Core Insights - Mattel has partnered with OpenAI to integrate generative AI into toy development and IP revitalization, marking OpenAI's first collaboration with a toy manufacturer [1][4] - The partnership aligns with OpenAI's strategy to promote AI across various industries and aims to enhance Mattel's expanding entertainment segment [1][4] Group 1: Partnership Details - The collaboration will focus on creating AI-powered products, with the first product expected to be released later this year [4] - Mattel emphasizes the importance of privacy and security in the development of these AI products [4] - Mattel's Chief Franchise Officer, Josh Silverman, stated that the company will retain control over its IP and will not license it to OpenAI [4] Group 2: Business Expansion - In 2023, Mattel achieved significant success with the movie "Barbie," targeting adult audiences [4] - The company is planning a series of films based on its classic toys, including Hot Wheels, Polly Pocket, Barney, Uno, and Masters of the Universe [4] - Mattel is also venturing into digital gaming, with plans to launch its first self-published game by 2026 [4] Group 3: Employee Tools - As part of the partnership, Mattel employees will gain access to tools like ChatGPT Enterprise to enhance creative brainstorming and product development efficiency [5]

Canalys （现 并入 Omdia ）的最新 数据显示 ， 2025 年第一季度全球云基础设施服务支出达 909 亿 美元，同比增长 21% 。 企业 逐渐 认识到，部署 AI 应用需重新聚焦云迁移，从而推动云投资持续增 长。云与 AI 基础设施的大规模投入，已成为 2025 年市场发展的关键主题之一。 与此同时，为加速企业 大规模采用 AI ， 领先云 厂 商正加快基础设施优化，尤其通过自研芯片的方式，旨在降低 AI 使用成 本、提升推理效率 。 2025 年第一季度，三大云 厂 商（亚马逊云科技 、微软 Azure 和谷歌云 ）的市 场排名保持不变，合计占据全球云支出的 65% 。三家头部云厂商的云相关支出合计同比增长 24% 。 2025年第一季度，头部云厂商的增长态势出现分化。微软Azure和谷歌云均实现超过30%的同比增长 （尽管谷歌云的增速较上一季度略有放缓），而亚马逊云科技（AWS）增长17%，较2024年第四季度的 19%出现放缓。这一减速主要受到供应侧限制影响，限制了其满足快速增长的AI相关需求的能力。对此， 头部云厂商持续加大对AI基础设施的投资，扩大算力容量，布局长期增长。 整体来看 ...

Core Viewpoint - Adobe's latest quarterly performance and sales outlook exceeded Wall Street analysts' expectations, but investor skepticism remains regarding its ability to compete against AI-focused companies like OpenAI's Sora and Runway in the creative software market [1][2][6]. Financial Performance - For the third fiscal quarter of 2025, Adobe expects overall sales to reach between $5.88 billion and $5.93 billion, surpassing the average analyst expectation of approximately $5.88 billion [1]. - Non-GAAP profit per share is projected to be between $5.15 and $5.20, compared to the average analyst estimate of $5.11 [1]. - Adobe's second fiscal quarter sales grew by 11% year-over-year to $5.87 billion, exceeding the average analyst expectation of $5.8 billion [8]. AI Integration and Product Development - Adobe has integrated generative AI features into its flagship products like Photoshop, Premiere, and Illustrator, creating a new "AI family bucket" model [2][8]. - The Firefly AI series has been used over 24 billion times, generating more than 24 billion units of AI content, indicating significant user engagement [3]. - Adobe's Firefly Video Model and "Text-to-Video" capabilities are being tested and integrated into its creative software workflow, enhancing video editing efficiency [9][10]. Market Position and Competitive Landscape - Despite a brief surge in stock price post-earnings, Adobe's shares have faced a decline of about 7% year-to-date, underperforming the S&P 500 index [6]. - Analysts express that the market may misunderstand Adobe's position in the face of AI competition, suggesting that the company's technological advancements are not being fully recognized [2]. - Adobe's strategy focuses on copyright compliance and workflow integration to capture market share in the AI application software sector, competing directly with emerging players like Sora and Runway [11][12]. Industry Trends - AI-related spending is becoming a top priority for enterprises, with expectations that AI-related expenditures will account for 27.7% of software budgets by mid-2025, increasing to 31.6% by 2026 [16].

Group 1 - Amazon announced an investment of over $20 billion in Pennsylvania to expand cloud computing infrastructure and advance AI innovation, expected to create at least 1,250 high-skilled jobs [2] - The investment will enhance Amazon's AI capabilities and infrastructure, reinforcing its position in the AI sector [2][4] - The investment plan will also generate thousands of jobs in the AWS data center supply chain [2] Group 2 - Amazon's entry into AI e-commerce appears delayed compared to competitors like Alibaba, which launched its AI business assistant in 2024 [3] - Challenges for Amazon in AI e-commerce include data privacy concerns, the need for improved algorithm accuracy, and increasing competition from emerging platforms like Temu [3] - Despite these challenges, Amazon possesses unique advantages such as vast consumer data, a mature e-commerce ecosystem, and strong cloud computing infrastructure through AWS [4]

生成式AI 一、 Meta开源发布V-JEPA 2世界模型：开启物理推理新时代 1. Meta开源V-JEPA 2世界模型，能理解物理世界，用100万小时视频数据训练，可实现零 样本规划和机器人控制； 2. 模型仅需62小时训练即可生成规划控制模型，在行为分类和预测方面达到顶级表现，成功 率达65%-80%； 3. Meta发布三个物理理解基准测试，揭示AI与人类在物理推理能力上仍存在差距，未来将发 展分层次和多模态JEPA模型。 https://mp.weixin.qq.com/s/M1mKgpz4ecCIC3xKq50k-A 二、 小扎AGI重新 组队， 科技公司的顶尖工程师们应来尽来 1. Meta CEO扎克伯格正组建"超级智能"小组，已成功挖角谷歌DeepMind首席研究员Jack Rae和其他AI顶尖人才； 2. Jack Rae是"压缩即智能"思想代表人物，曾负责Gemini的"思考"功能，在DeepMind工作 7年参与多个重要模型开发； 3. Meta为吸引AI人才提供7-9位数薪酬方案，计划建立约50人规模的团队， 或 以 百 亿美 元收购Scale AI 及 其 团队 。 https:/ ...

Core Viewpoint - Disney and Universal Studios have filed a lawsuit against AI image generator Midjourney for copyright infringement, marking a significant legal battle between Hollywood giants and generative AI companies [2][4]. Group 1: Lawsuit Details - The lawsuit is 110 pages long and accuses Midjourney of using a vast amount of film works to train its AI, allowing users to generate infringing images of iconic characters such as Darth Vader and Minions [4][6]. - Midjourney has generated infringing content from over 20 film IPs, including characters from "Toy Story," "Frozen," and "Shrek," by using prompts like "Disney style" and "Marvel style" [6][10]. - The characters listed in the lawsuit span over five decades, from 1977's Yoda to characters from 2023 releases [10]. Group 2: Midjourney's Business Performance - Midjourney, founded in 2021, reported a revenue surge to $300 million (approximately 2.2 billion RMB) last year, with a user base reaching tens of millions [10]. - The subscription service offers monthly fees ranging from $10 to $120, depending on the level of image generation speed [10]. Group 3: Industry Implications - The lawsuit is viewed as a milestone in the ongoing debate over generative AI and copyright infringement, with the film industry warning that AI infringements could disrupt the incentive mechanisms of U.S. copyright law [10]. - The U.S. film industry has a salary scale of $229 billion (approximately 1.66 trillion RMB) and supports 2.3 million jobs [10]. Group 4: Midjourney's Response and Industry Context - Midjourney has not yet responded to the lawsuit and claims to be an "independent research lab" with a team of fewer than 12 people [11]. - The Hollywood industry has had a conflicted stance on AI technology, having previously seen strikes over AI threats, while also utilizing AI in Oscar-nominated films for tasks like voice correction [13].

《科创板日报》12日讯，日前Meta AI宣布推出首个生成式全新视频编辑功能，并在Meta AI应用、 Meta.AI网站以及Edits应用中上线，用户可通过50多种预设AI提示词，一键实现视频中服装、场景、光 影等元素的智能变换。该功能目前已在美国及全球超12个国家上线，限时免费支持编辑10秒视频片段， 让用户能够通过预设的AI视频编辑提示词，实现创意构想并分享或发布编辑后的视频。 Meta AI推首个生成式AI视频编辑功能 秒换服装、场景、灯光，可免费编辑10秒 ...

Core Insights - The article discusses the success of "AI Baby" videos, which have gained millions of views and likes, driven by the innovative platform Hedra that recently raised $32 million in Series A funding, achieving a valuation of $200 million [11][18]. Group 1: Hedra's Technology and Features - Hedra's platform utilizes a "full-modal" approach, integrating various external models to streamline the content creation process, addressing the fragmentation in generative AI tools [11][28]. - The Character-3 model at the core of Hedra allows for the generation and editing of content in three simple steps: analyzing celebrity traits, generating baby images, and synchronizing audio with facial expressions [12][14][24]. - The platform's precision in lip-syncing and facial expression generation enhances the emotional depth of the content, overcoming the "uncanny valley" effect often seen in AI-generated media [22][24][25]. Group 2: Market Trends and User Engagement - The "AI Baby" concept leverages the appeal of cute aesthetics combined with adult narratives, creating a unique viral effect that resonates with audiences [12][16]. - Hedra has rapidly gained traction, amassing 3 million users since the launch of the Character-3 model in 2024, indicating a strong demand for its innovative content creation capabilities [18]. - The company aims to expand its applications into enterprise marketing, showcasing its potential to reduce production costs for customer service and virtual characters [20]. Group 3: Competitive Landscape - The generative AI space is highly competitive, with companies like Captions, HeyGen, Synthesia, and Runway also developing technologies for video and virtual character creation [21]. - Hedra differentiates itself by offering a comprehensive solution that integrates text input, image generation, voice synthesis, and multi-character animation, addressing the limitations of existing tools [28].

该行称，对谷歌而言，这一合作重要的两点原因如下： 智通财经APP获悉，谷歌云(GOOGL.US)与OpenAI达成协议将支持增加算力以满足不断扩张的计算需 求，摩根士丹利针对该事件发布了相关评论报告，认为若协议得到确认，将表明谷歌对其长期搜索业务 地位的信心，以及推动谷歌云加速发展，估值达到超18倍。该行给予谷歌"增持"评级，上调目标价至 185美元。 大摩认为，若谷歌云(GCP)与OpenAI的协议得到确认，其意义体现在两方面：1)尽管条款未知，但该行 认为OpenAI的推理需求正在快速增长。若这一合作推动谷歌云加速发展，其估值可能达到2026年每股 收益的18倍以上。2)计算能力是一项竞争优势，这表明谷歌对其长期搜索业务地位充满信心。 据报道，谷歌云与OpenAI于5月最终达成协议，OpenAI将使用谷歌云来增加计算能力，以满足其不断增 长的计算需求。若该合作得到确认，意味着OpenAI正继续将其基础设施基础多元化，此前是微软"独 家"合作，而OpenAO今年早些时候已与CoreWeave(CRWV.US)签署了协议，去年与甲骨文(ORCL.US)也 有合作。 该行尚未看到明确报告提及OpenAI将使用T ...

资料图 日本政府的税制调查会通过专家会议讨论了调整小额进口货物的免税政策。日本目前对1万日元以下的 进口货物免除消费税和关税。这被认为支持了中国"Temu"和"SHEIN"等…… 半导体存储器之一的DRAM价格正在持续上涨。指标产品的5月大宗交易价格连续2个月比上 月上涨10%。主要原因是认为全球存储器厂商停止和缩减包括指标产品在内的品类生产的预 期加强，产品短缺持续。部分需求方出现了押注短期价格上涨、开始囤积库存的趋势。 DRAM嵌入个人电脑（PC）、智能手机和数据中心的服务器设备等，用于数据的临时存储。 大宗交易价格由作为卖方的存储器制造商、作为买方的设备制造商和模块制造商按月或按季 度决定。 从5月份的大宗交易价格来看，作为指标的8GB DDR4产品为每个2.06美元左右，创出自 2024年8月以来的高点。容量较小的4GB产品也达到1.57美元左右，提高10%，涨幅连续2 个月达到10%。 关于长鑫存储技术，相关人士表示"将加速节电产品LPDDR4和属于下一代产品的DDR5的生 产"，包括中国企业在内的竞争激化，有可能影响到DDR5以后的价格。 不过，DDR4价格上升趋势的前景仍存在不确定性。由于PC和 ...