Intended purpose, prospective settings in which the AI system will be deployed, the specific set or types of users along with their expectations, and impacts of system use are understood and documented. Assumptions and related limitations about AI system purpose and use are enumerated, documented and tied to TEVV considerations and system metrics.
About
It is not necessarily possible to have advanced knowledge about all potential settings in which a system will be deployed. To help delineate the bounds of acceptable deployment, context mapping may include examination of the following:
intended, prospective,and actual deployment setting.
specific set or types of users.
operator or subject expectations.
concept of operations.
intended purpose and impact of system use.
requirements for system deployment and operation.
potential negative impacts to individuals, groups, communities, organizations, and society – or context-specific impacts such as legal requirements or impacts to the environment.
unintended, downstream, or other unknown contextual factors.
Actions
Pursue AI system design purposefully, after non-AI solutions are considered.
Define and document the task, purpose, minimum functionality, and benefits of the AI system to inform considerations about whether the project is worth pursuing.
Maintain awareness of industry, technical, and applicable legal standards.
Collaboratively consider intended AI system design tasks along with unanticipated purposes.
Determine the user and organizational requirements, including business and technical requirements.
Determine and delineate the expected and acceptable AI system context of use, including:
operational environment
impacts to individuals, groups, communities, organizations, and society
user characteristics and tasks
social environment.
Track and document existing AI systems held by the organization, and those maintained or supported by third-party entities.
Gain and maintain awareness about evaluating scientific claims related to AI system performance and benefits before launching into system design.
Identify human-AI interaction and/or roles, such as whether the application will support or replace human decision making.
Plan for risks related to human-AI configurations, and document requirements, roles, and responsibilities for human oversight of deployed systems.
Transparency and Documentation
Organizations can document the following:
Which AI actors are responsible for the decisions of the AI and is this person aware of the intended uses and limitations of the analytic?
Which AI actors are responsible for maintaining, re-verifying, monitoring, and updating this AI once deployed?
Who is the person(s) accountable for the ethical considerations across the AI lifecycle?
