你当前正在访问 Microsoft Azure Global Edition 技术文档网站。 如果需要访问由世纪互联运营的 Microsoft Azure 中国技术文档网站，请访问 https://docs.azure.cn。
Guidance for integration and responsible use with Image Analysis
Microsoft works to help customers responsibly develop and deploy solutions that use Image Analysis. We're taking a principled approach to upholding personal agency and dignity by considering the AI systems' fairness, reliability and safety, privacy and security, inclusiveness, transparency, and human accountability. These considerations are in line with our commitment to developing Responsible AI.
General guidelines for integration and responsible use
This article discusses Image Analysis and the key considerations for making use of this technology responsibly. The following are general recommendations for the responsible deployment and use of Image Analysis. Your context might require you to prioritize and apply these recommendations according to the needs of your specific deployment scenario. But in general, the following activities will assist you:
Understand what it can do: Fully assess the potential of any AI system you are using to understand its capabilities and limitations. Understand how it will perform in your particular scenario and context, by thoroughly testing it with real-life conditions and data.
Respect an individual's right to privacy: Only collect data and information from individuals for lawful and justifiable purposes. Only use data and information that you have consent to use, and only for the purpose(s) for which consent was given.
Legal review: Obtain appropriate legal advice to review your solution, particularly if you will use it in sensitive or high-risk applications. Understand what restrictions you might need to work within, and your responsibility to resolve any issues that might come up in the future.
Human-in-the-loop: Keep a human in the loop, and include human oversight as a consistent pattern area to explore. This means ensuring constant human oversight of Image Analysis, and maintaining the role of humans in decision-making. Ensure that you can have real-time human intervention in the solution to prevent harm. This enables you to manage situations where Image Analysis does not perform as expected.
Security: Ensure your solution is secure, and that it has adequate controls to preserve the integrity of your content and prevent unauthorized access.
Have a blocklist or an allow list: Instead of enabling all tags with the Image Analysis tag feature, focus on the specific ones most appropriate for your use case.
Recommendations for preserving privacy
A successful privacy approach empowers individuals with information, and provides controls and protection to preserve their privacy.
If the service is part of a solution designed to incorporate health-related data, think carefully about whether and how to record that data. Follow applicable state and federal privacy and health regulations.
Privacy managers should carefully consider what the retention policies should be for the extracted image metadata and insights, as well as the underlying images. The retention policies should be reflective of the intended use of the applications.
Don't share any data without explicit consent from affected stakeholders or data owners, and minimize the data that is shared.