations - On January 26, 2025, Feng Ji praised DeepSeek R1 as a "national fortune-level technological achievement," emphasizing its accessibility and affordability [5][6] - He articulated the value of DeepSeek across six dimensions: powerful, cheap, open-source, free, connected, and local [5] - On February 9, 2026, he described Seedance 2.0 as "the strongest on Earth," declaring the end of the "childhood era" of AIGC, indicating a shift towards commercial viability and productivity [13][34] Group 2: Impact on AI Perception - Feng Ji's commentary helped demystify AI technologies, making them relatable to the general public by framing them as tools for everyday use [8][36] - His insights coincided with critical moments in the AI sector, where confidence in Chinese models was essential for their acceptance and integration into daily life [9][34] Group 3: Technological Advancements - DeepSeek's introduction challenged the notion that developing powerful models requires exorbitant resources, demonstrating that algorithmic innovation can significantly reduce costs [16][17] - Seedance 2.0 represents a breakthrough in video generation, showcasing China's capability to compete in this challenging domain [18][19] Group 4: Market Dynamics and Global Competition - The article highlights a shift in the AI landscape, where smaller companies can now compete alongside industry giants due to reduced barriers to entry and innovative algorithms [32][33] - The global attention on Chinese AI advancements, as evidenced by Elon Musk's comments on Seedance 2.0, indicates a growing recognition of China's role in the international tech arena [27][28] Group 5: Cultural and Societal Integration - The 2026 Spring Festival saw a surge in AI tool usage for traditional activities, indicating a cultural shift towards integrating AI into everyday life [22][24] - This trend reflects a broader democratization of technology, where AI tools are accessible to a wider demographic, including older generations in smaller cities [25][24]