Group 1 - U.S. stock market experienced a significant drop on September 23, with major indices closing lower: Dow Jones down 0.19%, Nasdaq down 0.95%, and S&P 500 down 0.55% [1] - Major tech stocks saw declines, with Oracle down over 4%, Amazon down over 3%, Nvidia down over 2%, and Microsoft down over 1% [1] - The Nasdaq Golden Dragon China Index fell by 2.22%, with notable declines in Chinese concept stocks: Baidu down over 8%, Zhihu down over 5%, Bilibili down over 4%, and New Oriental and JD down nearly 3% [1] Group 2 - Federal Reserve Chairman Jerome Powell indicated increased risks in the job market, which influenced the Fed's recent decision to lower interest rates [1] - Powell described the Fed's policy shift as moving towards a "neutral" stance, emphasizing that there is no preset direction for future policies [1] - Current inflation levels remain slightly above target, with the August core PCE inflation rate expected to be 2.3%, primarily driven by tariff impacts rather than widespread inflationary pressures [1] Group 3 - Powell noted that some asset prices are relatively high compared to historical levels, with stock price valuations appearing quite elevated based on various indicators [1] - Signs of slowing consumer spending and weakened business confidence were highlighted, alongside a decrease in labor market vitality [1] - Powell anticipates that tariffs may lead to a temporary rise in inflation over the next few quarters, but the Fed aims to prevent transient price increases from becoming a persistent issue [1] Group 4 - International gold prices increased, with COMEX gold futures rising by 0.58% to $3796.9 per ounce, and COMEX silver futures up by 0.12% to $44.265 per ounce [2]

Market Overview - The market is expected to maintain a range-bound movement between 3700 and 3900 in the short term [1][16] - Major U.S. stock indices experienced a collective decline, with the Dow Jones down 0.19%, S&P 500 down 0.55%, and Nasdaq down 0.95% [4][3] - Large tech stocks faced pressure, with notable declines in Amazon (down 3.04%), Nvidia (down 2.82%), and Tesla (down 1.93%) [4] Economic Indicators - The preliminary Purchasing Managers' Index (PMI) for the U.S. showed a slowdown in both manufacturing and services, with the manufacturing PMI at 52.0, down from 53.0 in August [4] - The OECD projected a global economic growth rate of 3.2% for 2025, an increase of 0.3 percentage points from its June forecast [7] Corporate Actions - Lakala announced that Legend Holdings reduced its stake by 807.81 million shares, bringing its holding to 23.97% [11] - Xinhuajin is at risk of being subject to other risk warnings and potential delisting due to non-operational fund occupation amounting to 406 million yuan [12] - Weixinno plans to invest 5 billion yuan in a new display industry innovation center in collaboration with Kunshan Weixin [13] - Ganli Pharmaceutical signed a supply framework agreement worth no less than 3 billion yuan related to insulin products [14]

知乎COO张荣乐表示："知乎与CICAS组委会此次联合发起的，不仅仅是一场比赛。我们更希望借助双方的平台与专业力量，连接起科技从业者、高校 师生和广大科技爱好者，促进人工智能领域的创新链、人才链和产业链深度融合，为实体经济的数字化和智能化发展，搭一座桥、尽一份力。" 赛事报名时间为2025年9月10日24:00至10月31日24:00，符合条件的个人开发者、高校或企业团队可登录CICAS官网或通用人工智能专项赛报名网站了解 赛事详情。 （注：此文属于央广网登载的商业信息，文章内容不代表本网观点，仅供参考。） CICAS大赛是由中国人工智能学会主办的全国性人工智能行业应用赛事，今年共计将举办13场专项赛，其中通用人工智能（AGI）专项赛旨在以场景创 新为牵引，充分发挥"以赛助研、以赛促评、以赛带教、以赛定标、以赛引才、以赛育产、以赛办展"创新驱动作用，推动通用人工智能与未来产业、实体经 济高质量发展。会上，中国人工智能学会副秘书长、CICAS组委会执行秘书长余有成与知乎COO张荣乐共同参与了专项赛启动仪式。 本届专项赛采用"开放场景"竞赛模式，面向全国各地公开征集不少于120个通用人工智能场景的应用创新项目参赛 ...

Group 1 - The Zhihu-W TechClub conference focused on the theme "Reconstructing Technology Under AI," bringing together experts, executives, AI developers, and Zhihu contributors to discuss AI technology evolution and human-machine collaboration [1][4] - Zhihu announced the "AI Pioneers" and "Tech Translator Brand" lists, recognizing developers and brands leading technological and product innovation [3][11] - The conference featured discussions on the future of AI technology and its impact on human life, with insights from academic experts on AI development paths and the integration of brain-machine interfaces [4][5] Group 2 - The conference highlighted the importance of AI in various sectors, with examples from China Mobile's AI applications in agriculture, mining, and smart living [5] - A debate among Zhihu contributors addressed the dual nature of AI's impact on society, discussing whether AI enhances or diminishes human creativity and productivity [6][9] - The "AI Pioneers" list included notable figures from academia and industry, showcasing individuals who are driving AI advancements and innovation [13][14] Group 3 - The conference launched the 2025 National AI Application Scenario Innovation Challenge, focusing on general artificial intelligence (AGI) and encouraging participation from research institutions and technology companies [15][17] - The challenge aims to gather innovative projects across various AI application scenarios, emphasizing social concerns and enhancing workplace efficiency [17] - Zhihu aims to connect technology professionals, academia, and the public to transform advanced technologies into practical solutions for everyday life [17]

Core Viewpoint - The implementation of the "Identification Method for AI-Generated Synthetic Content" has led to a requirement for explicit labeling of AI-generated content across various platforms, with varying degrees of compliance observed among different applications and content dissemination platforms [1][2][6]. Group 1: AI Application Compliance - Ten generative AI applications, including Tencent Yuanbao and DeepSeek, have successfully implemented labeling for generated text, images, and videos as per the new regulations [1][2]. - The labeling styles for generated images and videos are generally consistent, often located in the upper left or lower right corners, and remain visible after downloading [4]. - Some applications, like Quark and Doubao, provide additional labeling upon download, enhancing visibility [4]. Group 2: Content Dissemination Platforms - Ten content dissemination platforms, including WeChat Video Number and Douyin, have introduced user self-declaration features, allowing users to label AI-generated content [5]. - However, the accessibility of these self-declaration features varies, with some platforms making it difficult for users to find and utilize them effectively [5]. - Platforms like Douban and Bilibili have shown promptness in adding significant labels to content suspected of being AI-generated, while others like Weibo and Zhihu have room for improvement in their automatic recognition capabilities [6]. Group 3: Expert Recommendations - Experts suggest that the successful implementation of the identification method requires collaboration among various stakeholders to enhance the detection and management capabilities of AI-generated content [7]. - Continuous improvement in technical capabilities and inter-platform cooperation is essential to avoid regulatory gaps and ensure compliance with the identification requirements [7]. - Strengthening user awareness regarding the importance of labeling AI-generated content is crucial for fostering a culture of transparency and accountability in information dissemination [7].

Market Performance - The three major U.S. stock indices closed higher, with the Dow Jones Industrial Average up 0.11%, the S&P 500 up 0.47%, and the Nasdaq Composite up 0.94% [1] - The S&P 500 and Nasdaq reached new closing highs [1] Technology Sector - Major tech stocks saw significant gains, with Google rising over 4%, making it the fourth U.S. publicly traded company to surpass a market capitalization of $3 trillion [1] - Tesla and Oracle both increased by over 3% [1] Chinese Stocks - Most popular Chinese stocks experienced an upward trend, with the Nasdaq Golden Dragon China Index rising by 0.87% [1] - Notable performers included Xiaoma Zhixing up 10.94%, Canadian Solar up 10.28%, and Li Auto, Bilibili, and Daqo New Energy each rising up to 6.9% [1] - Other significant gains included JinkoSolar up 4.90%, NIO up 4.34%, Weibo up 3.67%, and various companies like Zhihu, EHang, and Xpeng Motors at least up 2.22% [1] - Tencent Music, Zeekr, and Alibaba saw increases of at least 1.92% [1]

责任编辑：江钰涵 大会将揭晓"AI先行者"和"TECH TRANSLATOR企业"榜单，从社区视角发掘AI潜力股和优秀企业品 牌。此外，知乎还将联合中国人工智能学会，开启2025第三届全国人工智能应用场景创新挑战赛通用人 工智能赛道，挖掘AI开发新星。 新浪科技讯 9月12日午间消息，首届知乎Tech Club新知科技大会将于9月20日在杭州开幕。上海交大讲 席教授徐雷、格力电器市场总监朱磊、《最强大脑》科学评审魏坤琳等嘉宾将参会，围绕AI技术演 进、人机协同、科技重构等话题展开深度思辨。 ...

[Company Information](index=2&type=section&id=Company%20Information) [Board of Directors and Committee Members](index=3&type=section&id=1.1%20Board%20of%20Directors%20and%20Committee%20Members) The company's board, comprising executive, non-executive, and independent non-executive directors, experienced changes in independent director appointments and committee chairpersons during the reporting period - Mr. Sun Han-hui resigned as **independent non-executive director and from multiple committee roles** on May 27, 2025[3](index=3&type=chunk) - Dr. Cheng Li-lan was appointed as **independent non-executive director, Audit Committee Chairman, and other committee member** on May 27, 2025[3](index=3&type=chunk) - Ms. Ni Hong was appointed as **Remuneration Committee Chairman** on May 27, 2025[3](index=3&type=chunk) [Company Contact and Registration Information](index=3&type=section&id=1.2%20Company%20Contact%20and%20Registration%20Information) The company secretary is Ms. Liu Qihua, with Mr. Zhou Yuan and Ms. Liu Qihua as authorized representatives, and its headquarters are in Beijing, China, with a registered office in the Cayman Islands - The company secretary is **Ms. Liu Qihua**, and authorized representatives are **Mr. Zhou Yuan and Ms. Liu Qihua**[4](index=4&type=chunk) - Headquarters and principal place of business in China are located at No. 18 Xueqing Road, Haidian District, Beijing, China[4](index=4&type=chunk) - The registered office is at PO Box 309, Ugland House, Grand Cayman KY1-1104, Cayman Islands[4](index=4&type=chunk) [Professional Advisors and Services](index=4&type=section&id=1.3%20Professional%20Advisors%20and%20Services) The company's auditor is PwC, with legal advisors covering Hong Kong, US, and Cayman Islands law, and Guotai Junan Securities as compliance advisor - The auditor is PricewaterhouseCoopers[4](index=4&type=chunk) - Legal advisors cover Hong Kong, US, and Cayman Islands law[5](index=5&type=chunk) - The compliance advisor is Guotai Junan Capital Limited[5](index=5&type=chunk) [Share Information and Website](index=4&type=section&id=1.4%20Share%20Information%20and%20Website) The company's stock codes are 2390 on SEHK and ZH on NYSE, with its official website being ir.zhihu.com - SEHK stock code: **2390**, NYSE stock code: **ZH**[5](index=5&type=chunk) - Company website: ir.zhihu.com[5](index=5&type=chunk) [Key Highlights](index=5&type=section&id=Key%20Highlights) [Financial Performance Summary](index=5&type=section&id=2.1%20Financial%20Performance%20Summary) For the six months ended June 30, 2025, total revenue decreased by 23.7% to RMB 1,446.6 million, but operating and net losses significantly narrowed, achieving a net profit of RMB 62.4 million and an adjusted net profit of RMB 98.3 million, turning losses into profits Key Financial Data for the Six Months Ended June 30 (RMB in thousands) | Metric | 2025 | 2024 | Change (%) | | :--- | :--- | :--- | :--- | | Total Revenue | 1,446,559 | 1,894,673 | (23.7)% | | Gross Profit | 899,287 | 1,100,023 | (18.2)% | | Operating Loss | (143,609) | (408,619) | (64.9)% | | Net Profit / (Loss) | 62,372 | (246,391) | N/A | | Adjusted Operating Loss | (106,501) | (340,686) | (68.7)% | | Adjusted Net Profit / (Loss) | 98,280 | (180,283) | N/A | [Operational Data Summary](index=5&type=section&id=2.2%20Operational%20Data%20Summary) For the six months ended June 30, 2025, average monthly subscribing members were 13.7 million, a 6.7% decrease year-over-year Average Monthly Subscribing Members for the Six Months Ended June 30 (in millions) | Metric | 2025 | 2024 | Change (%) | | :--- | :--- | :--- | :--- | | Average Monthly Subscribing Members | 13.7 | 14.7 | (6.7)% | [Non-GAAP Financial Measures](index=6&type=section&id=2.3%20Non-GAAP%20Financial%20Measures) The company uses non-GAAP financial measures like adjusted operating loss and adjusted net profit/(loss) to exclude share-based compensation, intangible asset amortization from acquisitions, and non-GAAP tax effects, providing a clearer view of operational performance and period-over-period comparison - Non-GAAP financial measures exclude share-based compensation expenses, amortization of intangible assets from business acquisitions, and non-GAAP tax effects[9](index=9&type=chunk) Reconciliation of GAAP and Non-GAAP Financial Measures (RMB in thousands) | Metric | 2025 | 2024 | | :--- | :--- | :--- | | Operating Loss | (143,609) | (408,619) | | Add: Share-based compensation expenses | 30,128 | 58,453 | | Add: Amortization of intangible assets from business acquisitions | 6,980 | 9,480 | | Adjusted Operating Loss | (106,501) | (340,686) | | Net Profit / (Loss) | 62,372 | (246,391) | | Add: Share-based compensation expenses | 30,128 | 58,453 | | Add: Amortization of intangible assets from business acquisitions | 6,980 | 9,480 | | Add: Non-GAAP adjustments for income tax effects | (1,200) | (1,825) | | Adjusted Net Profit / (Loss) | 98,280 | (180,283) | [Business Review and Outlook](index=7&type=section&id=Business%20Review%20and%20Outlook) [Business Review for the Reporting Period](index=7&type=section&id=3.1%20Business%20Review%20for%20the%20Reporting%20Period) In the first half of 2025, Zhihu achieved profitability under both GAAP and non-GAAP measures, deepened strategic adjustments to enhance community trustworthiness, fostered high-quality content, a trusted creator network, and AI capabilities, while significantly improving its gross margin to 62.2% - In the first half of 2025, the company achieved profitability under both GAAP and non-GAAP measures, with an adjusted net profit of **RMB 98.3 million**, compared to an adjusted net loss of RMB 180.3 million in the same period of 2024[14](index=14&type=chunk) - Gross margin increased to **62.2%**[14](index=14&type=chunk) - The company continued to deepen its strategic adjustments, focusing on enhancing the trustworthiness of the Zhihu community and promoting the synergistic development of high-quality content, a trusted creator network, and AI capabilities[14](index=14&type=chunk) [Financial and Strategic Highlights](index=7&type=section&id=3.1.1%20Financial%20and%20Strategic%20Highlights) In the first half of 2025, the company achieved profitability and improved gross margin, consolidating its long-term competitive advantage through resource reallocation, organizational optimization, and enhanced user experience, while focusing on high-quality content, trusted creators, and AI synergy - Achieved an adjusted net profit of **RMB 98.3 million** in the first half of 2025, with gross margin rising to **62.2%**[14](index=14&type=chunk) - Deepened strategic adjustments, focusing on enhancing **community trustworthiness**, including resource reallocation, organizational optimization, and improved user experience[14](index=14&type=chunk) [AI Progress](index=7&type=section&id=3.1.2%20AI%20Progress) Zhihu views AI as a historic opportunity, with its community becoming a hub for AI-related topics, accelerating investments in AI applications, launching new features like Zhihu Direct Answer, and leveraging AI to enhance content identification, moderation, paid membership promotion, and vocational training sales efficiency - Zhihu community has become a **center for AI-related topics**, with high-quality professional content increasingly cited by leading external AI large models and AI search engines[16](index=16&type=chunk) - Zhihu Direct Answer active usage penetration significantly increased, and a public knowledge base was launched to **enhance user experience and reduce AI hallucinations**[17](index=17&type=chunk) - AI applied to content identification, content moderation, paid membership promotion, and vocational training sales services, comprehensively **improving operational efficiency**[17](index=17&type=chunk) [Zhihu Content Ecosystem](index=8&type=section&id=3.1.3%20Zhihu%20Content%20Ecosystem) Zhihu continues to produce high-quality content, enhancing content quality through its TopicRank algorithm and trusted content model, with cumulative platform content reaching 911.4 million items by June 30, 2025, a 9.7% year-over-year increase, while also commercializing short-form novel IPs and expanding into mid-to-long-form novels - Based on the TopicRank algorithm, a trusted content model is built, combining AI capabilities with community interaction to dynamically identify and classify content quality[18](index=18&type=chunk) - As of June 30, 2025, cumulative platform content reached **911.4 million items**, a **9.7% year-over-year increase**[18](index=18&type=chunk) - High-quality short-form novel IP commercialization potential unleashed, successfully entering the **short-drama market**, and expanding content supply to mid-to-long-form novels[19](index=19&type=chunk) [Zhihu Users and Creators](index=9&type=section&id=3.1.4%20Zhihu%20Users%20and%20Creators) In the first half of 2025, monthly active users remained stable with increased engagement and core user retention, and significantly longer daily active user visit durations, despite average monthly subscribing members decreasing by 6.7% to 13.7 million, while professional creators grew by 26.9% and AI-powered content creators' activity continued to rise - Monthly active user base remained stable, with **significantly increased monthly active user engagement, core user retention, and daily active user visit duration**[22](index=22&type=chunk) Average Monthly Subscribing Members (in millions) | Metric | 2025上半年 | 2024上半年 | Change (%) | | :--- | :--- | :--- | :--- | | Average Monthly Subscribing Members | 13.7 | 14.7 | (6.7)% | - The number of certified honorary creators increased by **26.9% year-over-year**, and the activity of AI-powered high-end content creators continued to increase[23](index=23&type=chunk) [Monetization Model](index=10&type=section&id=3.1.5%20Monetization%20Model) Total revenue for the first half of 2025 decreased by 23.7% to RMB 1,446.6 million, with paid membership services remaining the largest revenue source but declining by 7.1%, and vocational training revenue also decreasing as the company shifts towards new social interaction and knowledge-sharing driven models 2025年上半年總收入及付費會員收入（人民幣百萬元） | Metric | 2025上半年 | 2024上半年 | Change (%) | | :--- | :--- | :--- | :--- | | Total Revenue | 1,446.6 | 1,894.7 | (23.7)% | | Paid Membership Services Revenue | 819.8 | 882.4 | (7.1)% | | Vocational Training Services Revenue | 156.6 | 279.1 | (43.9)% | - Paid membership services remain the largest revenue source, but a decrease in average monthly subscribing members led to a **slight revenue reduction**[26](index=26&type=chunk) - Vocational training service revenue decreased primarily due to strategic optimization of self-operated course products and acquired businesses, shifting towards **new social interaction and knowledge-sharing driven models**[26](index=26&type=chunk) [Material Events After Reporting Period](index=10&type=section&id=3.2%20Material%20Events%20After%20Reporting%20Period) No material events occurred after the reporting period and up to the latest practicable date that could impact the company, other than those disclosed in this interim report - No material events occurred after June 30, 2025, and up to the latest practicable date that could impact the Group[27](index=27&type=chunk) [Business Outlook](index=10&type=section&id=3.3%20Business%20Outlook) For the second half, the company will deepen user engagement, enhance high-quality content creation, improve its commercial ecosystem, and prioritize accelerating AI integration across the platform and expanding trusted content and services to broader applications beyond the community - For the second half, the company will deepen user engagement, enhance high-quality content creation, improve its commercial ecosystem, and continue to focus on core user experience and community trustworthiness[28](index=28&type=chunk) - Prioritize two key areas: accelerating AI integration across the entire platform, and expanding trusted content and services to broader application scenarios beyond the community[28](index=28&type=chunk) [Management Discussion and Analysis](index=11&type=section&id=Management%20Discussion%20and%20Analysis) [Consolidated Operating Results](index=11&type=section&id=4.1%20Consolidated%20Operating%20Results) For the six months ended June 30, 2025, total revenue was RMB 1,446.6 million, a 23.7% year-over-year decrease, while operating loss significantly narrowed by 64.9% to RMB 143.6 million, and net profit reached RMB 62.4 million, reversing the net loss from the prior year Consolidated Operating Results for the Six Months Ended June 30 (RMB in thousands) | Metric | 2025 | 2024 | | :--- | :--- | :--- | | Total Revenue | 1,446,559 | 1,894,673 | | Cost of Revenue | (547,272) | (794,650) | | Gross Profit | 899,287 | 1,100,023 | | Total Operating Expenses | (1,042,896) | (1,508,642) | | Operating Loss | (143,609) | (408,619) | | Investment Income | 160,185 | 38,713 | | Net Profit / (Loss) | 62,372 | (246,391) | | Net Profit / (Loss) Attributable to Zhihu Shareholders | 62,388 | (247,585) | - Operating loss decreased by **64.9%** from RMB 408.6 million in the same period of 2024 to **RMB 143.6 million**[31](index=31&type=chunk) - Net profit was **RMB 62.4 million**, compared to a net loss of RMB 246.4 million in the same period of 2024[31](index=31&type=chunk) [Revenue Analysis](index=12&type=section&id=4.2%20Revenue%20Analysis) Total revenue for the first half of 2025 was RMB 1,446.6 million, a 23.7% year-over-year decrease, with paid membership remaining the largest revenue source at 56.7% of total revenue but declining by 7.1%, and marketing services and vocational training revenues also decreasing due to service product optimization and strategic adjustments Revenue by Business Line for the Six Months Ended June 30 (RMB in thousands) | Revenue Source | 2025 (RMB) | 2025 (%) | 2024 (RMB) | 2024 (%) | | :--- | :--- | :--- | :--- | :--- | | Marketing Services | 419,737 | 29.0 | 674,521 | 35.6 | | Paid Membership | 819,848 | 56.7 | 882,376 | 46.6 | | Vocational Training | 156,631 | 10.8 | 279,069 | 14.7 | | Others | 50,343 | 3.5 | 58,707 | 3.1 | | Total | 1,446,559 | 100.0 | 1,894,673 | 100.0 | - Marketing services revenue decreased primarily due to the company's continuous proactive optimization of service product offerings, with **profitability improvement as a strategic core**[36](index=36&type=chunk) - Paid membership revenue slightly decreased primarily due to a **decline in average monthly subscribing members**[36](index=36&type=chunk) - Vocational training revenue decreased primarily due to the company's **strategic optimization of self-operated course products and acquired businesses**[36](index=36&type=chunk) [Costs and Gross Profit](index=13&type=section&id=4.3%20Costs%20and%20Gross%20Profit) Cost of revenue decreased by 31.1% year-over-year to RMB 547.3 million, primarily due to reduced content and operational costs, personnel expenses, and cloud services and bandwidth costs, leading to a gross margin increase from 58.1% in 2024 to 62.2% in the first half of 2025, reflecting enhanced commercialization and operational efficiency - Cost of revenue decreased by **31.1%** to **RMB 547.3 million**, primarily due to reduced content and operational costs, personnel-related expenses, and lower cloud services and bandwidth costs resulting from improved technological efficiency[38](index=38&type=chunk) Cost of Revenue Details for the Six Months Ended June 30 (RMB in thousands) | Cost of Revenue Item | 2025 (RMB) | 2025 (%) | 2024 (RMB) | 2024 (%) | | :--- | :--- | :--- | :--- | :--- | | Content and Operational Costs | 280,040 | 19.4 | 396,159 | 20.9 | | Cloud Services and Bandwidth Costs | 80,323 | 5.6 | 129,163 | 6.8 | | Staff Costs | 72,406 | 5.0 | 119,657 | 6.3 | | Payment Processing Fees | 70,531 | 4.9 | 87,199 | 4.6 | | Others | 43,972 | 2.9 | 62,472 | 3.3 | | Total | 547,272 | 37.8 | 794,650 | 41.9 | - Gross margin increased from **58.1%** in the same period of 2024 to **62.2%** in the first half of 2025, primarily due to enhanced commercialization capabilities and improved operational efficiency[41](index=41&type=chunk) [Operating Expenses](index=14&type=section&id=4.4%20Operating%20Expenses) Total operating expenses for the first half of 2025 decreased by 30.9% year-over-year to RMB 1,042.9 million, with significant reductions in sales and marketing, research and development, and general and administrative expenses, driven by disciplined promotional spending, improved R&D efficiency, and lower expected credit loss provisions - Total operating expenses decreased by **30.9%** to **RMB 1,042.9 million**[42](index=42&type=chunk) - Sales and marketing expenses decreased by **27.7%** to **RMB 646.9 million**, primarily due to more disciplined promotional spending and reduced personnel-related expenses[42](index=42&type=chunk) - Research and development expenses decreased by **29.3%** to **RMB 287.5 million**, primarily due to improved R&D efficiency[42](index=42&type=chunk) - General and administrative expenses decreased by **47.6%** to **RMB 108.5 million**, primarily due to a decrease in the allowance for expected credit losses on accounts receivable[42](index=42&type=chunk) [Profitability Metrics](index=14&type=section&id=4.5%20Profitability%20Metrics) Both operating loss and adjusted operating loss significantly narrowed, while net profit and adjusted net profit turned profitable, with investment income substantially increasing due to unrealized gains from fair value remeasurement of an investment in a private company - Operating loss decreased by **64.9%** from RMB 408.6 million to **RMB 143.6 million**[43](index=43&type=chunk) - Adjusted operating loss decreased by **68.7%** from RMB 340.7 million to **RMB 106.5 million**[44](index=44&type=chunk) - Investment income increased to **RMB 160.2 million**, primarily due to unrealized gains from fair value remeasurement of an investment

Core Viewpoint - The 2025 National Science Popularization Month is set to commence in September 2025, with Zhihu launching its Science Season activities, focusing on the theme "AI Critical, Science Connection" to explore scientific development in the AI era [1] Group 1: Event Overview - Zhihu collaborates with Science China to initiate the "Connecting All Things with Science" creative activity, led by three academicians to promote scientific ideas and spirit [1] - The Zhihu Science Season has been held five times, serving as a significant platform for showcasing scientific achievements and transmitting scientific spirit [1] Group 2: Contributions from Academicians - Academician Pang Guofang shares insights on overcoming research bottlenecks, emphasizing that failures and unexpected data can lead to breakthroughs [4] - Academician Han Jisheng encourages young researchers to balance personal interests with national needs, highlighting the importance of both [4] - Academician Li Jian'an discusses a shift in understanding health, advocating for a holistic view of health as harmony between humans and their environment [5] Group 3: Creative Content and Themes - Zhihu Science Month introduces four thematic tracks to create engaging scientific content, focusing on human connection and warmth in science communication [6] - The tracks include "Wonders of Nature," "Physics Q&A," "Natural Observation," and "Silicon-Based Observation," aimed at fostering discussions in various scientific fields [6] - A notable question raised during the event involves the potential for computers to replicate human brain functions, sparking philosophical debates on consciousness [6] Group 4: Community Engagement - The Zhihu Science Season has become a vital event for the scientific community, attracting numerous professionals and enthusiasts from various scientific disciplines [8] - The event features prominent scientists and researchers who engage in discussions and share cutting-edge research, contributing to a vibrant academic environment [8] - Since 2021, Zhihu has hosted the Science Season annually, providing a platform for significant scientific discussions and advancements [8]

南京医科大学教授、主任医师，南京医科大学康复医学院名誉院士，美国医学科学院国际院士励建安则 重点解答了有关运动健康的几大问题。励建安坦言，对健康的理解发生了显著转变，"过去觉得没病就 是健康，现在更认同世界卫生组织对健康的深层定义，健康是人与环境的和谐相处。人的组织器官基本 上都有一个特质，你用它，它就能把自己的能力维系到尽可能长的时间。" 责任编辑：杨赐 当研究走到了瓶颈，该如何突破？中国质量检验检测科学研究院首席科学家、中国工程院院士庞国芳回 答称，每次失败的实验记录、优化的参数、甚至偶然发现的异常数据，都是积累的种子，而这些种子遇 到某个契机，就会像多米诺骨牌一样引发连锁突破，这就是积累的力量。 北京大学博雅讲席教授、北京大学神经科学研究所名誉所长、中国科学院院士韩济生以"德、才、勤"三 个字勉励年轻科研工作者继续努力。他称，"要发展个人的兴趣，同时结合国家的需要，年轻人对这两 种问题都要考虑，这两者是不可偏废的"。 新浪科技讯 9月9日下午消息，2025年9月是首个法定全国科普月，知乎启动2025科学季，并联合科普中 国率先发起"用科普联结万物"创作活动，中国科学院院士韩济生、中国工程院院士庞国芳、美 ...