移动通信网 程然
英国通信管理局Ofcom日前对数字平台祭出迄今最严格的在线安全规则,要求网站及主要社交媒体服务部署检测技术,主动识别并移除儿童性虐待材料。新规预计于2027年初正式生效,标志着英国《在线安全法》执法框架从“被动响应举报”向“主动技术筛查”的重大转变。Ofcom在线安全事务主管Oliver Griffiths直言:“我们对当前平台所做的努力并不满意”,并警告不合规企业将面临“严重后果”。
检测义务全面升级:从举报驱动到主动筛查
根据新规要求,更多数字服务——包括此前监管覆盖不足的色情网站——将强制采用哈希匹配技术,通过比对图像“数字指纹”与已知违法内容数据库来识别非法图像。这一技术已在主流平台中广泛应用,但此次英国首次将其确立为法律层面的强制性义务,覆盖范围显著扩大。
更值得关注的是,平台还需部署自动化工具来检测此前未被识别的儿童虐待内容和恐怖主义内容,彻底摆脱对用户举报的依赖。这一要求实际上将内容安全的举证责任从监管部门转移至平台方——平台不能再以“未收到举报”为由推卸责任,而必须证明自身具备主动发现违法内容的技术能力。此外,新规还要求平台封禁制作、上传或传播儿童性虐待内容的用户,并采取技术手段阻止其重新注册。
【关键数据】
- 新规生效时间: 2027 年初
- 日均犯罪量: 英国每天平均发生约 100 起儿童性虐待图像犯罪
- 青少年禁令: 16 岁以下用户禁用社交媒体,2027 年春季生效
- 覆盖平台: Snapchat、TikTok、Instagram、YouTube、Facebook、X 等
![]()
AI 成为治理焦点,违法内容治理压力加剧
Ofcom在声明中特别指出,在线违法内容数量持续攀升,而生成式AI技术“使这一问题显著恶化”。深度伪造技术使得制作逼真的儿童性虐待图像变得前所未有的容易,传统基于已知内容数据库的哈希匹配手段在面对海量新生成内容时显得力不从心——这正是新规要求平台部署“检测未识别内容”自动化工具的直接动因。
英国防止虐待儿童协会NSPCC援引警方数据指出,英国每天平均发生约100起儿童性虐待图像犯罪,这一数字为监管加码提供了强有力的民意和立法支撑。英国在线安全事务大臣Ian Murray表示:“我们正在快速行动,确保法律跟上新技术的步伐。”从立法节奏看,英国正试图在AI生成内容泛滥之前完成监管框架的全面落地。
监管组合拳:16岁以下社交禁令同步推进
此次检测义务新规并非孤立动作。英国政府今年早些时候已公布计划,将禁止16岁以下用户访问Snapchat、TikTok、Instagram、YouTube、Facebook和X等社交平台,该限制预计于2027年春季生效,同时覆盖直播服务和AI聊天机器人。Ofcom明确表态:“这项禁令将在保护英国儿童免受虐待和剥削方面发挥重要作用,我们随时准备在生效后立即执法。”
需要关注的是,Ofcom同时宣布将不再推进关于儿童直播的单独监管提案,理由是即将实施的年龄验证要求已能覆盖相关风险。这一取舍显示出监管机构在规则设计上正寻求精简路径,避免多重复核造成的执行混乱。从全球视角看,英国这一系列动作可能为欧盟《数字服务法》乃至其他司法辖区的平台内容治理提供参照样本——平台主动检测义务是否会成为国际通行标准,值得持续追踪。
出处:Ofcom targets platforms over child abuse content
英文原文UK watchdog Ofcom tightened online safety rules for digital platforms, requiring websites and major social media services to deploy detection technology to identify and remove child sexual abuse material. Expected to take effect in early 2027, the new rules will require more digital services, including pornography sites, to use hash matching to identify known illegal images by comparing digital fingerprints against databases. Platforms will also need to implement automated tools to detect previously unidentified child abuse and terrorist content, moving beyond systems that rely on material being reported by users. Companies must also ban users who generate, upload or share child sexual abuse content and prevent them from returning to their services. The requirements build on existing obligations under the UK’s Online Safety Act . Ofcom stated the growing volume of illegal material online, “made significantly worse by AI,” meant existing measures were insufficient. The National Society for the Prevention of Cruelty to Children (NSPCC) pointed to police data showing an average of 100 child sexual abuse image crimes take place across the country every day. “We are not satisfied that enough is being done,” said Oliver Griffiths, director of Ofcom’s Online Safety Group, warning companies that fail to comply should expect “serious consequences”. UK online safety minister Ian Murray added: “We are moving fast to make sure the law keeps pace with new technology.” Earlier this year, the UK government outlined plans to ban users under the age of 16 from accessing social media platforms including Snapchat, TikTok, Instagram, YouTube, Facebook and X. Set to come into force in spring 2027, the restrictions will also apply to livestreaming services and AI chatbots. Ofcom clarified it would not pursue separate proposals on children livestreaming, citing the forthcoming age-verification requirements. “The ban will play an important role in helping keep children in the UK safe, including from abuse and exploitation, and we stand ready to enforce it as soon as it takes effect,” the regulator added. The post Ofcom targets platforms over child abuse content appeared first on Mobile World Live .