ALGORITHMIC-IMPACT-ASSESSMENT

Algorithmic Impact Assessment AI safety + governance framework.

Definition

Algorithmic Impact Assessment (AIA) is the structured assessment process for evaluating + documenting AI systems impacts + risks before deployment, foundational concept Canadian Government Directive on Automated Decision-Making 2019 (mandatory AIA government AI systems automated decisions affecting individuals or businesses) + adopted multiple jurisdictions (New Zealand + Estonia + multiple ; EU AI Act Article 27 'Fundamental Rights Impact Assessment' FRIA equivalent), evaluates fairness + bias + privacy + cybersecurity + societal impacts + mitigation measures. Framework + standard + guidance objectives AI safety + governance + risk management + transparency + fairness + accountability + explainability + multiple AI ethics principles adoption multi-stakeholder process + voluntary adoption industry + emerging regulation alignment EU AI Act + US Executive Orders + multiple national jurisdictions + private sector adoption Microsoft + Google + Meta + OpenAI + Anthropic + other major AI labs + corporate compliance teams 2020s+.

Origin

Canadian Government Directive on Automated Decision-Making 2019 mandates AIA ; EU AI Act FRIA Article 27 equivalent 2027+ ; multiple jurisdictions adoption.

Example in context

Canadian federal agency considers deploying AI system to triage citizen service applications: per Directive on Automated Decision-Making 2019 mandatory conducts Algorithmic Impact Assessment AIA, evaluates system across ~60 questions covering 6 areas (impact level + risk + design + project + algorithm characteristics + data), assigns Impact Level (Level I-IV), prescribed mitigation measures + transparency + human oversight obligations per AIA conclusions before deployment authorization.

Related terms

• AI Audit Framework — complementary AI assessment.