Core Viewpoint - The rise of generative artificial intelligence is reshaping the information ecosystem, leading to the proliferation of low-quality content, termed "digital sludge," which poses a significant risk to the cognitive development and value formation of minors in China [1][2]. Group 1: Impact of "Digital Sludge" - "Digital sludge" erodes deep cognitive abilities, as exposure to fragmented and low-logic information inhibits the development of critical thinking skills in adolescents [2][3]. - It distorts values and social cognition, creating a false environment that interferes with minors' ability to discern right from wrong and build healthy value systems [2][3]. - The phenomenon exacerbates the deterioration of the digital cultural ecosystem, as low-quality content crowds out quality content, leading to a vicious cycle that undermines cultural creativity and aesthetic standards [2][3]. Group 2: Factors Contributing to the Proliferation of "Digital Sludge" - The combination of technological advancements and market logic has lowered the barriers for producing low-quality content, with platforms prioritizing user engagement through algorithms that favor emotionally charged content [3]. - The traditional filtering mechanisms of professional "gatekeepers" in media have been significantly weakened in the algorithm-driven environment, allowing low-quality content to thrive [3]. Group 3: Governance Strategies - Establish a comprehensive governance system that integrates technology, legal frameworks, education, and social forces to address the challenges posed by "digital sludge" [5]. - Implement "algorithm for good" principles to ensure platforms take responsibility for content distribution, shifting from a "traffic-first" to a "value-first" approach [5][6]. - Create a collaborative regulatory framework involving multiple departments to enhance the precision and agility of governance, focusing on proactive measures rather than reactive responses [6]. Group 4: Educational Initiatives - Enhance the digital literacy and critical thinking education for minors by integrating systematic media literacy into school curricula, focusing on algorithm awareness and information verification [7]. - Empower parents through digital literacy support programs to help them guide their children effectively in navigating the digital landscape [7]. - Foster a cultural atmosphere that values deep thinking and rational dialogue through community engagement and experiential educational projects [7].

Core Viewpoint - The rapid growth of AI-generated content poses significant challenges for minors, who are increasingly exposed to low-quality and misleading information, necessitating a comprehensive approach to enhance their media literacy and critical thinking skills [2][3][5]. Group 1: Current Situation - The number of underage internet users in China has surpassed 196 million, highlighting the urgency of addressing the issue of "digital waste" [2]. - The prevalence of AI-generated fake news and low-quality content is impacting the cognitive development and value formation of young people [3][4]. Group 2: Recommendations for Improvement - A collaborative system involving higher education institutions, primary and secondary schools, media, and enterprises is recommended to improve minors' ability to discern and critique AI-generated content [2][3]. - Schools should integrate media literacy education into daily teaching, breaking the limitations of single subjects to create a cross-disciplinary training system [3][4]. Group 3: Role of Parents and Reading - Parents need to enhance their own media literacy to effectively guide their children in navigating complex online content and recognizing quality information [5]. - Emphasizing the importance of reading classic and quality literature can serve as a protective measure against the influx of low-quality digital content, fostering a more robust cognitive framework for children [5].

Research Background and Significance - The rapid flow of information has led to unprecedented speed and influence of rumor dissemination, necessitating a stable and professional fact-checking capability to maintain a healthy information ecosystem and protect user rights [7][9]. Methodology and Data Sources - The report employs quantitative statistics, trend comparisons, questionnaire surveys, and case studies to analyze the evolution of rumors from 2015 to 2025, focusing on total volume, type distribution, and dissemination patterns [10][11]. Ten Core Findings 1. The rumor ecosystem has evolved through three phases: 2015-2019 focused on health and food safety rumors; 2020-2022 centered on major public events; and 2023-2025 characterized by AI integration and content diversification [13]. 2. The dissemination of rumors has undergone a "channel revolution," transitioning from text-based to video and algorithm-driven platforms, significantly altering the spread and impact of misinformation [13]. 3. AI technology has lowered the barriers for producing false information, leading to a "probability truth" challenge where the focus shifts from content authenticity to source credibility [13]. 4. Public trust and recognition have deteriorated as highly realistic content blurs the lines of truth, increasing anxiety and apathy towards the concept of "truth" [13]. 5. A significant portion of the public exhibits cognitive closure, with 93% of respondents preferring clear answers, making them susceptible to emotionally driven misinformation [14]. 6. The public's ability to identify health-related rumors is notably weak, with 55.63% of respondents scoring low due to the complexity and misleading nature of health misinformation [14]. 7. Fact-checking methods have evolved through three stages: manual verification, algorithmic identification, and AI-driven real-time verification, enhancing efficiency by up to 90% [14]. 8. The governance model has shifted from independent platform efforts to a collaborative ecosystem involving multiple stakeholders, including authoritative institutions [14]. 9. The effectiveness of fact-checking has been quantitatively assessed, showing a 36.5% reduction in public panic during the pandemic due to effective rumor management [14]. 10. "Pre-fact-checking" strategies have proven effective, particularly for policy-related rumors, with 92% of respondents showing a habit of verifying through official channels [15]. Evolution of Rumors Over the Past Decade - The chapter outlines the significant changes in the forms, media, and social impacts of rumors over the past decade, reflecting technological advancements and societal developments [19]. Phases of Rumor Evolution - The rumor landscape is divided into three phases: - 2015-2019 saw a rise in everyday life rumors amid a lack of scientific literacy [22]. - 2020-2022 shifted focus to major public health events, with video content becoming the dominant form of rumor [26]. - 2023-2025 is marked by AI's role in content creation, complicating the identification of misinformation [28]. Lifecycle of Rumors - The lifecycle of rumors includes four stages: emergence, diffusion, peak, and decline, with some rumors re-emerging periodically [37]. Types of Rumors - "Evergreen" rumors maintain relevance due to their broad applicability, while "flash-in-the-pan" rumors arise from specific events and have shorter lifespans [39][40]. Upgrading of Rumor Dissemination Methods and Public Perception Changes - The evolution of media technology has restructured rumor dissemination, transitioning from text-based platforms to visual and algorithm-driven short video platforms, significantly altering public attitudes and behaviors towards misinformation [45]. Channels of Rumor Dissemination - The transition from traditional text-based platforms to visual content and algorithm-driven platforms has redefined the mechanisms of rumor spread [46]. Public Attitudes and Behavioral Changes - Public attitudes towards rumors have shifted significantly over the past decade, influenced by life anxieties and the trust dynamics within social networks [58].