中证报中证网讯（王珞）昆仑万维（300418）近日公布的2025年第三季度报告显示，公司第三季度实现 归属于上市公司股东的净利润为1.90亿元，环比扭亏为盈。 公司第三季度报告显示，在AI技术研发与应用落地的推动下，昆仑万维AI战略已从技术积累期迈入商 业回报期，其全球化的AI产品矩阵正展现出增长潜力与商业价值。 第三季度环比扭亏为盈 国际竞争力持续增强 "依托强大的创新能力与完善的产品矩阵，公司在AI生态构建与商业化进程中取得显著成效，整体经营 表现强劲。"昆仑万维在报告中表示。 昆仑万维2025年前三季度营业收入增长，主要由于报告期公司短剧、广告及AI软件技术业务发展，收 入增加。 昆仑万维相关负责人表示："公司的整体战略是在多个细分场景推出更好的AI产品，通过吸引更多用户 扩大用户基数，重点是让C端用户感受到产品的实用性，从而夯实公司在AI布局中的用户基础。现在AI 产品已经广泛普及，目前正处于市场激烈竞争阶段，公司在基本盘方面已经站稳脚跟，未来发展值得期 待。" 报告期内，公司在AI大模型、AI智能助手、AI音乐与音频、AI视频、AI社交及短剧等多个领域均取得 了突出进展。在AI智能助手领域，公司正式 ...

Core Insights - Tech investors are facing a paradox as both Meta Platforms Inc and Microsoft Corp reported better-than-expected quarterly results, yet their stock prices are under pressure [1] Company Performance - Meta Platforms Inc and Microsoft Corp both exceeded earnings expectations for the quarter, indicating strong performance in the AI sector [1] - Despite the positive earnings reports, both companies are experiencing stock price declines, highlighting a disconnect between financial performance and market sentiment [1] Market Reaction - The market's reaction to the earnings reports of these tech giants suggests a cautious outlook among investors, potentially due to broader economic concerns or market volatility [1] - The contrasting performance of these companies in the stock market raises questions about investor confidence in the tech sector, particularly in relation to AI investments [1]

Core Insights - The article discusses a year-long research project aimed at teaching 100 elderly individuals how to use large AI models, exploring the impact of AI on their lives and how they redefine their understanding of technology [1][6][43] - The research highlights the diverse experiences of elderly users, emphasizing their emotional needs and the importance of trust in human-machine interactions [20][31][42] Group 1: Research Overview - The project was initiated by Fudan University's Aging Research Institute, AI for Good and Smart Elderly Care Research Center, and Tencent, focusing on integrating AI into the lives of older adults [1][6] - The study involved a structured process of teaching, using, tracking, and interviewing elderly participants to understand their interactions with AI [6][10] Group 2: User Profiles and Data Collection - The research included a diverse group of 100 elderly individuals, with tailored teaching methods to accommodate varying technological backgrounds across different regions [10][11] - A comprehensive data collection framework was established, resulting in over 10,236 pieces of data, including voice and text records, to analyze user behavior and needs [11][14] Group 3: Emotional and Social Dynamics - Many elderly participants expressed initial reluctance to adopt new technology, viewing it as non-essential to their fulfilling lives [15][16] - Emotional connections were emphasized, with participants preferring human interaction over technology, indicating a desire for companionship rather than replacement [18][19] Group 4: Trust Calibration in AI Usage - Trust calibration emerged as a critical factor in the interaction between elderly users and AI, with many participants adjusting their trust levels based on their experiences [20][21] - Different types of trust calibration were identified, including limited correction, collaborative reciprocity, and cognitive fixation, affecting how users engaged with AI [21][22] Group 5: Gender Dynamics in Technology Adoption - Gender roles within families influenced the time and resources available for elderly women to engage with technology, leading to disparities in usage [27][29] - Self-deprecating beliefs about technology competence were more prevalent among elderly women, impacting their willingness to explore AI [30] Group 6: Long-term Engagement and Emotional Needs - The relationship between elderly users and AI evolved from initial curiosity to emotional reliance, with many finding comfort and companionship in their interactions [31][32] - Users expressed a desire for AI that understands and empathizes with their daily lives, highlighting the need for technology that is not just functional but also emotionally supportive [36][41] Group 7: Future Directions for AI Development - The findings suggest that future AI technologies should not merely simplify existing tools for the elderly but should be designed to understand and respond to their unique emotional and social needs [44][45] - The research advocates for a more inclusive approach to technology design, emphasizing the importance of empathy and understanding in creating user-friendly AI for older adults [44][45]

Financial Highlights - In Q3 2025, Kunlun Wanwei achieved revenue of 2.072 billion yuan, a year-on-year increase of 56.16% [1][2] - The net profit attributable to shareholders reached 190 million yuan, marking a significant year-on-year growth of 180.13% [1][2] - For the first three quarters of 2025, total revenue was 5.805 billion yuan, up 51.63% year-on-year, with overseas business revenue accounting for 93.3% of total revenue [1][2] AI Business Developments - Kunlun Wanwei has launched several AI models since July, including Skywork-R1V 3.0 and Matrix-3D, showcasing strong innovation capabilities [1][2] - The company has also introduced the AI Developer (Vibe Coding Agent) in its overseas product lineup [3] DramaWave Performance - DramaWave, the short drama platform, has seen rapid growth, ranking third in overseas short drama revenue as of August 2025, with monthly downloads exceeding 4 million [3] - In the last 30 days, DramaWave's advertising material volume has increased from approximately 25,000 to 30,000 daily [4] Audience Insights - DramaWave's user demographics show a predominance of female users at 77.59%, with the largest age group being 25-34 years old at 37.96% [9] - In contrast, the audience for anime dramas on the platform is primarily male, with male users making up between 82.25% and 90.91% of the audience for high-exposure anime materials [12]

OpenAI is now the world’s most valuable private company, but when Microsoft originally invested $1 billion in the startup in 2019, it was less than a sure bet. Microsoft CEO Satya Nadella faced pushback even from the company’s cofounder and original CEO, Bill Gates, he recalled during an interview on tech-focused YouTube channel TBPN. “Remember this was a nonprofit, and I think Bill [Gates] even said, ‘Yeah, you’re going to burn this billion dollars,’” Nadella said. Yet Nadella and the Microsoft team we ...

Core Insights - The collaboration between Xiamen University and Tencent has produced a highly regarded paper titled "FlashWorld: High-quality 3D Scene Generation within Seconds," which has gained significant attention both domestically and internationally, ranking first on the Huggingface Daily Paper list and receiving endorsements from prominent AI figures [2][4]. Group 1: FlashWorld's Performance - FlashWorld achieves 3D scene generation in 5 to 10 seconds on a single GPU, representing a speed increase of up to 100 times compared to previous methods [4]. - The generated scenes can be rendered in real-time on web user interfaces, surpassing the quality of other closed-source models [4]. - In comparative tests, FlashWorld produced stable, complete, and high-quality rendering results, being five times faster than the quick mode of Marble and eliminating the need for backend GPU connections like RTFM [6][10]. Group 2: Technical Approach - FlashWorld utilizes a technology route based on 3DGS for scene output, allowing for local web rendering, which is a significant advantage over video models that require heavy loads [8]. - The method combines a multi-view diffusion model with a three-dimensional focus, enhancing visual quality through a distillation process that ensures multi-view consistency and reduces denoising steps [10][12]. - The training process includes dual-mode pre-training and cross-mode post-training, which enhances the model's ability to generalize across various scenes, styles, and trajectories without needing ground truth data [13][16]. Group 3: Experimental Results - FlashWorld has demonstrated superior performance in generating structured scenes, such as fences, which were previously challenging to achieve [18]. - The model excels in generating fine details, such as hair, from text inputs, showcasing its capability in dense perspective reconstructions [21]. - In benchmark tests, FlashWorld outperformed other methods in speed and quality, achieving the highest average scores in various qualitative metrics [23][24].

Core Insights - Meta CEO Mark Zuckerberg emphasizes the importance of investing heavily in artificial intelligence, suggesting that over-investment is preferable to under-investment [2][4] Investment Strategy - Meta has made a significant $14.3 billion investment in Scale AI to revamp its AI unit, now called Superintelligence Labs [2] - The company is expanding its data centers and forming cloud-computing partnerships with Oracle, Google, and CoreWeave to support its AI ambitions [3] Financial Performance - Meta's capital expenditures (capex) for the year are now projected to be between $70 billion and $72 billion, an increase from the previous guidance of $66 billion to $72 billion [5] - In the third quarter, Meta reported a revenue increase of 26% year-over-year, reaching $51.24 billion, surpassing analyst expectations of $49.41 billion [8] Market Position - Despite the positive revenue growth, Meta's stock fell approximately 8% following the earnings call, contrasting with Alphabet's stock, which rose by 6% [6] - Zuckerberg indicated that if Meta were to have excess computing power, it could potentially offer this capacity to third parties, although this is not currently a concern [6] Future Outlook - Zuckerberg believes that the substantial investments in AI will yield significant returns over time, contributing positively to Meta's core business [4][8] - In a worst-case scenario, if Meta ends up with excess data center capacity, it would lead to asset depreciation, but the company expects to utilize this capacity effectively over time [7]

The company said on the post-release earnings call that “demand again exceeded supply across workloads,” and Microsoft now expects to be capacity-constrained through at least the end of its fiscal year. CEO Satya Nadella said on the earnings call that the company “say[s] no to some of the demand that may be something that we could serve, but it’s not in our long-term interest.”Microsoft now sees Azure growing about 37% this coming quarter — fast enough to keep the company bumping up against its own power li ...

Microsoft CEO Satya Nadella departs following a meeting of the White House Task Force on AI Education in the East Room of the White House in Washington on Sept. 4, 2025.Microsoft was hit with outages in its Azure cloud and 365 services on Wednesday, hours before the company's scheduled earnings release. Users reported problems on social media accessing their sites and services running on Microsoft's products, and the company's website was inaccessible. The problems began around 11:40 a.m. ET, according to D ...

10月29日晚间，昆仑万维（300418）发布2025年第三季度报告。数据显示，前三季度公司实现营业收入 58.0亿元，同比增长52%。公司AI相关业务收入同比大幅增长，进一步巩固了行业领先地位。同时，公 司实现海外业务收入54亿元，同比增长58%，海外收入占比达93%，同比提升3.6个百分点，国际竞争力 持续增强。公司整体毛利率达67%，继续保持在较高水平。在盈利层面，2025年第三季度实现归母净利 润1.9亿元，同比增长180.13%，环比扭亏为盈。 公司在AGI基础研究领域也取得了多重学术突破。9月19日，公司发表的论文《Improving Multi-Step Reasoning Abilities of Large Language Models with Direct Advantage-Based Policy Optimization》，被机器 学习顶会NeurIPS 2025选为Spotlight论文。 AI智能助手方面，公司推出新一代AI智能体，致力于不断降低专业研究与应用开发的门槛。公司正式 发布Skywork Deep Research Agent V2。9月，天工超级智能体在海外版产 ...