As Artificial Intelligence (AI) becomes increasingly integral to our daily lives, it is crucial that AI systems are designed to offer a helpful, safe, and trustworthy experience for all.
This is why Microsoft employs Responsible AI practices in developing and deploying technology. Responsible AI aims to prioritize people in our design process, considering both the benefits and potential harms that AI systems may have on society.
Our work is guided by six core Responsible AI principles, and we take a cross-company approach through cutting-edge research, top-tier engineering systems, and excellence in policy and governance.
Responsible AI Principles
These six principles lay the foundation for all of our AI efforts across the company:
-
Fairness – Microsoft aims to design AI systems that ensure quality of service, resource availability, and minimize potential stereotyping based on demographics, culture, or other factors.
-
Reliability and safety – Microsoft develops AI systems in alignment with their design ideas, values, and principles to avoid creating harm.
-
Privacy and security – With an increased reliance on data to develop and train AI systems, we've established requirements to ensure that data is not leaked or disclosed.
-
Inclusiveness – Microsoft's goal is for AI systems to empower and engage global communities. Collaborations with under-served minority communities help in planning, testing, and building these systems.
-
Transparency – Creators of AI systems should be open about the usage and limitations of AI, ensuring that everyone understands AI behavior.
-
Accountability – Accountability for technology impact is emphasized. For Microsoft, this involves consistently applying our principles in all activities.
Microsoft's Responsible AI standard
Microsoft's Responsible AI Standard comprises company-wide rules to ensure AI technologies are developed and deployed in line with Microsoft's AI principles.
Microsoft integrates strong internal governance practices across the company, updating our Responsible AI Standard to improve its concreteness, actionability, and integration into existing engineering practices.
A thoughtful, cross-discipline approach consults experts within and beyond Microsoft to ensure inclusiveness and forward-thinking. This Responsible AI Standard serves as a framework for the practice of responsible AI and evolving regulatory requirements.
Use of data
Our approach to privacy and data protection is grounded in our belief that customers own their own data and ensuring any product or service we provide is built with privacy by design from the ground up. We've defined clear privacy principles that include a commitment to be transparent in our privacy practices, to offer meaningful privacy choices, and to always responsibly manage the data we store and process.
To learn more, see responsible AI in action at Microsoft.
