An AEDT audit is the statistical examination New York City's Local Law 144 requires for automated tools that screen, rank, or score job candidates. Here's what it is — and isn't.
An AEDT is a tool that uses machine learning, statistical modeling, or data analytics to substantially assist or replace discretionary decision-making in hiring or promotion. Common examples: AI resume screeners, interview-ranking systems, and skills-assessment scorers. A plain keyword filter or ATS database search is generally not an AEDT.
The audit must be independent — performed by someone with no financial interest in the tool, its vendor, or the employer (DCWP Final Rules). The auditor works from the tool's outputs on real candidate data; source code access is not required.
DCWP moved to proactive investigations in January 2026 and the first fines have been issued. Penalties run $500 for the first violation and $500–$1,500 per day after, per tool. Six months of non-compliant use across one tool is a $67,500–$270,000 exposure range.
Automated statistical analysis (72-hour turnaround) plus independent sign-off is the fastest path. Start with a self-assessment toolkit to confirm coverage, then order the analysis.