Algorithmic discrimination protections call for identifying potential biases in automated systems so that they do not unfairly treat people. They appear as one principle in the United States Blueprint for an AI Bill of Rights, alongside requirements concerning safety, privacy, explanations, and human recourse. section 10.29
The principle remains deliberately broad: the Blueprint positions it as guidance rather than a directive, leaving practical implementation and enforcement uncertain. The National Institute of Standards and Technology’s AI Risk Management Framework reaffirmed related principles, but did not by itself specify every implementation detail. section 10.29
Policy framework
Figure 10.6 organizes algorithmic discrimination protections within five principles: safe and effective systems, discrimination protections, data privacy, notice and explanation, and human alternatives or fallback. section 10.29
