1 / 6
Modes of AI use: Automation, Agency, Augmentation
In this lesson
Modes of AI use: Automation, Agency, Augmentation
Classify real-world AI deployment scenarios into automation, agency, or augmentation modes based on the degree of human oversight, system autonomy, and documented knowledge limits,
You'll be able to
- Classify real-world AI deployment scenarios into automation, agency, or augmentation modes based on the degree of human oversight, system autonomy, and documented knowledge limits, applying the MAP function's requirements for characterizing how system output may be utilized and overseen by humans[^7].
- Evaluate the appropriateness of each mode for a given production context by assessing likelihood and magnitude of beneficial and harmful impacts on individuals, groups, and organizations, using the MAP function's framework for impact characterization based on expected use and past uses in similar contexts[^6].
- Apply NIST AI RMF GOVERN and MAP function principles to define roles and responsibilities for human-AI team configurations in each mode, documenting operator proficiency requirements and decision-making processes that counter systemic biases[^5].
- Create risk profiles for automation, agency, and augmentation deployments by mapping each mode's incident communication requirements, performance measurement needs, and feedback integration practices across the MANAGE and MEASURE functions[^1][^3][^4].