Group 1 - The 22nd National Cybersecurity Publicity Week revealed the first real-world testing results for AI large models, identifying 281 security vulnerabilities, with over 60% being unique to large models, including risks like prompt injection and information leakage [1] - Attackers are studying AI learning preferences and deliberately feeding false information, with organized efforts to "data poison" AI by fabricating expert identities and creating fake research reports to manipulate AI outputs [1] - The regulatory framework is evolving, with the release of the 2.0 version of the "Artificial Intelligence Security Governance Framework" on September 15 [1] Group 2 - Ant Group's consumer finance division utilizes multimodal perception and collaboration between large and small models to accurately identify counterfeit documents and synthetic voices, achieving a 98% accuracy rate in fake document recognition [2] - New security assessment systems from Green Alliance Technology enable automated deep scanning of over 140 mainstream models, identifying risks related to content safety, adversarial attacks, data leakage, and component vulnerabilities [2] - The "AI Era Cybersecurity Talent Development Report (2025)" indicates a projected global cybersecurity talent gap of 4.8 million by 2025, with a 19% year-on-year increase, and highlights the need for cybersecurity professionals in the U.S. and China [2]

Core Insights - The third "Tianwang Cup" Cybersecurity Competition concluded successfully on September 23 in Tianjin, showcasing the technical prowess and professional offensive and defensive capabilities of top cybersecurity teams from across the country [1][3] Group 1: Event Overview - The competition was organized by the Tianjin Municipal Government and supported by various governmental and technological institutions, highlighting its significance in the domestic cybersecurity landscape [3][4] - The event focused on key areas such as digital security, artificial intelligence security, and vehicle networking security, aiming to establish a comprehensive technical offensive and defensive system [3][4] Group 2: Participation and Results - A total of 132 teams with 530 participants passed the qualification review, with 42 top teams advancing to the finals after rigorous selection [4] - The competition awarded 4 first prizes, 7 second prizes, and 11 third prizes, reflecting the high level of expertise displayed by the participating teams [1][4] Group 3: Industry Implications - The event aims to strengthen the security framework essential for the development of the digital economy, particularly as new technologies like artificial intelligence and smart vehicles emerge [4] - The "Tianwang Cup" is positioned to foster collaboration among government, industry, academia, and research institutions, promoting technological transformation and talent cultivation in the cybersecurity sector [4]

Core Viewpoint - The 2025 National Cybersecurity Awareness Week in Henan Province aims to enhance public awareness and skills regarding cybersecurity through various activities and collaborations, emphasizing the importance of collective responsibility in maintaining network security [1][5]. Group 1: Event Overview - The event was launched from September 15 to 21, 2025, featuring a combination of online and offline activities to promote cybersecurity knowledge and skills [1]. - A micro-drama titled "Confession of Safety Code" was performed to illustrate common cybersecurity risks such as AI face-swapping and excessive personal information collection [2]. - Various departments collaborated to create a coordinated cybersecurity framework, enhancing public trust in network security [2]. Group 2: Thematic Activities - The event included six themed days focusing on different demographics and risks, such as "Campus Day," "Telecom Day," and "Personal Information Protection Day," to integrate cybersecurity knowledge into daily life [3]. - Activities included student performances demonstrating phishing website identification and community simulations of scam calls, effectively educating participants [3]. Group 3: Community Engagement - The event served as a platform for interaction between the public and cybersecurity professionals, allowing for the identification of community safety concerns [4]. - Participants from various sectors, including students and industry representatives, expressed commitment to enhancing cybersecurity awareness and practices [4]. Group 4: Policy and Infrastructure Development - Henan Province has been advancing its cybersecurity framework, highlighted by the implementation of the "Henan Cybersecurity Regulations" and the establishment of the first national cybersecurity technology museum [5]. - The ongoing efforts aim to improve public capability in addressing common cybersecurity risks and foster a strong societal commitment to network security [5].

Core Insights - The fifth "Great Wall Cup" Cybersecurity Competition, focusing on artificial intelligence security, was successfully concluded, organized by internet information offices and education departments from Beijing, Tianjin, Hebei, and Inner Mongolia [1] - The competition aimed to enhance practical skills in areas such as large model data security, information system security, adversarial sample attacks, deep forgery, and content security, promoting innovative and secure applications of AI across various sectors [1] Group 1: Competition Overview - The competition featured two categories: university and social groups, attracting 2040 teams and 5229 participants from 31 provinces and cities, with a doubling in the number of registered teams and participants compared to the previous year [1] - The finals saw 70 university teams and 30 social teams compete in Beijing, culminating in a six-hour contest [1] Group 2: Winners and Recognition - The Inner Mongolia University team and the Zhengzhou University team won the first prize in the university category, while the State Grid Jibei Electric Power Company team secured the top spot in the social category [1]

Core Viewpoint - The China-ASEAN Artificial Intelligence Security Forum emphasizes the importance of AI in driving transformation across various sectors while addressing the associated security challenges [1][2] Group 1: Event Overview - The forum was held on September 17 in Nanning, featuring speeches from key figures including Chen Gang and Qi Xiangdong, and the establishment of the Guangxi Artificial Intelligence Security Research Institute [1] - A strategic cooperation agreement was signed between the Nanning government and Qi Anxin Group to enhance AI security collaboration [1] Group 2: AI Security Focus - AI is recognized as a double-edged sword that can drive economic and social development but also presents new challenges [1] - The focus on AI security is aligned with national security interests, public welfare, and individual privacy needs, presenting significant opportunities for companies in the AI security sector [1] Group 3: Strategic Directions - Qi Anxin Group aims to address AI security issues by ensuring safety across four key areas: framework, data, personnel, and supply chain [2] - The collaboration with Guangxi will focus on three main areas: innovation in security technology for AI scenarios, promoting mutual recognition of AI security standards, and creating a platform for talent exchange and training in AI security [2] Group 4: Collaborative Efforts - The forum included discussions among experts on AI governance between China and ASEAN, fostering a consensus on cooperation in AI security [2]

Core Points - The "Artificial Intelligence Security Industry Self-Discipline Initiative" was jointly released by the China Cybersecurity Association and over 60 enterprises and research institutions, marking a significant industry consensus in the AI field and a shift from "regulation" to "self-discipline" [1] - The initiative emphasizes that security is the "lifeline" of AI development and calls for a collaborative effort to build a "controllable, trustworthy, and reliable" AI ecosystem, covering seven key areas including shared responsibility, integration of technology and management, data compliance, ethical standards, and innovative cooperation [1] - Major tech companies such as Alibaba, Baidu, Huawei, and others participated in the initiative, which stresses the importance of implementing security responsibilities throughout the entire lifecycle of AI development, particularly in avoiding algorithmic bias, preventing data misuse, and ensuring user privacy [1] - The initiative serves as both an industry commitment and a practical action guide, proposing the establishment of comprehensive lifecycle technology security standards and promoting transparency in content labeling and enhanced detection and evaluation [1] Industry Context - The rapid integration of AI technology into daily life highlights the critical need for industry self-discipline mechanisms, as AI applications span from smart voice assistants to autonomous driving and medical diagnostics, raising increasing concerns about safety and ethics [2] - The release of this initiative is a proactive response from the industry to public concerns and aims to safeguard the healthy development of AI in the future [2]

来源：环球网 《倡议》提出，要在人工智能的研发、提供、使用等全链条中强化安全理念，建立健全安全责任体系； 要强化技管集合，共建安全能力；要深化协同共治，共建风险治理能力；要推动形成透明、负责任的行 业生态；要践行智能向善，推动建立行业自律机制；要加强技术创新与治理创新，在数据安全、伦理规 范等领域积极布局；要扩展全球视野，共促开放合作，携手全球伙伴共同促进技术发展，协同应对风险 挑战，助力构建人类命运共同体。 《倡议》由中国网络空间安全协会会同行业企业共同发起，涵盖科研院所、互联网头部企业、网络安全 厂商、人工智能企业等60余家单位，聚焦人工智能发展面临的突出安全风险和治理挑战，形成安全共 识，为进一步推动人工智能产业在创新和安全中实现良性互动，为人工智能健康发展提供有力支撑。 展望未来，轻松健康集团将以此次《倡议》发布为契机，进一步深化在人工智能安全治理方面的探索与 实践。也将携手行业伙伴，共同推动AI治理标准化与体系化建设，探索"安全+健康"的新路径，助力人 工智能产业实现高质量、可持续发展。 【环球网财经综合报道】9月15日，2025年国家网络安全宣传周开幕式在云南省昆明市举行。本届网安 周主题为"网络 ...

Core Insights - Irregular, an AI security company, has raised $80 million in a new funding round led by Sequoia Capital and Redpoint Ventures, bringing its valuation to $450 million [1] Group 1: Company Overview - Irregular, formerly known as Pattern Labs, is a significant player in the AI assessment field, with its research cited in major AI models like Claude 3.7 Sonnet and OpenAI's o3 and o4-mini [2] - The company has developed the SOLVE framework for assessing model vulnerability detection capabilities, which is widely used in the industry [3] Group 2: Funding and Future Goals - The recent funding aims to address broader goals, focusing on the early detection of new risks and behaviors before they manifest [3] - Irregular has created a sophisticated simulation environment to conduct high-intensity testing on models before their release [3] Group 3: Security Focus - The company has established complex network simulation environments where AI acts as both attacker and defender, allowing for clear identification of effective defense points and weaknesses when new models are launched [4] - The AI industry is increasingly prioritizing security, especially as risks from advanced models become more apparent [4][5] Group 4: Challenges Ahead - The founders of Irregular view the growing capabilities of large language models as just the beginning of numerous security challenges [6] - The mission of Irregular is to safeguard these increasingly complex models, acknowledging the extensive work that lies ahead [6]

9月17日，第22届中国—东盟博览会暨中国—东盟商务与投资峰会在广西南宁市隆重召开。在本次博览 会同期举办的中国—东盟人工智能安全前沿论坛上，广西人工智能安全研究院暨奇安信中国—东盟人工 智能安全研究院首次亮相，同时奇安信大模型安全护栏新品正式对外发布。本次亮相的人工智能安全研 究院聚焦三个方向：一是立足人工智能场景开展"数网智"安全技术创新；二是推动人工智能安全规范互 认互通；三是打造人工智能安全人才交流与培育平台，最终形成我国面向海外开展网络安全业务的龙头 品牌。而大模型安全护栏以"守护智能未来，构建可信AI"为核心理念，无需客户改造大模型，即可提供 全链路安全防护，从而帮助千行百业的大模型应用筑牢安全屏障，推动AI技术在安全、可控、可信、 合规的前提下释放真正价值，为中国和东盟国家的数智化发展保驾护航。（奇安信集团） ...

9月15日，2025年甘肃省网络安全宣传周网信科技主题展在兰州音乐厅广场举行。 该展览以"网络安全为人民，网络安全靠人民——以高水平安全守护高质量发展"为主题，重点展示人工智能安全、数据安全等 领域的创新成果与示范应用，同步开展知识讲座、新技术新产品发布等活动，现场还设置了互动体验项目吸引市民群众参与。 当日，网络安全产品和服务供需洽谈会、互联网人才招聘会等活动同步举办。 人民网记者 周婉婷摄影报道 责编：张青津、姚凯红 ...