quick hit
- The Illinois Department of Human Rights has released draft regulations to implement a new prohibition on AI discrimination in employment in Illinois, clarifying potential notification and record-keeping obligations for employers regarding the use of AI.
- The draft rule would broadly apply to all employers under the Illinois Anti-Discrimination Act and would require notification whenever AI is involved in a covered employment decision, regardless of whether it results in unlawful discrimination.
- Under the draft regulations, the required notice would need to include specific information such as the name of the AI product, the employment decisions it affects, its purpose, the data it collects, the types of jobs it covers, and contact details.
The draft rule, Subpart J: Use of Artificial Intelligence in Employment, implements the amendments to the Illinois Human Rights Act (IHRA) made under Illinois House Bill (HB) 3773, the 2024 Act, that prohibits the use of AI (including generative AI) in a manner that discriminates against employees on the basis of a protected characteristic, even if such discrimination is unintentional. HB 3773 goes into effect on January 1, 2026 and requires the IDHR to issue regulations implementing the substantive requirements of the Act.
The new draft rules, which were shared at a stakeholder meeting and have not yet been formally released for public comment, will clarify notification and record-keeping requirements for employers regarding the use of AI to make employment decisions.
Target entity
The draft regulations broadly apply to “employers” under the IHRA and their agents, including recruiters and other third parties acting on behalf of employers. This notification requirement would be triggered if an employer “uses” AI to “influence or facilitate” “covered employment decisions,” including hiring, hiring, promotion, renewal, selection of training or apprenticeships, termination, discipline, tenure, or terms, privileges, or conditions of employment.
The draft rule broadly defines the “use” of AI to include “any instance in which the output of an artificial intelligence system influences or facilitates a covered employment decision.” Covered AI is defined in HB 3773 and refers to “a machine-based system that infers how to produce an output, such as a prediction, recommendation, or decision, from received input for an explicit or implicit purpose.”
Notification requirements
The draft regulations would require employers to notify employees and prospective employees when AI is used to “influence or facilitate” employment decisions, regardless of whether the use of AI has the “purpose or effect of exposing the employee to unlawful discrimination.”
The draft rule provides that AI that could trigger employer notification requirements sets out a wide range of situations in which it is commonly used, including the “use of computer-based assessments and tests,” such as puzzles and games, to make predictions about employees or prospective employees, or to measure the mental or physical skills or abilities, personality or aptitudes of employees or prospective employees, or to otherwise “screen, evaluate, classify, and/or recommend prospective or current employees.”
Additionally, the draft rules would require notification when AI is used to target job postings or advertisements to specific groups or populations, screen resumes, analyze facial expressions, word choice, and voice during online or video interviews, or analyze data obtained from third parties.
However, notification is not required for AI that is used solely for business operations other than hiring decisions, such as generating marketing copy, or for regular software that does not infer or produce output that influences hiring decisions (such as basic word processing, spreadsheets, firewalls, spam filtering, etc.) unless these tools are used to infer or recommend hiring outcomes. Similarly, even if a system has AI capabilities, no notification is required if the employer does not use those capabilities to make covered decisions.
Contents of notification
If notification is required, employers must be specific and transparent. Employers must disclose:
- Names of AI products and their developers and vendors.
- Employment decisions that AI influences or facilitates (e.g., hiring, hiring, discipline, etc.).
- The purpose of the AI system and the categories of personal information or employee data it collects or processes, along with a practical explanation (e.g. summarizing or scoring resumes, analyzing video interviews, evaluating chat interactions with recruitment bots).
- Types of occupations in which AI is used.
- A contact person (e.g., human resources representative or hiring manager) for questions regarding AI and its use.
- The right to request a reasonable accommodation and how to request one. and
- Required legal language.
The notice must be plainly written, provided in a language commonly spoken in the workplace, and reasonably accessible to employees with disabilities.
Timing and method of notification
Employers will be required to notify current employees annually or within 30 days of “the implementation of a new or significantly updated AI system.” For prospective employees, job postings must include notice about the use of AI.
The draft regulations would require different forms of notification, taking into account modern physical and digital work environments and consistency across common communication channels with employees and prospective employees. The notice must (1) be published in the employee handbook or manual; (2) A “conspicuous location'' on the employer's physical premises where the notice is normally posted. (3) “Conspicuously Placed” on the Employer’s Intranet or External Website; (4) any job announcement or posting;
Recordkeeping requirements
The draft rule would further extend record-keeping requirements to the use of AI, requiring employers to keep notices, postings, and disclosures about AI and “records of such use” for four years. The draft regulations also extend several existing retention periods to four years and require employers to retain all relevant records until a charge is adjudicated, regardless of when the charge was filed.
Expanding AI regulations
The draft IDHR regulations represent a new step in AI regulation in Illinois, which will bring employee protections in place with respect to the expanded use of AI technology starting January 1, 2026, similar to other jurisdictions such as California, Colorado, and New York City, although California Governor Gavin Newsom vetoed a bill in October 2025 that would have broadly required employers to notify when using AI. Notably, in addition to the new law's notification provisions specifically addressed in the draft regulations, Illinois law explicitly makes it illegal for employers to use AI in a manner that discriminates against employees or prospective employees, whether or not the discrimination is intentional.
With the federal government rolling back regulations, additional state and local regulation regarding the use and impact of AI technology is expected. However, President Donald Trump recently signed an executive order restricting state and local governments from enacting new laws and regulations related to AI or claiming federal preemption in this area. This order is expressly intended to limit regulation in order to promote the growth of U.S. industry.
next step
Depending on the feedback that IDHR may receive, it is possible, although unlikely, that the draft regulations will undergo significant changes before they are published in April 2016. illinois register and began a public comment period. In any event, the draft regulations provide insight into how IDHR proposes to address implementation of HB 3773's notification provisions.
As such, Illinois employers may want to take note of the new draft regulations and review their use of AI tools to determine whether and how the use of these tools impacts employment decisions. Although HB 3773 does not require a formal bias or impact assessment like New York City or Colorado's AI laws, an active assessment could help uncover whether an employee's use of AI results in discriminatory outcomes, even if such outcomes were unintended. Additionally, employers may want to begin preparing to comply with the notice requirements, including drafting potential notices and implementing steps to provide notices in accordance with the new requirements.
