人工智能应用风险评估工具是北京人工智能安全与治理实验室研制的人工智能风险、伦理与治理在线检测、评估与服务的工具。工具依据世界各国政府间组织、政府、科研机构、企业发布的人工智能伦理与治理原则提案，从合法合规、对环境与社会有益、以人为本、公平公正、问责、透明性、可解释性、隐私与数据保护、安全性、可控性、可靠性等方面对人工智能产品与服务进行评估，并对其中可能存在的风险隐患给出解释、相关案例与治理建议。从而为人工智能企业、科研机构开展自律自治、政府开展人工智能伦理与治理监管与服务提供基础设施。

Artificial Intelligence is considered to be a disruptive science and technology that are changing our society. We should extend its positive and good use, and at the same time, avoid misuse, abuse, as well as potential negative side effects. AI Risk Assessment is designed to help AI research, projects, products, and services evaluate their AI from various perspectives for potential technical, ethical and social risks, and provide automated reports on the evaluation and status of their AI from the AI governance perspective. Relevant cases and suggestions will also be provided to help to make the right decisions. AI should be beneficial not only to human, but also to environment, ecology, and the future of human-AI symbiotic society.