Key Considerations for Responsible Development and Fielding of Artificial Intelligence

The “Key Considerations”  stem from the National Security Commission on Artificial Intelligence’s (NSCAI) line of effort dedicated to Ethics and Responsible Artificial Intelligence (AI). The recommended considerations and practices found in this framework provided the foundation for integrating responsible AI throughout the Commission’s recommendations in its final report delivered to Congress and the Executive Branch on March 1, 2021. 

The Key Considerations include near-term recommendations and topics that agencies should give priority consideration, practices that should be implemented, policies that should be defined or updated to reflect new AI considerations, and future work that should be pursued – from R&D to international cooperation. These recommendations, which operationalize AI principles adopted by all government agencies, span 5 areas: 

  • Aligning Systems and Uses with American Values and the Rule of Law
  • Engineering Practices
  • System Performance
  • Human-AI Interaction & Teaming
  • Accountability and Governance.

The Key Considerations was intended for U.S. government departments and agencies critical to national security in the context of the NSCAI’s mandate and final report. However, the Key Considerations are relevant for any entity designing, building or using AI technologies, including industry, academic, and civil society organizations. They also inform opportunities for international collaboration and cooperation.

Abridged Key Considerations

Previous Versions

The Key Considerations was originally published in July 2020. After integrating feedback following public and interagency consultations, a revised version was provided as a supplement to the Commission’s March 1, 2021, Final Report.   The Commission received and incorporated feedback on its recommendations, including the Key Considerations, through public plenary sessions, and discussions with leading academics and representatives from civil society. These representatives include the Association for the Advancement of Artificial Intelligence (AAAI) and the Partnership on AI (PAI). We thank the Montreal AI Ethics Institute for hosting a full-day discussion and feedback session in September 2020 with members of the technical community and civil society in the US and abroad. We also thank technical and policy experts in the US Government who provided feedback on areas needing additional R&D. Input was provided by representatives from the Department of Defense (DoD), Intelligence Community (IC), Department of Homeland Security (DHS), Federal Bureau of Investigation (FBI), Department of Energy (DoE), Department of State (DoS), and Department of Health and Human Services (HHS), as well as the GSA AI Community of Practice. The NSCAI archive hosts the prior full and abridged versions of the Key Considerations document.
 

To learn more about the Key Considerations, listen to this podcast from the Chair of NSCAI’s Ethics and Responsible AI Line of Effort.