前置拦截违规素材84万+ 巨量引擎强化AI广告风险管控

Core Insights - The article discusses the measures taken by the company to enhance advertising safety through AI technology and a comprehensive governance system, highlighting the interception of over 840,000 AI-related advertising violations in Q3 of this year [1][2]. Group 1: Advertising Safety Measures - The company has established a full-process governance system to address three major risks associated with AI advertising: violation of public order, infringement, and false advertising [2]. - A multi-dimensional feedback mechanism called the Consumer Complaint Rate (CCR) has been introduced, which integrates user comments, reports, and complaints, resulting in a 40% decrease in negative feedback on advertisements this year [4]. Group 2: Technological Innovations - The company has developed a self-researched multi-modal large model that can review 90% of advertising materials within 10 minutes, improving review efficiency by 75% compared to traditional methods [7]. - The large model can automatically correct common violations in advertising materials and provide detailed rejection reasons, enhancing the overall efficiency and experience for advertisers [7]. Group 3: Targeted Services for Vulnerable Groups - The company has launched specialized service assurance projects for elderly and minor groups, providing preemptive support for users facing consumption disputes due to advertisements [9]. - The initiative has served over 3,000 individuals, achieving a user satisfaction rate of 92.5%, and includes educational efforts to reduce risks for these vulnerable groups [9]. Group 4: Collaborative Governance - Experts emphasize the need for cross-domain collaboration in AI governance, advocating for continuous regulatory enhancement and the establishment of a "co-governance alliance" involving regulators, platforms, and ethical committees [11].