2026.02.27 本文字数：3393，阅读时长大约5分钟 作者 |第一财经齐琦 2月27日，A股小金属板块延续强势表现，实现连续三个交易日走高。 期货市场同步狂热，沪锡主力2604合约午后强势崛起，盘中暴涨超8%至45.32万元/吨，最高位触及45.38万元/吨。 然而，在板块大面积上涨背后，各类小金属产业逻辑悄然生变，价格也分化明显。其中，钨价、锡价刷新历史新高，镁金属却仍在低位徘徊，部分企业甚 至出售资产缓解经营压力。本轮小金属行情的驱动力是什么？不同品种之间缘何分化？ 小金属开年大涨50%，多股翻倍 本周三，A股小金属板块行情被点燃，导火索是特朗普政府计划用AI为关键矿产设定参考价格。该AI定价模型初期聚焦锗、镓、锑和钨，未来将扩展至其 他矿物。 除了消息层面的催化，小金属供给端约束与需求端结构性爆发，成为板块持续升温的关键。 据Choice数据统计，截至2月27日，申万小金属行业年内累计涨幅近50%，在124个申万二级行业中暂列首位。 成份股中，25只个股涨势强劲，年内股价涨幅中位数达42.42%。其中，有12只股票年内累计涨幅已超50%，其中翔鹭钨业、章源钨业和中钨高新股价已经 翻倍，年内累计涨 ...

| | | 分时 多日 1分 5分 15分 30分 * | | | | F9 盘前盘后 盛加 九转 画线 工具 | | (2) (2) | 有色ETF华宝 | | 159876 | | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | | | | | | | 159876(有色ETF华富) 09:48 价 1.230 浴膜 0.026(2.16%) 均价 1.217 成效量 5537 | | | 2.199 | DEG | | +0.026 +2.169 | | | | | | | | | | | SZSE CNY 9:48:42 文星中 | | 原ノ ■ + | | | MJ 277 | | | | | | | 1.4699 | 净值出处 | | 华宝中证有色会属ETF | | × | | | | | | | | | 今年 | 23,49% 120日 | 67.12% | | | | | | | | | | 0.73% | ਟੇਜ | 5.13% 250日 | 125,72% | | R | | | | | | | ...

其他蓝筹股方面，汇丰控股(00005)涨5.47%，报142.7港元，贡献恒指140.04点；龙湖集团(00960)涨 4.6%，报10.24港元，贡献恒指1.83点；信义光能(00968)跌3.58%，报3.5港元，拖累恒指0.95点；万洲国 际(00288)跌3.27%，报10.05港元，拖累恒指4.61点。 港股三大指数今日走势分化，恒指盘中一度涨近1%，恒科指数则冲高回落，尾盘再度翻绿。截至收 盘，恒生指数涨0.66%或175.4点，报26765.72点，全日成交额2367.65亿港元；恒生国企指数涨0.3%，报 9034.75点；恒生科技指数跌0.19%，报5260.50点。 天风证券指出，短期看，港股在前期估值修复后延续高位震荡格局，南向资金持续净流入与结构性活跃 度对市场形成一定支撑，指数仍具阶段性韧性与结构性机会。但外部环境约束仍然存在。中期维度下， 对港股整体判断维持谨慎乐观，结构优于指数的特征或仍将延续。 蓝筹股表现 海底捞(06862)领涨蓝筹。截至收盘，涨6.19%，报17.51港元，成交额5.39亿港元，贡献恒指3.12点。春 节假期（2月15日至23日，共9天），海底捞全国门店累计 ...

Core Viewpoint - Instacart's stock price fell by 6% following reports of an investigation by the Federal Trade Commission (FTC) into its AI pricing tool, Eversight, which has been criticized for significant price discrepancies among similar grocery items [1] Group 1: FTC Investigation - The FTC has issued a civil investigative demand to Instacart, seeking information about its Eversight pricing tool [1] - A study involving 437 shoppers across four cities found an average price difference of about 7% for the same grocery items on Instacart [1] - The FTC expressed concern over media reports regarding Instacart's alleged pricing practices, stating it does not comment on ongoing investigations [1] Group 2: Instacart's Response - Instacart stated that recent reports inaccurately conflate A/B price testing, dynamic pricing, and monitoring pricing, misrepresenting how its pricing tests operate [2] - The company clarified that it does not use personal information or demographic data to set prices on its platform, emphasizing that retail partners control the base prices [2] - Instacart addressed concerns related to Target, explaining that it uses publicly available price information as a starting point and adds a fee to cover operational costs, noting that it has ended certain pricing tests at Target stores [2]

Core Viewpoint - The Federal Trade Commission (FTC) is investigating Instacart due to criticisms regarding its AI-driven pricing tool, Eversight, which has led to significant price discrepancies for different shoppers purchasing the same groceries [1][2]. Group 1: Investigation and Criticism - The FTC has issued a civil investigative demand to Instacart, seeking information about the Eversight pricing tool [1][2]. - A study conducted by Groundwork Collaborative, Consumer Reports, and More Perfect Union revealed that nearly 75% of tested items on Instacart showed different prices for different users [3]. - The price differences for the same basket of goods from the same retailer could reach up to 7%, potentially resulting in an annual cost difference of approximately $1,200 for consumers if this discrepancy persists [3]. Group 2: Company Response - Instacart stated that only a limited number of partner retailers are conducting online pricing tests, which do not utilize personal, demographic, or individual behavioral data [2][3]. - The company emphasized that prices do not change based on supply and demand or real-time factors [2][3].

Core Viewpoint - The release of the "Internet Platform Antitrust Compliance Guidelines (Draft for Comments)" marks a significant step in China's regulatory framework for platform economies, particularly following the "Double Eleven" shopping festival, addressing new competitive issues that emerged during this period [1][10]. Group 1: Regulatory Positioning - The new guidelines and the previously issued "Price Behavior Rules" form a comprehensive regulatory system, with the former focusing on platforms with significant market power while the latter applies to all operators [2][3]. - The "Price Behavior Rules" aim to maintain basic price order and transparency, while the new guidelines specifically target monopolistic risks and behaviors of dominant platforms [3]. Group 2: Key Breakthroughs in the New Guidelines - The new guidelines introduce an "ecological" perspective, emphasizing the responsibility of platform managers in maintaining healthy platform ecosystems [6]. - The guidelines require platforms to conduct self-examinations of their algorithms, addressing risks associated with algorithmic collusion and ensuring compliance through dynamic monitoring [7][8]. - Eight specific risk scenarios are provided in the guidelines, enhancing operational clarity for platforms and addressing potential anti-competitive behaviors [9]. Group 3: Addressing Issues from "Double Eleven" - The guidelines prohibit irrational price wars initiated by dominant platforms, establishing boundaries for competitive practices during promotional events [11]. - The use of AI for price discrimination and customer profiling is restricted, ensuring that platforms cannot justify differential pricing based on user data [12]. - The guidelines protect merchants' pricing autonomy, preventing platforms from coercing them into participating in promotional activities or bearing costs that should be the platform's responsibility [13][14]. - Restrictions on "blocking and shielding" behaviors are detailed, promoting interconnectivity and reducing operational costs for merchants [15]. Group 4: Conclusion - The "Internet Platform Antitrust Compliance Guidelines" establish a more refined and forward-looking compliance framework, guiding platform economies towards high-quality development driven by technology and ecological cooperation [16].

Core Insights - The rise of AI and usage-based billing is fundamentally reshaping the business models and organizational structures of SaaS companies, as highlighted by Metronome's rapid growth and adaptation in this new landscape [1][21]. Pricing Transformation - Pricing is no longer just a financial action but an integral part of the product experience, necessitating a shift in how companies approach billing systems [3][18]. - The traditional SaaS pricing model has evolved through three stages: On-Prem (perpetual licensing), Cloud (seat-based subscriptions), and now to the AI era, which focuses on value generated [4][5]. Challenges of Usage-Based Billing - Implementing usage-based billing presents significant challenges, including the need for real-time monitoring, complex pricing logic, and the necessity for financial-grade data accuracy [7][9][10]. - Companies must adapt their entire operational framework to align with usage-based pricing, requiring a comprehensive redesign of their business engines [11][13]. Organizational Restructuring - The shift to usage-based billing necessitates a redefinition of roles across departments, including sales, customer success, product teams, and finance, to ensure alignment with customer usage and value delivery [14][16][17]. - CEOs play a crucial role in driving this transformation by setting clear timelines and responsibilities for the transition to usage-based models [17]. Value as a Brand Strategy - Pricing strategies are increasingly viewed as a market weapon, with companies leveraging innovative pricing models to enhance brand perception and customer engagement [20]. - The AI-driven market is entering a phase where the ability to effectively implement usage-based pricing will determine competitive advantage and market leadership [22][23]. Conclusion - Usage-based billing is not merely a pricing strategy but a foundational element of future AI enterprise organizational structures, requiring updates across product design, sales incentives, financial logic, and technical systems [23].

Core Viewpoint - Traditional pricing models are becoming ineffective due to value misalignment and cost pressures, leading to a rising demand for disruptive pricing strategies in software companies [3][6]. Group 1: Trends in AI Pricing - A study of over 240 software companies revealed five key trends in AI pricing, indicating a shift from traditional fixed pricing to hybrid pricing models [4][11]. - The proportion of fixed fee subscriptions decreased from 29% to 22%, while hybrid pricing models increased from 27% to 41% [11]. - More than half of the surveyed companies (53%) have integrated AI functionalities into their core software products [9]. Group 2: Challenges and Considerations - Many companies are unprepared for the rapid changes in pricing models, with 75% of software companies adjusting their pricing strategies in the past year [51]. - There is a significant personnel gap in pricing analysis and market insight, with many companies still relying on outdated tools like Excel [52][53]. - The complexity of pricing structures, especially with the introduction of AI, leads to confusion among buyers, who prefer direct communication over static price lists [50][48]. Group 3: Future of Pricing Models - The industry is transitioning from ownership to rental and then to usage-based pricing, which could fundamentally change how software companies operate [57]. - Companies are increasingly leaning towards outcome-based pricing, which ties pricing to the results delivered to customers [56][36].

Core Viewpoint - Traditional pricing models in the software industry are becoming ineffective due to value misalignment and cost pressures, leading to a rising demand for innovative pricing strategies, particularly in SaaS and AI hybrid products [3][6]. Group 1: Trends in AI Pricing - A study of over 240 software companies revealed five key trends in AI pricing, indicating a shift from fixed and seat-based pricing to hybrid pricing models [4][11]. - The proportion of companies using fixed fee subscriptions decreased from 29% to 22%, while those adopting hybrid pricing rose from 27% to 41% [11]. - More than half (53%) of respondents are integrating AI features into their core software products, highlighting the increasing convergence of AI and software [9][10]. Group 2: Hybrid Pricing Models - Hybrid pricing, which combines subscription and usage-based models, has become the mainstream approach, allowing companies to meet diverse customer needs while maintaining simplicity [16][20]. - Companies like Clay have successfully implemented hybrid pricing strategies, offering small discounts and allowing unused credits to roll over, enhancing customer retention [17][20]. - The popularity of hybrid pricing stems from its ability to integrate into existing pricing structures without causing significant disruption [18][20]. Group 3: Challenges in Pricing Transition - As more AI products adopt hybrid pricing, companies face challenges in developing suitable pricing strategies, as there are numerous potential combinations [21]. - The transition to outcome-based pricing is slow, with only 5% of respondents currently using this model, while 25% expect to adopt it by 2028 [27]. - Companies must address four critical factors (CAMP: Consistency, Attribution, Measurability, Predictability) to successfully implement outcome-based pricing [35][36][37][38]. Group 4: Price Transparency - The trend towards price transparency is often overestimated, as many companies still struggle with complex pricing structures and fear that pricing will overshadow their value proposition [39][42]. - While companies with lower average contract values (ACV) tend to publish pricing information, this practice is less common among larger firms [44]. - Increased pricing complexity, such as hybrid models with AI credits, leads buyers to prefer direct communication over relying solely on online pricing [46]. Group 5: Preparedness for Pricing Changes - The rapid evolution of AI technology necessitates a reevaluation of existing pricing models, with 75% of software companies adjusting their pricing strategies in the past year [48]. - Many companies lack the necessary personnel and tools to support strategic pricing decisions, resulting in a gap in capabilities [49][50]. - As companies grow, pricing often becomes a contentious issue among various departments, leading to a lack of clear ownership and strategic direction [52]. Group 6: Future of Pricing Models - There is optimism regarding usage-based and hybrid pricing models as transitional phases towards more sophisticated outcome-based pricing [53]. - The evolution of pricing models reflects a broader shift in the software industry from ownership to rental and then to usage-based models, ultimately aiming to align supplier accountability with customer outcomes [54].