通信,拉近世界的距离!
您的位置:首页 >资讯 >业界新闻 >查看资讯

IEEE庆祝AI伦理工作十周年:从EAD报告到29项标准体系

移动通信网 程然

人工智能的快速演进让偏见、数据隐私、虚假信息乃至失控风险成为全球性议题,但鲜为人知的是,IEEE 标准协会在这些问题进入公众视野之前,就已经开始了系统性的应对。2016 年,IEEE SA 发起“自主与智能系统伦理考量全球倡议”,并发布《伦理对齐设计》报告。今年恰逢这份报告首份草案诞生十周年,IEEE SA 近期上线了专题网站、播客系列与博客,回顾这一工作在全球负责任 AI 进程中的定位与影响。

从伦理原则到设计实践:EAD 的开创性定位

EAD 的使命是推动所有参与自主与智能系统设计的从业者,将人类福祉置于优先位置。报告由全球多学科专家社区共同开发,提出了八项基础性原则,涵盖尊重人权、将福祉提升作为开发成功标准、赋予个体数据自主权、有效性、透明性、问责制、滥用风险意识以及系统创建者的胜任能力。值得注意的是,报告刻意区分了“伦理”与“道德”的边界,在工程实践中将古典伦理传统与基于价值的设计方法论结合,为开发 AI 工具的工程师提供可操作的建议。

EAD 首创性地将应用伦理前置到 AI 系统的设计阶段,而非产品发布后再做补救。这一思路在当时几乎没有先例。该倡议创始执行总监、现任 IEEE Planet Positive 2030 全球总监 John C. Havens 评价称,EAD 是第一份超越“原则清单”形式的报告,且出自 IEEE 这样具有公信力的组织,“在发布之时,业界没有任何类似的东西”。超过 100 位领先 AI 伦理专家参与了首版撰写,2017 年发布的第二版与首版均公开征集意见,累计收到数百页公众反馈。到 2019 年,贡献者已超过 400 人,社区规模仍在扩张。

【关键数据】

- 倡议启动时间: 2016 年

- 基础性原则数量: 8 项

- IEEE 7000 系列标准: 已立项 29 项,已发布 11 项(且仍在持续扩充)

- 专家贡献者规模: 截至 2019 年超过 400 人

- 公开反馈体量: 两版报告累计收到数百页公众意见

IEEE AI伦理十周年

标准体系与认证矩阵的成型

在 EAD 框架基础上,IEEE 陆续发展出 IEEE 7000 系列标准。目前该系列共 29 个项目,其中 11 项已正式发布,覆盖范围相当具体:IEEE 7000-2021 规范伦理系统设计模型流程,IEEE 7001-2021 针对系统透明性,IEEE 7002-2022 处理数据隐私,IEEE 7010-2020 评估对人类福祉的影响,IEEE 7012-2025 定义机器可读的个人隐私要求,IEEE 7014-2024 则触及“模拟人类共情”这类前沿议题。这套标准的共同特征,是把抽象的伦理要求转化为工程环节可执行的流程与指标。

标准之外,IEEE SA 还构建了配套的认证体系。CertifAIEd 项目提供面向个人和产品的两类认证:前者培养评估自主与智能系统是否符合 IEEE 方法论与伦理标准的专业人员;后者由认证评估员对组织的 AI 产品进行审核,通过 IEEE 符合性评估程序后颁发认证标志。此外,面向儿童 AI 产品的“负责任 AI for Children”认证也已推出,通过对儿童类 AI 产品进行独立测试与验证,确认其满足设计与伦理要求。

伦理基础设施的十年启示

在组织层面,IEEE 新近成立了“面向全球繁荣的 AI 系统”产业协作组织,为行业提供践行负责任 AI 所需的资源、知识与工具,这标志着其工作重心正从标准制定向生态赋能延伸。纵观十年历程,IEEE 的路径清晰可辨:以 EAD 报告确立价值框架,以 7000 系列标准落地为工程规范,再以认证体系形成市场约束,最终通过产业协作推广普及。

当前全球 AI 治理进入规则密集期,欧盟 AI 法案等监管框架相继生效,各国标准组织也在加速布局。IEEE 十年积累的意义在于证明了一条“软标准”路径的可行性——不依赖强制立法,而是通过设计前置、标准牵引与认证背书,让伦理要求内化为产品开发流程的一部分。对国内产业链而言,这套方法论在出海合规与国内 AI 治理框架对接上都有直接参考价值,值得持续追踪。


出处:IEEE Celebrates 10 Years of Incorporating Ethics Into AI Systems

英文原文
Policymakers, technologists, and members of the public have raised concerns about the harm AI’s rapid advancement might cause to society. Their concerns include bias, data privacy violations, the spread of misinformation, and losing control of AI agents. Those issues are relatively new to many people, but the IEEE Standards Association has been working for the past decade on ways to prioritize the responsible development of AI systems. In 2016 IEEE SA launched the IEEE Global Initiative for Ethical Considerations in Autonomous and Intelligent Systems (A/IS) and published the Ethically Aligned Design report. EAD ’s mission is to encourage everyone involved in designing autonomous and intelligent systems to prioritize human well-being by applying ethical considerations for the benefit of humanity. Developed by a multidisciplinary community of experts from around the world, the report outlined eight foundational principles for A/IS. They include human rights, people’s well-being, and transparency. More than 100 leading AI ethics experts provided input. EAD ’s Version II report was released in 2017. Both versions solicited public input and received hundreds of pages of feedback, according to John C. Havens , who was the global initiative’s founding executive director, and is now the global director for IEEE Planet Positive 2030 . And by 2019, there were more than 400 contributors and the community continues to grow. “ Ethically Aligned Design was the first of its kind in the sense of a report that was more than a list of principles and from an organization with the credibility of IEEE,” Havens says. “At the time it was released, there wasn’t anything like it.” Since its launch, a series of standards has been developed to provide practical guidelines on designing, building, and deploying technology that prioritizes human values. Several AI ethics certification programs have been introduced. This year marks the 10th anniversary of EAD ’s first draft. To celebrate, IEEE SA recently launched a website , a podcast series , and a blog that reflect on the report’s achievements and impact. Eight guiding principles The word ethics is often used interchangeably with morality , but that is not the case in responsible AI systems design, according to EAD ’s first version. The report includes insights from classical ethics traditions and value-based design methodologies to offer actionable recommendations for engineers developing AI tools. The report established eight foundational general principles for designing autonomous systems: respecting human rights, prioritizing increased well-being as success criteria for development, empowering individuals with data agency to have control over their identity, effectiveness, transparency, accountability, awareness of misuse, and competence for the system creators. The principles shift applied ethics to the design stage of AI systems, compared with being a concern after they’re released, the authors wrote. A suite of standards The IEEE 7000 series of standards has been developed based on EAD ’s framework. Currently the series consists of 29 projects, 11 of which have been published, and more are still being added. They cover topics such as ethical system design model process ( IEEE 7000-2021 ), AI system transparency ( IEEE 7001-2021 ), data privacy ( IEEE 7002-2022 ), algorithmic bias considerations ( IEEE 7003 -2024), the impact on people’s well-being ( IEEE 7010-2020 ), machine-readable personal privacy requirements ( IEEE 7012-2025 ), and systems that mimic human empathy ( IEEE 7014-2024 ). AI ethics certifications IEEE SA offers a suite of ethics in AI programs including IEEE CertifAIEd , which includes two certifications : one for individuals and one for products. The professional certification provides individuals with skills to assess autonomous and intelligent systems for adherence to IEEE’s methodology and ethics criteria. “ Ethically Aligned Design was the first of its kind in the sense of a report that was more than a list of principles and from an organization with the credibility of IEEE.” —John C. Havens, founding executive director of the global initiative The product certification program assesses whether an organization’s AI tool or system conforms to the criteria in the IEEE Ontological Specifications. A CertifiAIEd assessor evaluates the product to ensure it meets the criteria. Upon completion of the assessment, the owner submits it to the IEEE Conformity Assessment Program , which can certify it and issue the certification mark. IEEE SA also recently released its Responsible AI for Children certification program . This program certifies that AI products for children have been independently tested and verified to meet design and ethics requirements. New industry connections group The IEEE Artificial Intelligence Systems (AIS) for Global Flourishing Collaborative Industry Collections group was formed to provide the resources, knowledge, and tools needed to prioriti
微信扫描分享本文到朋友圈
扫码关注5G通信官方公众号,免费领取以下5G精品资料
5G通信公众号
  • 1、回复“YD5GAI”免费领取《中国移动:5G网络AI应用典型场景技术解决方案白皮书》
  • 2、回复“5G6G”免费领取《5G_6G毫米波测试技术白皮书-2022_03-21》
  • 3、回复“YD6G”免费领取《中国移动:6G至简无线接入网白皮书》
  • 4、回复“LTBPS”免费领取《《中国联通5G终端白皮书》》
  • 5、回复“ZGDX”免费领取《中国电信5GNTN技术白皮书》
  • 6、回复“TXSB”免费领取《通信设备安装工程施工工艺图解》
  • 7、回复“YDSL”免费领取《中国移动算力并网白皮书》
  • 8、回复“5GX3”免费领取《R1623501-g605G的系统架构1》
  • 本周热点本月热点
    通信人才网

    最热通信招聘

    更多