对上海的澳洲来讲肯定是一个收益水平很高的一个县但是在这条线上价格的下探是旅客的一个需求还有竞争关系双方的一个作用一个必然的情况然后地区方面就是台湾的体能价值减了50%左右而客工里收入提升了20%这个就是正常的供需关系的变化港澳的话其实价值增幅有30%但是跳价水平还是减了6% 这个也可以理解因为其实大家可以逐渐的感受到现在去港澳其实对内地的人来说可能已经逐渐没有以前那么大的吸引力了这个也是客观的一个人文社会这种变化还有一个旅游趋势变化导致的一个结果那么国内的话其实相对来讲 差异是比较大的我觉得同样没有什么不用再多讲了吧差异我这边主要是两点吧第一个就是投入有一个大幅的增长因为这么多年来呢虽然整体民航业经营不是很顺利但是飞机还是在持续的买尤其现在有这个国产的飞919时刻还在不停的增长但是这些增长的时刻呢其实并没有在北昌华盛那样的大机场 然后同时国际恢复疲软终于最后的格局就是国内运力投入大幅增长竞争加剧导致需要进一步拼抢拉锯客源那么另一方面的旅客的消费能力确实是在下降然后高铁竞争加剧因为我们行业内原来说跟高铁的竞争大概在600到800公里但是现在我觉得基本上大家默认已经在200到1200公里因为高铁的速度和他们网络破 ...

Investment Rating - The report raises price targets for Alchip (to NT$3,988), TSMC (to NT$1,388), and MediaTek (to NT$1,688), reflecting greater AI ASIC and advanced packaging revenue assumptions in 2025 and 2026 [9][10]. Core Insights - The AI ASIC total addressable market (TAM) is forecasted to grow from US$12 billion in 2024 to US$30 billion in 2027, driven by strong competition from NVIDIA's AI GPU [2]. - Alchip is identified as the new Top Pick due to its significant price upside (45%) and higher exposure to AI (75% of revenue) compared to TSMC (30% price upside) [6][10]. - TSMC is expected to allocate 27% of CoWoS capacity for ASIC in 2025, with total AI semiconductors accounting for 30% of revenue in 2026 [6]. Summary by Sections ASIC Projects - AWS 3nm Trainium project is expected to significantly benefit Alchip, with a revenue forecast increase of 23% for 2026 due to the 3nm project upside [2][16]. - Google TPU v7 project is anticipated to contribute approximately US$2.1 billion in revenue for MediaTek in 2026 [3][30]. - Microsoft’s Maia200 project is expected to favor Marvell over GUC, leading to a price target trim for GUC [4][62]. Revenue Forecasts - Alchip's total revenue is projected to reach US$3,075 million in 2026, reflecting a 73% year-over-year growth [21]. - MediaTek's ASIC revenue forecast for 2026 has been raised from US$2 billion to US$2.5 billion [37]. - TSMC's AI revenue is expected to account for approximately 34% of its total revenue by 2027 [56]. Competitive Landscape - The competition between US and Asian ASIC service providers is intensifying, with Alchip and MediaTek emerging as key players [27][29]. - TSMC is positioned to benefit regardless of whether ASIC or GPU solutions dominate the market, with expectations of significant revenue contributions from both [45][50]. Market Trends - The report anticipates a 43% compound annual growth rate (CAGR) for the cloud AI semiconductor market from 2023 to 2030, potentially reaching US$500 billion by 2030 [59]. - The demand for advanced packaging solutions like CoWoS is expected to double in 2025, indicating strong growth in AI semiconductor production [46][54].

各位投资者朋友大家下午好我是平安证券护手团队的分析师盛德晨很高兴今天给大家带来我们2025年到债市的年度策略那我们的题目叫不预设顺势而为 主要基于的背景就是明年无论是内部还是外部面临的不确定性都比较大那更多其实任何的预设都有可能会被打破那更多还是跟随着整体的利率的一个趋势以及寻找的一些路标去做交易 那明年主要是面临两个大的不确定一个就是内部一个就是外部那内部方面像我们现在的政策基调也调整成了适度宽松的货币政策更加积极的货币政策财政政策所以明年稳增长是一个明年的内部的最大的一个主题那外部层面 特朗普上台之后特朗普2.0对美国的无论是对美国的经济还是货币政策以及传导到人民币汇率传导到咱们国内的基本面都可能会造成不小的一个影响所以在这样一个背景之下我们这篇年度策略报告 四个部分那第一个部分呢就是复盘了一下24年整体的一个表现大类资产海内海外那第二部分就是对于国内的一个小专题就是在这种稳当涨发力期市场应该如何走出方向我们也寻找了一些路标第三个部分就是海外的一个小专题 就是我们探寻了特朗普2.0版本对美国货币政策以及对国内基本面的影响也做了一些测算那最后一个部分就是落脚到投资策略那我就跟大家开始汇报首先第一部分复盘一下 ...

各位投资者朋友大家好我是平安证券的雇收组分析师陈维宁我们今天很高兴在这里给大家分享我们25年的RAISE年度策略我们的策略项目叫布局之年这个其实也是我们对于明年RAISE市场的一个看法 那今天我们的分享会分成三个部分第一个部分先给大家展示一下现在瑞士市场是一个什么样的状态第二部分带大家回顾一下24年瑞士的行情表现第三部分再进入到策略展望的部分首先截至到12月3号瑞士市场的市值规模已经接近1400亿元了流通市值是接近一半的一个规模 那今年其实也是一个RAIDS的一个供给大年截止到12月3号已经募集的这一些规模是达到了588亿如果再加上一些可能在12月继续募集的一些个券的话那今年的供给规模很有可能会突破600亿元那和往年的每年新增供给300到400亿元的这个规模相比呢今年是明显的供给的量比较高的一年 那在存量的项目当中呢占比比较高的是交通产业和消费三者合计占比超过了65%嗯那大家也都知道RACE是一个高分红的资产那它具体分红有多高呢我们给大家一个概念大概就是产权类的RACE的现金分派率是5.02%那是比较接近红利股的股息率略低略微低一点点 比长久期的信用债则是明显高出162个BP另外RAISE的一个主要的价值就是 ...

影视作品我们那些大几亿投入的然后大卡斯大IP大制作比如说像繁华信念啊还有最近上的爱奇艺的我是醒醒这些的剧集都还是能够吸引到用户并且用户的付费意愿会比较强 那另外一方面在亚宁的另一头是我们看到可能以短剧和小游戏为代表的一些进入门槛比较低,相对玩起来比较轻度,但是又有一定课金付费的一个重度挖掘能力的一些消费内容。 得到了用户碎片化时间为首的一些倾向吧所以其实更多的用户线上娱乐需求是向头部集中和尾部下沉的一个分布状态这是我们观察到的第二个趋势那公司在这两侧的布局也会显得更加的关键和重要然后第三个趋势我们在持续关注的是AI技术的一个助力效率提升 最近其实A股的一个活跃很多也是AI应用的大家的一个期待吧那往后看的话虽然现在AI在2B端的一个适合游戏内容的生产场景中应用比较多但相对成熟的2C端的应用还没有太多普及我们看到了一些端倪比如说最近比较火的AI陪伴玩具还有可能就是AI陪伴式对话等等包括AI Agent 那这些可能会在明年会有更多端倪,甚至会帮助我们的公司有更多商业模式的一个衍生。然后最后的话也是最近较火的股市经济,如果往回退一步来讲就是我们一直在讲的IP产业链的一个发展,可能能看到更多的一个成熟的迹象。 今年的股 ...

Summary of Key Points from the Conference Call Industry Overview - The conference discusses the advancements and future potential of humanoid robots driven by AI technology, particularly in areas such as multi-modal perception, large language models (LLM), and precise motion control [1][2][3]. Core Insights and Arguments - **AI Breakthroughs**: Recent advancements in AI have significantly enhanced humanoid robots' capabilities in environmental perception, cognitive decision-making, and task execution, expanding their application scenarios [1]. - **Current AI Limitations**: Existing AI technologies lack a connection to world models, leading to insufficient understanding of physical laws and limitations in human output capabilities [4]. - **Humanoid Robots as Future Direction**: The design of humanoid robots aligns with human needs, increasing their acceptance in the market and showcasing their potential for versatility across various jobs and complex environments [5]. - **AI and Physical Robots Integration**: The combination of AI with physical robots allows for improved data collection and model training, enhancing the understanding of physical laws and human emotions, which is crucial for consumer market applications [6]. - **Commercialization Bottlenecks**: Key challenges include hardware discrepancies, insufficient AI capabilities for complex tasks, and the need for greater market acceptance [8]. - **Application Prospects**: Humanoid robots are expected to see widespread use in elder care within five years and gradually replace human labor in industrial production [9][11]. - **China's Advantages**: China benefits from strong policy support, robust hardware capabilities, and significant industrial cluster effects, positioning it to potentially lead the global humanoid robot market [25]. Additional Important Content - **Historical Context**: Humanoid robots have a long history, with early designs dating back to ancient China and Leonardo da Vinci's mechanical knight [7]. - **Industrial Applications**: Humanoid robots are currently being tested in controlled environments like factories, with companies like Tesla and XPeng exploring their use for tasks such as material handling [13][14]. - **Cost Control and Domestic Alternatives**: The high cost of humanoid robots is attributed to evolving hardware configurations and the lack of mass production. However, there is significant potential for cost reduction through domestic manufacturing [18][20]. - **Market Acceptance Challenges**: The introduction of humanoid robots into family care settings faces both technical and emotional barriers, requiring time for consumer adaptation [12]. - **Future Market Expectations**: The anticipated price for humanoid robots could reach levels comparable to large consumer electronics, driven by advancements in component manufacturing and cost reduction strategies [20]. Conclusion - The conference highlights the transformative potential of humanoid robots in various sectors, particularly in elder care and industrial applications, while also addressing the challenges that need to be overcome for widespread adoption. The strong support from the Chinese government and advancements in technology position the country favorably in the global humanoid robot market.

传媒周一谈：AI、IP、首发经济投资机会再梳理 20241216 摘要 • 传媒互联网板块投资机会主要集中在 AI 应用（AI 广告、社交陪伴、教育、 影视及 AI 玩具）、IT 产业链和电影及内容出海等赛道，这些领域在 2025 年将持续保持高景气。 • 尽管近期传媒板块有所上涨，但长期来看，建议投资者关注高景气赛道， 尤其是在大产业逻辑下不断验证和催化的领域，短期调整可能是再次布局 良机。 • 首发经济为传媒互联网行业带来新的发展机遇，与 AI 应用、IP 及内容出 海等方向相契合，将成为推动相关行业增长的重要动力，尤其在演艺、会 展、游戏、影视等领域。 • 大模型时代 AI 应用持续看好，多模态（尤其视频生成）、AA 型技术突破 及 AI 硬件结合（智能眼镜、AI 玩具）等近期重要突破将推动用户增长和 商业化进程。OpenAI、字节跳动等公司在视频生成领域进展显著。 • 国内外 AI 助手领域竞争激烈，质朴 AI、OpenAI、谷歌等公司均在提升 AI 助手能力，以满足用户学习、工作、生活及情绪需求，提升效率和便利性。 • 重点布局方向包括 AI 视频生成（上海电影、芒果超媒等）、AI 社交陪伴 及玩具 ...

超常规？债先牛？A 股怎么走 20241215 摘要 • 当前市场震荡，主要源于基本面和资金面因素的博弈。宽松的货币财政政 策利好风险资产，但企业利润下滑和居民信心不足限制了股市上涨。 • 资金流向呈现分化：北向资金和 ETF 流入增加，但国债价值上升和汇率贬 值显示中性偏空。融资余额上升但活跃度下降，表明部分投资者被套。 • 建议中等仓位参与区间震荡，优先配置大盘股，关注人工智能和并购重组 等题材性机会及相关 ETF，同时关注近期交易火热的债券型 ETF。 • 债券市场牛市，因国债发行利率下降导致存量债券价格上涨，部分资金流 向债市，短期内可能对 A 股造成压力。国债 ETF 三年期年化收益率已达两 位数。 • 中国经济受投资、出口和消费三大因素影响，目前投资低迷，出口受外部 环境影响，消费受政策刺激但市场反应波动。中央经济工作会议提出扩大 内需、科技创新等政策。 • 房地产市场下行，未来或将通过发行更多国债拉动经济增长。扩大内需措 施包括发放消费券、家电下乡等，以刺激消费增长。 • 股市未来走势预计慢涨慢跌，需营造稳健投资环境，提升市场活跃度。人 工智能有望成为新的经济突破口，相关领域 ETF 值得关注。 ...

字节引领应用机会，AI 板块投资机会梳理 20241215 摘要 • 字节跳动在 AI 领域快速布局，C 端应用如"豆包"（DAU 达 1,000 万）和 "集梦"（DAU 达 20 万）增长迅速，B 端合作也积极推进，火山引擎大会 将重点展示大模型应用及合作计划。 • 字节跳动在 AI 玩具领域积极储备，测试产品交互体验良好，预计明年将 是重要发展节点，下周五将在上海举办 AI 玩具沙龙。 • 字节跳动的大模型布局加速了国内 AI 发展，带来算力、玩具营销等新投 资机会，相关公司如掌阅科技、海天瑞声、蓝色光标等将受益。 • 智能眼镜市场迎来爆发，预计明年出货量将达数百万甚至千万级别，产业 链公司如歌尔股份、深圳易到信息、恒玄等将迎来增长机遇。 • 智能眼镜成为 AI 硬件落地重要场景，因其便捷的交互方式、视觉感知匹 配、舒适的佩戴体验和强大的拓展性。 • 智能眼镜产业链关键环节值得关注，包括与国际头部芯片厂商合作的公司、 上游 SOC 环节公司、核心元件领域公司等。 • 传统 3C 产业链未来一年展望乐观，短期内受益于政策补贴，长期内受益 于 AI 手机布局，推荐歌尔股份、东山精密、蓝思科技等公司。 Q&A ...

Summary of AI Companion Products Conference Call Industry Overview - The conference focused on the AI companion products industry, highlighting both domestic and international market performances. Notable overseas products include Carter AI (with over 10 million DAU) and Chao Ti (approximately 3 million DAU), while domestic products like Xin Yi Ming Tian and Mao Xiang (around 1 million DAU) dominate the role-playing category. Overall, user willingness to pay remains low, and the profitability model is still immature [1][2]. Core Insights and Arguments - **Business Models**: The primary business models for AI companion products include charging for conversation counts, premium version subscriptions, and creator fees. Currently, only Tokyo Xin Ye has achieved relatively stable profitability through multimodal capabilities and advertising revenue, while others struggle with low user payment willingness [1][3][4]. - **User Demand**: Demand for AI companion products stems from two main categories: sexual attraction or hormonal guidance, and emotional companionship or venting. Target users include those with social anxiety, individuals lacking a listening ear, and those with psychological trauma [1][5]. - **Market Development**: The future market for AI companion products is expected to evolve towards a combination of hardware and software, with hardware's importance in consumer applications increasing. However, current models and hardware technologies require further development to achieve significant breakthroughs [1][6][7]. - **Unique Value Proposition**: Even with advancements in AI assistant capabilities, AI companion products will continue to provide unique emotional support and interaction experiences for specific user groups, ensuring they are not completely replaced [1][8]. Additional Important Content - **Regulatory Challenges**: AI companion products face stringent content regulation challenges, particularly regarding the protection of minors. Companies must establish comprehensive content review mechanisms to balance user engagement and safety [1][11][13]. - **R&D Investment**: Key investment points in R&D include training model effectiveness and improving dialogue, image, and sound solutions. The costs vary significantly depending on whether companies use open-source models or develop their own [1][10]. - **Ethical and Safety Audits**: There are significant differences in ethical reviews between domestic and international markets. Domestic regulations are stricter, particularly concerning political content and explicit topics, while international markets may have more lenient standards [1][15][16]. - **Market Dynamics**: Many large model companies focus on data accumulation and user scale rather than immediate profitability, which poses sustainability challenges as the industry is still in the market share acquisition phase without a stable profit model [1][17][18].