Get emailed when this bill changes status, is amended, or advances.
Disclaimer: This page provides general informational summaries only and does not constitute legal advice. AI-generated content may contain errors. Always consult a qualified attorney for guidance specific to your situation.
Read full disclaimer →
The bill aims to create a framework for independent bias audits of automated employment decision tools, focusing on fairness and accountability in hiring practices. Employers and employment agencies will be required to assess the impact of these tools on various demographic categories, ensuring compliance with bias mitigation standards.
Key Provisions
Employers must conduct a bias audit of automated employment decision tools at least once a year.
Audits must assess the impact on various demographic categories and provide recommendations.
Employers cannot use tools without a recent bias audit.
Independent auditors must be free from conflicts of interest.
Exclusions for small demographic categories must be justified in audit summaries.
Latest Legislative Action
Reported from Assembly Comm. as a Substitute, 2nd Reading
Bill Sponsors
Name
Role
District
Balvir SinghD
Sponsor
HD-007
Roll Call Votes
A ·
Assembly Science, Innovation and Technology Committee: Reported with Substitution ·
May 16, 2024
5 Yea 0 Nay 2 OtherPassed ✓
Compliance Checklist
Conduct a bias audit of automated employment decision tools Who: Employers and employment agencies Deadline: Annually, with the first audit due within one year of the law's enactment Penalty: Cannot use the automated tool if the audit is not conducted
Ensure independent auditors are free from conflicts of interest Who: Employers and employment agencies Deadline: Ongoing, with each audit Penalty: Audit results may be invalidated if conflicts are present
The bill mandates that employers and employment agencies conduct a bias audit for any automated employment decision tool they use, with audits required at least once a year. The audits must evaluate the tool's impact on various demographic categories, including race, gender, and disability, and provide actionable recommendations to mitigate identified biases. Non-compliance with the audit requirement will prevent the continued use of the automated tool. Key definitions include 'automated employment decision tool,' which encompasses any machine-based system influencing employment decisions, and 'independent auditor,' who must not have conflicts of interest. This legislation aligns with similar initiatives in other states aimed at promoting algorithmic accountability in employment practices. The bill also allows for the exclusion of categories representing less than two percent of the data from impact ratio calculations, with justification required.
We use cookies for analytics to understand how visitors use this site. We also use essential cookies for site functionality.
See our Privacy Policy for details.