The Illinois Department of Human Rights has proposed regulations requiring employers to notify employees and job applicants whenever artificial intelligence is used in hiring, promotions, discipline, performance monitoring, and other employment decisions. The proposed rules, published in the Illinois Register on May 15, 2026, flesh out the notice mandate in HB 3773, the AI amendment to the Illinois Human Rights Act that took effect January 1, 2026.

Illinois joins a growing group of states, including Colorado, Connecticut, and California, that are converging on transparency and disclosure as the primary regulatory tool for AI in employment, rather than the bias audit mandates that proved politically and legally difficult to implement.

What the Proposed Rules Require

Scope. Notice is required whenever an employer “uses” AI to influence or facilitate a “covered employment decision,” which is defined to include recruitment, hiring, promotion, renewal of employment, selection for training or apprenticeship, discharge, discipline, tenure, or the terms, privileges, or conditions of employment. Notice is required regardless of whether the AI use has a discriminatory purpose or effect.

The rules provide concrete examples of AI uses that trigger notice: resume screening, computer-based assessments and personality tests, targeted job advertisements, facial expression or voice analysis in interviews, third-party data analysis, AI-driven productivity monitoring and scoring, and automated work-task assignment that affects compensation.

Notably, notice is not required when AI is used for general business purposes (e.g., drafting a job posting or generating promotional images) rather than to influence an employment decision, or when a system has AI features but the employer does not use them for employment decisions.

How and when to give notice. Employers must provide notice through all applicable channels: employee handbooks, physical workplace postings, intranet and external websites (with a conspicuous homepage link), and in every job posting. Current employees and bargaining representatives must receive notice annually and within 30 days of adopting any new or substantially updated AI system.  Prospective employees must be notified in the job posting itself.

Required Content

• The AI system’s developer, product name, and vendor;
• Which employment decisions the AI influences (g., recruitment, hiring, discipline);
• The system’s purpose, including the categories of personal information or employee data it collects or processes;
• The types of job positions for which the AI tool will be used;
• A point of contact for employee questions; and
• The right to request a reasonable accommodation and instructions for doing so.

Trade secrets and information protected by law need not be disclosed.

Accessibility and recordkeeping. Notices must be in plain language, available in the languages commonly spoken by the employer’s workforce, and accessible to employees with disabilities. Employers must retain notices, postings, and records of AI use for three years.

The Broader Trend: States Choose Disclosure Over Audits

Illinois is part of a widening state-level shift toward transparency requirements and away from prescriptive bias audit mandates for AI in employment.

Colorado recently replaced its landmark 2024 algorithmic bias law (SB 205), which would have required bias audits and impact assessments, with a narrower transparency-focused law (SB 189) signed on May 14, 2026 and largely effective January 1, 2027. The revised law requires employers to give notice about AI tools used in hiring, promotions, and termination and grants workers the right to correct inaccurate data and request meaningful human review to the extent commercially reasonable. Colorado’s law extends beyond employment to cover AI used in education, finance, healthcare, housing, and insurance decisions. Enforcement rests solely with the attorney general, with no private right of action.

Connecticut enacted SB 5, which requires employers deploying automated employment-related decision technology to provide employees and applicants with written notice before the technology influences an employment decision, effective October 1, 2027. Required disclosures include the technology’s trade name, purpose, categories of personal data analyzed, and deployer contact information. Like Colorado, enforcement is through the attorney general with no private right of action, and anti-bias testing may serve as a defense to discrimination claims.

California’s civil rights regulations took effect October 1, 2025 covering automated decision systems in employment. While California stopped short of mandating bias audits, the rules make clear that the presence or absence of bias testing will be treated as evidence in discrimination claims—what some practitioners have characterized as a backdoor audit mandate.

New York City remains the outlier, requiring bias audits under Local Law 144, though it defines the covered tools narrowly.

The shift toward disclosure has been driven by a combination of tech industry advocacy, White House efforts to preempt stricter state AI laws, and legislative recognition that audit mandates face legal and practical headwinds.

Practical Takeaways

1. Inventory AI tools now. Identify every AI system used in or adjacent to employment decisions, including tools used by third-party vendors or staffing agencies, and map each to the categories of “covered employment decisions” under the proposed rules.
2. Prepare compliant notices. Draft notices that address each of the six required content elements, and plan for distribution across all required channels (handbooks, postings, websites, and job listings).
3. Engage vendors. Request from AI vendors the specific information required for notice: developer identity, product name, data categories processed, and system purpose. Review vendor contracts for information-sharing obligations.
4. Build a recordkeeping process. Establish a system for retaining AI-related notices and records of AI use for the three-year retention period.
5. Monitor the comment period. The proposed rules may change based on public comments. Employers should consider submitting comments on provisions that raise practical compliance concerns.
6. Think multi-state. Employers operating in multiple states should begin harmonizing compliance across Illinois, Colorado, Connecticut, California, and any jurisdiction with AI disclosure laws, since the notice content and timing requirements vary.

Whether or not these proposed rules are adopted in their current form, the direction is clear: states are building a compliance baseline around AI transparency in employment, and employers that invest now in notice infrastructure, vendor coordination, and multi-state tracking will be better positioned as these requirements continue to expand.

Please contact Sam Mitchell or your SGR attorney for more information.