21世纪经济报道记者闫硕 这个春节，元宝、千问、豆包掀起AI红包大战，主打C端健康场景的蚂蚁阿福也迅速入局。另一方面， 短视频平台的爆款内容让部分AI产品成功出圈，大模型这股风正加速吹向下沉市场。 在皖北一座小县城，用大模型当"健康顾问"已并不鲜见。49岁的秀梅前段时间体检时查出甲状腺结节， 跑了好几家医院，医生都建议静养观察，但她心里始终不踏实。侄子给她推荐了一款垂类大模型，自 此，她便频繁咨询，看着"不必过于担心"的回复，这款产品成了她康复路上的一颗"定心丸"。 后来，侄子推荐给她一款大模型APP，"我最开始问的是'良性甲状腺结节会不会转成恶性的'，输入问题 后几秒便有了回复。"秀梅说，看着"恶变概率非常低，通常在5%以下，很多研究甚至认为低于1%"的回 复，而且其中还附了多份权威资料链接，她松了一口气，反复读了好多遍。此后她又追问康复期间的各 类注意事项，AI的答复总能接住她的细碎担忧。 用户对大模型的应用，早已不局限于健康咨询场景。记者通过走访线下药店了解到，有不少人已经开始 拿着大模型推荐的药品进行咨询、购买。"不少人过年期间存有忌讳就医的习俗，即便出现小病小痛也 不愿前往医院，大多会选择自行购药缓 ...

Core Viewpoint - The discussion highlights the ongoing debate in the AI medical field, emphasizing that while AI has potential, it is not yet fully reliable, and the importance of continuing to train and value human doctors remains critical [1][2]. Group 1: Perspectives on AI in Healthcare - Huang Wei acknowledges that both Zhang Wenhong and Wang Xiaochuan present valid points regarding the role of AI in medicine, with Zhang focusing on the current limitations of AI and the need for ongoing doctor training [2]. - The core argument from Zhang Wenhong is that AI should not be fully relied upon at this stage, and there should be a pathway for doctors to grow and evolve in their roles [2]. - Wang Xiaochuan's perspective emphasizes patient autonomy and the ideal scenario where AI can assist without compromising the growth of medical professionals [2]. Group 2: AI Reliability and Patient Interaction - Huang Wei points out that many patients may use AI tools like DeepSeek to self-diagnose, but they often overlook the high error rates associated with such AI models, which do not take responsibility for their conclusions [2]. - The discussion illustrates the tension between the ideal use of AI in healthcare and the current reality where AI cannot replace the need for human doctors [2].

Core Insights - ByteDance launched Doubao-Seed-2.0 series on February 14, enhancing its large model capabilities for complex real-world tasks [1] - Doubao-2.0 Pro flagship version achieved top scores in IMO, CMO math competitions, and ICPC programming contests, surpassing Gemini 3 Pro in Putnam benchmark tests [1] - The model has improved knowledge coverage in niche areas and performed well in scientific knowledge tests, ranking comparably with Gemini 3 Pro and GPT 5.2 [1] - Doubao-2.0 has upgraded its multimodal understanding capabilities, excelling in visual reasoning, spatial perception, and long-context understanding tests [1] Application and Features - Doubao-2.0 enhances understanding of time series and motion perception, enabling real-time video analysis and environmental interaction for applications in fitness guidance, fashion advice, and companionship [2] - The model's agent capabilities achieved top scores in instruction following, tool invocation, and Search Agent evaluations, with a record score of 54.2 in HLE-Text [4] - Doubao-2.0 Pro is available on Doubao App, PC client, and web version, with a cost-effective pricing model based on input length [4] - The pricing for Doubao-2.0 Pro is set at 3.2 yuan per million tokens for inputs under 32k and 16 yuan per million tokens for outputs, offering a significant cost advantage over Gemini 3 Pro and GPT 5.2 [4] - Doubao-2.0 Lite offers high cost-performance, with input pricing at only 0.6 yuan per million tokens, surpassing the previous generation model Doubao-1.8 [4]

配音：王帅龙 新华社音视频部制作 2月12日，中国互联网巨头字节跳动发布了人工智能（AI）视频生成模型Seedance 2.0。该模型完成度获 得高度评价，包括更精准的构图理解和运用多角度镜头切换呈现电影级视觉效果，已给包括马斯克在内 的众多名人留下了深刻印象并在中国爆红。 编导：胡心懿 0:00 《参考消息》2月14日综合《日本经济新闻》、路透社、西班牙《机密报》网站报道《中国AI视频模型 再掀热潮》。 ...

Core Insights - ByteDance has launched the Doubao-Seed-2.0 series, optimizing for large-scale production environments to handle complex real-world tasks [1] - The Doubao-2.0 Pro flagship version has achieved top-tier performance in various competitions, surpassing Gemini 3 Pro in mathematical reasoning and problem-solving capabilities [1] - The model has enhanced knowledge coverage in niche areas and performed well in scientific knowledge tests, ranking alongside Gemini 3 Pro and GPT 5.2 [1] - Doubao-2.0 has upgraded its multimodal understanding capabilities, achieving industry-leading results in visual reasoning and complex document comprehension [1] Performance Enhancements - The model has improved its understanding of time series and motion perception, enabling real-time video analysis and environmental interaction for applications in fitness guidance and companionship [2] - Doubao-2.0 Pro has demonstrated top-level performance in instruction following and tool usage, achieving a score of 54.2 in the HLE-Text evaluation, significantly outperforming other models [4] Market Positioning - Doubao-2.0 Pro is available on the Doubao App, desktop client, and web version, with a pricing model based on input length, offering a cost advantage over Gemini 3 Pro and GPT 5.2 [4] - The Doubao-2.0 Lite version provides high cost-performance, with input pricing at 0.6 yuan per million tokens, outperforming the previous generation model Doubao-1.8 [4]

中国的基础来构建产业体系，而美国在政策层面和企业回流（reshoring，制造业回归国内） 方面行动迟缓也对此产生了影响。 中山淳史：特朗普政府在对委内瑞拉的军事行动中据称使用了美国Palantir公司的最先进AI技 术。原本属于军民两用的这类技术或许更该应用于企业。中国的产业主导地位不断增强，而 美国制造业衰退显著。而这类技术或许可以成为美国重振制造业的重要工具…… 中山淳史 ：美国调查公司欧亚集团（Eurasia Group）将"'电气国家'中国"列为今年十大风 险中的第二位。英文版的表述为"Overpowered（压倒性的力量）"。 高度依赖美国的日本企业自然会受到影响。美国越是远离EV和可再生能源，日本厂商就越难 以在"21世纪型"产业领域追求规模经济。 美国掳走委内瑞拉总统马杜罗的事件，暴露出美国觊觎石油利益的野心。这也有可能进一步 增加了日本企业未来方向的不确定性。这场堪比电影情节的军事行动使用了最先进技术，同 时也导致地缘政治动荡加剧。 据称，此次行动使用了被称为"动态本体（Dynamic Ontology）"的人工智能（AI）技术。美 军根据大量信息、卫星数据，实时掌握了马杜罗及其周围人物的所 ...

Core Insights - ByteDance officially launched Doubao Model 2.0, which includes Pro, Lite, Mini, and Code models, designed to adapt flexibly to various business scenarios [1][6][4] Model Specifications - Doubao 2.0 Pro targets deep reasoning and long-chain task execution, directly competing with GPT 5.2 and Gemini 3 Pro [1][6] - Doubao 2.0 Lite balances performance and cost, surpassing the capabilities of the previous main model, Doubao 1.8 [1][6] - Doubao 2.0 Mini is optimized for low latency, high concurrency, and cost-sensitive scenarios [1][6] - The Code version (Doubao-Seed-2.0-Code) is specifically designed for programming tasks and works best in conjunction with the AI programming product TRAE [1][6] Availability and Integration - Doubao 2.0 Pro is now available on the Doubao App, desktop, and web versions, allowing users to experience it in 'expert' mode [1][6] - Doubao 2.0 Code has been integrated with the AI programming product TRAE, and the Volcano Engine has launched API services for the Doubao 2.0 series aimed at enterprises and developers [1][6]

Core Insights - Doubao 2.0 has officially entered its second phase, focusing on systematic optimizations for large-scale production environments, enhancing capabilities in efficient inference, multimodal understanding, and complex instruction execution to better tackle real-world complex tasks [1] Product Overview - The Doubao 2.0 series includes three general-purpose agent models (Pro, Lite, Mini) and a Code model, designed to flexibly adapt to various business scenarios: - Doubao 2.0 Pro targets deep reasoning and long-chain task execution, fully competing with GPT 5.2 and Gemini 3 Pro [2] - Doubao 2.0 Lite balances performance and cost, with comprehensive capabilities surpassing the previous generation model Doubao 1.8 [2] - Doubao 2.0 Mini is aimed at low latency, high concurrency, and cost-sensitive scenarios [2] - The Code version (Doubao-Seed-2.0-Code) is specifically designed for programming scenarios, showing enhanced performance when used in conjunction with TRAE [2]

Core Insights - The article emphasizes the importance of understanding what will remain unchanged in the next decade, alongside the rapid evolution of AI technologies [1][2]. Group 1: AI Evolution and Business Impact - The pace of AI development is described as surpassing human cognitive limits, with significant advancements observed throughout the year, including DeepSeek and OpenClaw [1]. - Entrepreneurs are advised to focus on securing their lower limits while betting on higher potential outcomes in the AI era [24]. Group 2: Key Discussions from the Podcast - The podcast "善友探索流" features deep conversations with various explorers, aiming to identify the unchanging foundations of business amidst rapid changes [3]. - A notable discussion highlights the distinction between being a "tool" versus delivering "results" in AI applications, stressing that clients prioritize outcomes over the technology used [11][24]. - The concept of products having "life" in the AI era is introduced, suggesting a shift from mechanical approaches to a more holistic view of creation [11][25]. Group 3: Future Business Perspectives - The article encourages a higher-level perspective on business, suggesting that true breakthroughs often come from elevating one's vision and finding a lifelong purpose [12]. - The notion of extreme idealism as a significant business barrier is presented, indicating that a strong vision can provide a competitive edge [15][25].

Core Insights - The U.S. military conducted a large-scale military operation against Venezuela on January 3, forcibly detaining President Maduro and his wife, reportedly utilizing the AI model "Claude" developed by Anthropic [1][4] - The deployment of "Claude" was achieved through a collaboration between Anthropic and Palantir, a big data analytics company commonly used by the U.S. Department of Defense [2][5] - Anthropic's spokesperson stated that any use of "Claude" must comply with their usage policy, which prohibits applications that promote violence or weapon development [2][5] Company Insights - Anthropic is the first AI model developer to be used for classified operations by the U.S. Department of Defense, raising concerns about the implications of AI in military actions [3][6] - The recent military action has prompted discussions within the U.S. government about potentially canceling a $200 million contract with Anthropic due to concerns over the use of "Claude" [2][5] - Anthropic has positioned itself as a safer alternative in the AI industry, emphasizing its commitment to AI safety, which is now challenged by the military's use of its technology [2][5] Industry Insights - The increasing application of AI models within the Pentagon indicates a growing trend towards integrating advanced technologies in military operations, with potential implications for future conflicts [3][6] - The military's use of AI tools spans a wide range of functions, from document summarization to controlling autonomous drones, highlighting the versatility and potential risks associated with AI in defense [3][6] - The international response to the U.S. military action has been critical, with multiple countries condemning the operation and calling for adherence to international law and the principles of the United Nations Charter [3][6]