Quick Hits

• On September 29, 2026, the SBA removed disparate impact liability and race-conscious affirmative action requirements from 13 CFR Part 112, effective immediately, without prior notice and comment.
• The revised rule states Title VI of the Civil Rights Act of 1964 prohibits only intentional discrimination and rests on the same legal authorities as the earlier agency rescissions, Alexander v. Sandoval, Loper Bright, Students for Fair Admissions, and Executive Order 14281.
• The SBA rule preserved the government’s and litigants’ ability to use disparate outcome data to prove intentional discrimination, reinforcing that data collection and preservation obligations remain squarely in place for employers even as the federal enforcement posture changes.

The SBA’s rule change follows the same template that the U.S. Departments of Justice, Interior, Transportation, Agriculture, and Labor used over the preceding ten months, but it also removes affirmative action provisions that the other rescissions did not need to reach.

The SBA Rule

The SBA amended 13 CFR Part 112, its Title VI implementing regulation for federally assisted programs, in three respects. It removed the affirmative action provisions at Sections 112.3(b)(3) and 112.7(d), which authorized, and in some cases required, measures conscious of race, color, and national origin to overcome the consequences of practices that had limited participation, even where the recipient had never used discriminatory policies. It deleted the disparate impact language in Section 112.4 that extended the employment-discrimination prohibition to practices that merely “tend” to have a discriminatory effect, and removed the companion illustration at Section 112.7(a).

Same Pattern, One Additional Layer

The SBA rule follows the pattern in other federal agencies’ actions to rescind disparate impact regulations under Title VI. Indeed, the rule notes that DOJ reviewed and approved it. The SBA relies on the same reasoning as the other agencies: Title VI’s “single, best meaning” is that it prohibits intentional discrimination only. The SBA describes its disparate impact regulations as in “considerable tension” with Title VI and the Supreme Court of the United States’ precedents (including Students for Fair Admissions, Inc. v. President & Fellows of Harvard College). The SBA also invokes Executive Order 14281, which directs the elimination of disparate impact liability “to the maximum degree possible.” SBA issued its regulation without notice and comment and with an immediate effective date, invoking the Administrative Procedure Act’s exception at 5 U.S.C. § 553(a)(2) for rules relating to public property, loans, grants, benefits, or contracts.

The SBA rule goes further than the other rescissions by also addressing affirmative action. Prior sections of its rules authorized and sometimes required funding recipients to use race-conscious measures, such as racial preferences, special consideration of race, and affirmative steps to achieve racial balance, without requiring the narrow tailoring to a compelling governmental interest that strict scrutiny demands. The SBA characterized these provisions as encouraging “the exact conduct the Equal Protection Clause forbids.” The rescission of those provisions is a separate action from the disparate impact removal, and SBA took the position that each amendment operates independently, meaning that invalidation of one would not affect the others.

Disparate Impact Is Out, but the Same Data Can Still Support Liability

Consistent with every prior Title VI disparate impact rescission (by the U.S. Departments of Justice, Interior, Transportation, Agriculture, and Labor (DOL)), SBA stated that eliminating disparate impact liability “does not preclude the use of data on disparate outcomes to help prove intentional discrimination.” It drew the same distinction the DOL rule drew most clearly: using statistical disparity to help establish liability for intentional discrimination as an evidentiary matter “materially differs from using such disparity to impose liability for an unintentional disparate impact.”

That language is now in six federal agency rescission rules. The federal government has not told funding recipients or employers that their data is safe to disregard. It has told them that the same statistics once used to establish disparate impact liability are now positioned as evidence of intentional discrimination, the theory that remains fully enforceable.

SBA’s own cost-benefit analysis acknowledged that the removed provisions were already “effectively inoperative,” which only sharpens the point. The practical change for funding recipients may be modest, but the evidentiary reframing of disparate outcome data, from a standalone basis for liability to a tool for proving intent, is not.

What This Means for Employers and Funding Recipients

The SBA rule narrows part 112 regulations under Title VI. It does not affect Title VII, which independently codifies disparate impact liability, or the state and local statutes that continue to apply it. Employers with SBA relationships may wish to consider:

• Identifying the SBA relationship at issue, i.e., determining whether the organization is applying for or receiving SBA financial assistance covered by part 112, participating in an SBA contracting program, or both.
• Continuing to evaluate intentional and disparate impact discrimination risk by, for example, conducting and structuring adverse impact analyses under attorney-client privilege before a problem surfaces, not after.
• Accounting for continued private and state-level exposure. Private plaintiffs retain a right of action under Title VII of the Civil Rights Act of 1964, and state and local agencies continue to apply disparate impact analysis under their own statutes.
• Recognizing what the data now proves. The demographic and selection data that employers collect and maintain does not become safe because disparate impact liability is rescinded. That same data is now positioned as potential proof of intentional discrimination, the theory the U.S. Equal Employment Opportunity Commission (EEOC) says it is prioritizing.

Ogletree Deakins’ Diversity, Equity, and Inclusion Compliance Practice Group, Government Contracting and Compliance Practice Group, and Workforce Analytics and Compliance Practice Group will continue to monitor developments and will provide updates on the Diversity, Equity, and Inclusion Compliance, Employment Law, Government Contracting and Compliance, and Workforce Analytics and Compliance blogs as additional information becomes available.

This article and more information on how the Trump administration’s actions impact employers can be found on Ogletree Deakins’ Administration Resource Hub.

Follow and Subscribe
LinkedIn | Instagram | Webinars | Podcasts

Quick Hits

• AI tools can be extremely helpful in supporting the need for constant safety oversight in manufacturing environments
• However, employers retain the responsibility to comply with OSHA regulations, even if the employer implemented an AI process designed to protect worker safety.

AI has many positive use scenarios in a manufacturing setting. A camera notices that a worker walked into a press area without safety glasses. A sensor on a conveyor motor picks up a vibration pattern that has previously indicated an impending failure. A forklift slows down as a pedestrian comes around a blind corner. In this AI age, manufacturing employers implement AI resources not only to increase efficiency, but to better promote workplace safety—striving for fewer injuries, fewer disruptions, and a stronger safety culture. But one principle applies to AI in any setting: AI works best as a tool that supports human judgment, not one that replaces it. Without real human oversight, AI safety systems can leave gaps that could lead to worker injuries and OSH Act compliance shortfalls.

How AI Is Supporting Worker Safety

Used well, AI can extend what a safety team is able to see, predict, and respond to. Applications of AI are growing each day, but common uses in manufacturing can include:

• Personal protective equipment (PPE) compliance: Cameras paired with AI models can flag missing hard hats, eye and face protection, or proper gloves.
• Safety alarms: Systems can watch the restricted areas around hazardous machinery or blind spots and send warnings when employees enter an area where collision could occur.
• Predictive maintenance: AI can analyze sensor data and spot signs of equipment wear or failure before a breakdown, reducing the need for emergency maintenance.
• Incident analytics: AI can sort through large volumes of incident and near-miss reports to find trends that would be hard to see manually.
• Collaborative robots: “Co-bots” can take over repetitive or hazardous tasks and reduce workers’ exposure to strain and repeat use injuries.

Each of these tools can make a safety program stronger, but relying too heavily on any of them can create inadvertent gaps in even the most well-intentioned safety program. These gaps create openings for unintended compliance shortfalls and—potentially—employee injuries. Employers may want to continue intentionally monitoring their safety programs—even those supported by AI tools—to identify these gaps and create solutions to fill them.

Why Human Oversight Still Matters

However, AI does not replace competent or qualified persons, hazard assessments, or enforcement of safety rule violations. If employees assume “the system will catch it,” basic safety practices can slip. And it is the employer that remains responsible: the duty to comply with the OSH Act’s general duty clause (Section 5(a)(1)) and applicable Occupational Safety and Health Administration (OSHA) regulations cannot be handed off to a vendor or an algorithm.

Mapping AI to Commonly Cited OSHA General Industry Standards

OSHA has not issued regulations specific to AI use in safety programs. However, the following OSHA general industry standards come up often in manufacturing and are among OSHA’s Top Ten Most Frequently Cited Standards for the most recently reported year. AI can assist with each, but people must still oversee these processes to ensure compliance with OSHA requirements.

The pattern is the same across the board: AI can improve detection and efficiency, but human involvement is necessary to verify the accuracy of written programs, conduct training, carry out inspections, and implement enforcement programs. After all, it is still the employer that will be liable for an OSH Act violation, even if the employer implemented an AI process designed to protect against those violations.

Top Five Tips for Effective Use of AI in Manufacturing Safety Systems

Employers thinking about adopting AI safety tools, or already using them, may want to consider the following tips:

• Robust safety teams. Assessing the current human safety team before implementing AI tools for any safety tasks is a critical early step. Employers may want to identify whether the team is comprised of the right people, with the right experience and knowledge of the workplace, to identify potential hazards and whether AI tools can assist in reducing those risks.
• Intentionality. Consider starting with a traditional hazard identification and assessment. This step includes intentionally identifying the hazards that may exist before implementing an AI process, as well as the hazards that could be caused by the AI tool itself, and implementing only AI processes that are necessary or are tied to a specific safety-related goal.
• Integration, not replacement. Because human oversight is crucial in all safety programs, it is important to note that AI safety tools are not a substitute for an employer’s existing safety and health programs. Human involvement could take many forms, including:
  • Defining alert ownership. Factors include deciding who reviews alerts, how quickly they must respond, and what they are expected to do.
  • Validating and auditing performance. Standard elements in this step include testing systems under real site conditions, and auditing them again after changes to equipment, layout, or processes.
  • Keeping a human in the loop for stop-work decisions. Employers may want to remind their employees that people, not AI, should make the final call on whether to stop or restart work.
  • Training workers and supervisors. Employers may want to ensure that workers and supervisors understand what the tools do, where they fall short, and why ordinary safety practices still apply.
  • Documenting oversight activities. Keeping records of validation, audits, alert responses, and corrective actions demonstrates that employees exercise oversight of the safety programs.
  • Encouraging reporting without retaliation. Employers may want to invite workers to report AI failures or concerns and make sure no one faces retaliation for doing so.

Conclusion

AI has real potential to make manufacturing workplaces safer. It can catch hazards people might miss, predict failures before they happen, and take workers away from dangerous tasks. But the technology is a safety ally, not a safety manager. Employers that pair AI with strong human oversight, clear accountability, and a solid OSHA compliance program will be in the best position to reduce injuries and manage legal risk.

Ogletree Deakins’ Artificial Intelligence and Innovation Practice Group and Workplace Safety and Health Practice Group will continue to monitor developments and provide updates on the Artificial Intelligence and Innovation, Manufacturing, and Workplace Safety and Health blogs as additional information becomes available.

Follow and Subscribe
LinkedIn | Instagram | Webinars | Podcasts

Quick Hits

• A New Jersey federal district court rejected a concrete company’s bid to overturn an arbitrator’s award in favor of an employee who was discharged after testing positive for marijuana.
• The court found that the odor of marijuana alone is not sufficient to establish reasonable suspicion of cannabis use, possession, or impairment as required for an adverse action under CREAMMA.
• Under CREAMMA, an employer may take an adverse employment action against an employee based on—but not solely due to—the presence of cannabinoid metabolites in the employee’s bodily fluids.

Under CREAMMA, marijuana is legal for recreational use. The law prohibits employers from taking adverse employment actions solely based on a positive marijuana test. Employers can require an employee to undergo a drug test when there is reasonable suspicion of an employee’s use of marijuana while on duty, when there are observable signs of marijuana intoxication, or following a work-related accident subject to investigation by the employer.

Background on the Case

In April 2025, as an employee for County Concrete Corp. drove his personal vehicle during a lunch break past a supervisor and coworker on company property, the supervisor and coworker claimed that they detected a strong smell of marijuana from the vehicle. The supervisor ordered the employee to take a drug test, which returned positive for marijuana. The company fired the employee, who had previously tested positive for marijuana in December 2023. The employer had a two-strike policy, stating that two positive drug tests could result in termination of employment. The employee was represented by the International Brotherhood of Teamsters Local 863, which filed a grievance that proceeded to arbitration.

In December 2025, the arbitrator awarded the employee reinstatement to his former position with full back pay, seniority, and benefits. She found that the company failed to establish the employee’s use of, possession of, or impairment from cannabis at the workplace.

County Concrete filed a petition to vacate the arbitrator’s award. It argued that the arbitrator disregarded controlling New Jersey law and exceeded her contractual authority derived from the collective bargaining agreement.

The company argued that the arbitrator ignored CREAMMA when she, in issuing her decision, considered the nonbinding best practices for employers published by the New Jersey Cannabis Regulatory Commission, including relying on observable, physical signs of impairment while on duty and documenting evidence within twenty-four hours or before drug test results arrive, and designating a point person with sufficient training to determine impairment. The union and the employee argued that an odor on clothing or in a car only established that marijuana may have been present at some prior time, not that an employee consumed the drug on an employer’s property or was impaired while on duty.

Federal District Court Ruling

The U.S. District Court for the District of New Jersey noted that, in cases like this one concerning a collective bargaining agreement, courts must give deference to the arbitrator’s decision even if the arbitrator made errors in law or fact. An arbitrator’s award can only be vacated if it was procured by corruption, fraud, or undue means; if there was evident partiality or corruption in the arbitrator; if the arbitrator was guilty of misconduct in refusing to postpone a hearing or hear evidence; or if the arbitrator exceeded his or her powers.

In reviewing the arbitrator’s decision, the district court concluded that the arbitrator did not manifestly disregard CREAMMA or exceed her contractual authority. In confirming the award, the court explained that odor, standing alone, does not establish reasonable suspicion that an employee was using cannabis while performing his or her job duties or that the employee was impaired on employer property. Rather, “[a]n odor, standing alone, establishes only that cannabis may have been present at some point.” The court noted that, in this case, “[t]here was no testimony that Grievant slurred his speech, no evidence that his eyes were glassy, and no evidence that he was impaired while on duty, and no report or evidence of poor work performance.” Further, the employer “allowed the Grievant to complete his shift until 4:30 pm that day without any reports of being impaired.”

The court made further reference to the fact that, contrary to best practices identified by the New Jersey Cannabis Regulatory Commission, the employer did not document the observance until more than a week later, and neither the supervisor nor coworker who observed the marijuana smell were trained in drug impairment recognition. The court found the arbitrator’s award was aligned with established legal precedent and was “rationally derived” from the collective bargaining agreement.

Next Steps

This case shows that, in New Jersey, the odor of marijuana and a positive drug test, without additional evidence, is not sufficient to justify an adverse employment action against an employee suspected of cannabis use, possession, or impairment. It also highlights the importance of drug impairment recognition training for individuals assessing an employee’s potential marijuana use.

Employers in New Jersey may wish to review their employee handbooks and collective bargaining agreements to ensure compliance with state laws regarding drug testing and off-duty marijuana use. Courts in New Jersey are likely to show deference to an arbitrator’s interpretation of a collective bargaining agreement except in rare circumstances.

Ogletree Deakins’ Arbitration and Alternative Dispute Resolution Practice Group and the Drug Testing Practice Group will continue to monitor developments and will post updates on the Arbitration and Alternative Dispute Resolution, Drug Testing, and New Jersey blogs as additional information becomes available.

Leslie A. Lajewski is a shareholder in Ogletree Deakins’ Morristown office.

Steven J. Luckner is a shareholder in Ogletree Deakins’ Morristown office.

This article was co-authored by Leah J. Shepherd, who is a writer in Ogletree Deakins’ Washington, D.C., office.

Follow and Subscribe
LinkedIn | Instagram | Webinars | Podcasts

Quick Hits

• California is leading the charge to regulate AI use in the workplace with a suite of new laws targeting workplace monitoring and automated decision-making.
• California’s AB 1883 bans AI-powered workplace surveillance tools that predict employees’ emotional states or collect neural data, and AB 1331 prohibits employers from using workplace surveillance tools to monitor employees in bathrooms.
• SB 947, the “No Robo Bosses Act,” prohibits employers from relying solely on automated decision systems for discipline or termination decisions, effective July 1, 2027.

The bills include Assembly Bill (AB) 1883, which bans AI-driven workplace monitoring tools that make predictions about employees’ emotional states or collect employees’ “neural” data; AB 1331, which prohibits workplace surveillance in bathrooms; Senate Bill (SB) 947, the “No Robo Bosses Act,” which prohibits employers from relying solely on automated decision systems for discipline or termination. Together, these laws increase employers’ potential liability for using AI-powered tools in the workplace.

AB 1883—Workplace Surveillance Tools

AB 1883, titled “Workplace surveillance tools,” was enacted as employers are being offered an increasing array of tools that monitor everything from text and audio to biometric factors and behaviors. The law follows the European Union’s lead as the European AI Act already bans AI systems used for emotion recognition in the workplace and in educational settings.

AB 1883 bans the use of “workplace surveillance tool[s]” that use AI to either: (1) recognize, or make inferences or predictions about, an individual’s emotional state; or (2) collect neural data. “Workplace surveillance tools” are defined in the law as “any system, application, instrument, or device that collects” information about employees’ activities, communications, biometrics, or behaviors in a way other than direct observation by a human. Such tools include video or audio surveillance, time-tracking, geolocation tools, or others that use “a photo-optical system.” Additionally, the law bans the collection of employees’ “neural data,” meaning information that measures employees’ central or peripheral nervous systems.

Violations can result in statutory or civil penalties of up to $500 per violation. Further, although the law does not provide a private right of action, employers may face liability under the California Private Attorney General Act (PAGA), which allows individuals to pursue Labor Code-based penalty claims on behalf of the state.

AB 1331—Workplace Surveillance of Employees

AB 1331 also targets workplace monitoring or surveillance tools that collect “employee data, activities, communications, actions, biometrics, or behaviors by means other than direct observation by a person.” The law prohibits employers from using a workplace surveillance tool to monitor or surveil employees in workplace bathrooms, with limited exceptions, and allows employees to leave surveillance devices behind when entering those areas. AB 1331 permits employers to require employees to use a “workplace surveillance tool, including a badge or personal alarm system” to “access a locked or secured bathroom” so long as it does not have an audio or video recording capability and “does not have physically embedded or attached artificial intelligence.” AB 1331 shares an enforcement structure with AB 1883, including civil penalties of up to $500 per violation.

SB 947—‘No Robo Bosses Act’

SB 947, known as the “No Robo Bosses Act,” will take effect on July 1, 2027. The law is a newer version of a similar law that Governor Newsom vetoed in 2025. The law prohibits employers from relying solely on an automated decision system (ADS) for discipline or termination decisions, requiring a human reviewer to independently corroborate the ADS output. SB 947 also bars using an ADS to infer a worker’s protected characteristics or to predict and retaliate against a worker for exercising legal rights. Additionally, the law requires employers to provide post-use notice to affected employees when an ADS is used in employment decisions.

Increasing Scrutiny of Monitoring Tools

Employers have increasingly embraced workplace monitoring tools to increase productivity, efficiency, and workplace health and safety. An array of monitoring tools is now available that exceed traditional text, audio, internet, and keystroke activity recording. These newer technologies collect and analyze employees’ location data; biometric factors such as heart rate, sweating, and brainwaves (often via wearable devices); linguistic data; and visual recordings that capture micro-expressions, body language, and eye movements. Data can then be analyzed using AI to make predictions about employees, potentially their mental states. This can allow employers to intervene when an employee is in danger or provide support when necessary.

AB 1883 targets many of these AI-driven tools, but the final version of the law is more narrowly tailored than prior versions. It allows non-AI-powered workplace monitoring tools that are not used for the prohibited purposes and explicitly permits employers to use a “workplace surveillance tool” to “ensure safety.” This addresses concerns from some opponents that the bill would potentially have banned basic security surveillance tools and safety measures, such as sensors that can detect distracted or fatigued drivers.

Still, it will be necessary for employers to conduct due diligence on any monitoring tools being implemented or considered to determine what predictions or outputs the tools produce and whether the data collected could be considered “neural data.”

Evolving AI Legal Landscape

The latest package of bills once again puts California at the forefront of AI regulation in the United States. Whereas, under the current administration, there has been a dearth of federal workplace AI legislation, regulations, or other rulemaking, and no action to address AI-powered monitoring tools is expected in the foreseeable future.

The signings come on the heels of Governor Newsom signing two AI executive orders. N-10-26, also signed on September 30, 2026, declares that all state agencies and departments refer to AI technologies as “artificial intelligence” or “AI” regardless of “any rebranded or different terminology used by the federal government.” Signed on September 18, 2026, N-9-26 directs the state to explore increased oversight of AI development, including potentially requiring a “kill switch” for frontier AI models.

In addition, California, Colorado, Connecticut, Illinois, New York City, and Texas all have laws that restrict the use of automated decision tools for significant or consequential employment decisions, such as hiring, promotion, and termination. Those laws could arguably apply to AI-powered employee monitoring tools, particularly if the output of such tools is used to influence key employment decisions. It is expected that states will continue to legislate in this area.

Next Steps

California continues to be on the frontier of AI legislation. Employers should pay close attention to these new California AI laws and the evolving regulatory landscape governing the use of monitoring and AI-powered tools in the workplace. Before implementing a specific tool, employers may want to thoroughly vet what data the tool collects, how that data will be used, and whether the tool uses AI to generate any output from that data, as these new laws place additional compliance requirements on employers.

Employers using automated decision systems for discipline or termination decisions may want to prepare for SB 947’s requirements, including the human reviewer mandate effective July 1, 2027. More broadly, employers may further wish to consider the impact on employee morale and workplace culture when considering the use of a monitoring tool.

Ogletree Deakins’ Artificial Intelligence and Innovation Practice Group will continue to monitor developments and will provide updates on the Artificial Intelligence and Innovation, California, Cybersecurity and Privacy, and Employment Law blogs as additional information becomes available.

In addition, the Ogletree Deakins Client Portal provides subscribers with timely updates on California state laws, including California’s updated laws on Automated Employment Decisions. Premium-level subscribers have access to comprehensive law summaries, policies, and templates. Snapshots and Updates are complimentary for all registered client users. For more information on the Client Portal or a Client Portal subscription, please email clientportal@ogletree.com.

Follow and Subscribe
LinkedIn | Instagram | Webinars | Podcasts

Quick Hits

• Senate Bill 690 limits website-related California Penal Code Section 638.51 claims to actions brought by the attorney general, effective January 1, 2027, with a two-year retroactive reach.
• The law curbs private trap-and-trace lawsuits involving cookies, pixels, and similar tracking technologies, though constitutional challenges and other claims may remain.
• Companies may want to continue reviewing tracking tools, third-party data sharing, consent practices, privacy policies, and cookie-banner disclosures.

The amendment to Section 637.2 provides that: “An action against a private actor for a violation of Section 638.51 alleged to arise from conduct occurring on an internet website, online application, or mobile application may be brought under this section only by the Attorney General.”

For context, Section 638.51 provides that a person may not install or use a pen register or a trap and trace device without first obtaining a court order. A “pen register” is a “device or process that records or decodes dialing, routing, addressing, or signaling information transmitted by an instrument or facility from which a wire or electronic communication is transmitted, but not the contents of a communication.” Penal Code ¶ 638.50(c) defines a “trap and trace device” as a “device or process that captures the incoming electronic or other impulses that identify the originating number or other dialing, routing, addressing, or signaling information reasonably likely to identify the source of a wire or electronic communication, but not the contents of a communication.” California courts have concluded that software (such as cookies) may qualify as a pen register or trap and trace device, and that third-party pixels and other tracking technologies used on websites may also fall within the statutory definition. As a result, plaintiffs have used Section 638.51 to bring a wave of lawsuits, often brought as putative class actions.

The relief provided by the California Legislature was to amend 637.2 of the California Penal Code, which eliminates a private right of action under California Penal Code Section 638.51 and mandates that only the attorney general can bring trap and trace claims. The law takes effect on January 1, 2027, and applies retroactively for two years, which means it applies to lawsuits that have been filed over the last two years. It seems like a perfect result.

Not so fast, this is California after all. The retroactive application will likely be challenged on constitutional grounds. In addition, the plaintiffs’ bar believes that this law does not provide a blanket immunity and that conduct that constitutes intentional, surreptitious, and independently tortious interception and exploitation of consumer communications will still be subject to Section 638.51. Finally, this law does not affect claims under Penal Code Section 631(a), which addresses wiretapping, and Penal Code Section 632, which addresses recording confidential communications. These are other common causes of actions in lawsuits challenging the operation of a website.

The most important takeaway is that, even with this change in California law, careful review and maintenance of company websites remains a priority. Key points include using cookie banners on company websites and crafting consent and privacy policies; reviewing website analytics or tracking tools; and determining whether data is collected and shared with third parties, and whether opt-in consent for tracking tools may be appropriate in certain high-risk states.

Ogletree Deakins’ California offices and Cybersecurity and Privacy Practice Group will continue to monitor developments and will post updates on the California and Cybersecurity and Privacy blogs as additional information becomes available.

Follow and Subscribe
LinkedIn | Instagram | Webinars | Podcasts

Quick Hits

• In 2025, Oregon enacted SB 537, which expanded the types of healthcare employers are required to establish a workplace violence prevention plan, conduct safety and security assessments, and train employees on responding to threats of violence.
• In June 2026, Oregon OSHA published a proposed rule implementing the law, which is expected to take effect in January 2027.
• Oregon OSHA’s proposed rule would broaden the definition of “workplace violence” and explicitly require home health agencies and home hospice programs to have workplace violence prevention programs, in addition to hospitals and ambulatory surgical centers, which were already required to have such programs.

Oregon OSHA’s proposed rule implementing SB 537 defines “workplace violence” as “any act or threat of physical violence, assault, homicide, harassment, or intimidation or any other threatening behavior that occurs in the workplace.” This broad definition means compliant prevention plans would have to address a wider range of conduct than physical attacks alone.

The proposed rule would add a new provision that specifies that it is illegal to discriminate against healthcare workers for reporting workplace violence that occurs on the premises of a healthcare employer or in the home of a patient receiving home healthcare services.

The proposed rule would require healthcare employers to conduct periodic security and safety assessments to identify threats of workplace violence and develop and implement a workplace violence prevention plan based on the security and safety assessments. Healthcare employers would be required to give each employee a written copy of the workplace violence prevention plan, including a statement explaining that employees who report workplace violence have a right to be protected from retaliation. 

The proposed rule also would require healthcare employers to provide annual workplace violence prevention training to employees and any contracted security staff who work on an employer’s premises.

Recordkeeping Requirement

Under the proposed rule, healthcare employers would be required to maintain a record of all workplace violence incidents that occur on the healthcare employer’s premises or in the home of a patient receiving home healthcare services. The record would be required to include:

• the name and address of the premises on which the incident occurred;
• the date, time, and location where the incident occurred;
• the name, job, and department or ward assignments of the employee who was affected;
• a description about the person who committed the assault as a patient, visitor, employee, or other category;
• a description of injuries like minor soreness, surface abrasions, scratches, bruises, severe lacerations, bone fracture, head injury, loss of limb, or death;
• a description of any weapon used;
• the number of employees, including nursing staff, in the immediate area of the assault when it occurred; and
• the actions employees and the healthcare employer took in response to the incident.

Next Steps

Oregon OSHA’s proposed rule reflects a growing trend toward comprehensive, prevention-focused workplace violence protections in healthcare settings. The department said that the proposed rule will “tentatively” be adopted in October 2026 and take effect in January 2027. While waiting for OSHA Oregon to finalize the regulatory guidance, employers may wish to audit their current program and conduct a gap analysis, prioritize compliance efforts, and develop an implementation timeline ahead of compliance deadlines.

Ogletree Deakins’ Workplace Violence Prevention Practice Group will continue to monitor developments and will post updates on the Healthcare, Oregon, and Workplace Violence Prevention blogs as additional information becomes available.

Emma J. Redden is a shareholder in Ogletree Deakins’ Portland, OR, office.

Jacqueline M. Duvall is of counsel in Ogletree Deakins’ Kansas City office.

Kenneth M. Rock is of counsel in Ogletree Deakins’ Portland, OR, office.

This article was co-authored by Leah J. Shepherd, who is a writer in Ogletree Deakins’ Washington, D.C., office.

Follow and Subscribe
LinkedIn | Instagram | Webinars | Podcasts

Quick Hits

• Beginning January 1, 2028, employers already subject to California’s mandatory harassment prevention training requirements must incorporate an anti-hate speech component into that training.
• AB 1803 does not create a separate training obligation and does not increase the existing one-hour and two-hour training minimums. Employers may fold the new content into their current programs.
• The bill does not define “hate speech,” a gap the Assembly Committee on Labor and Employment flagged during the legislative process, and one that employers should watch as agency guidance develops.

Background on California’s Harassment Prevention Training Mandate

California employers with five or more employees are already required to provide at least two hours of harassment prevention training to supervisory employees and at least one hour to nonsupervisory employees every two years. New nonsupervisory employees generally must be trained within six months of hire, and new supervisory employees within six months of assuming a supervisory role. The required training currently must address sexual harassment, prevention of abusive conduct, and harassment based on gender identity, gender expression, and sexual orientation. AB 1803 amends this existing framework rather than replacing it.

What AB 1803 Adds

Beginning January 1, 2028, covered employers must include anti-hate speech training as a component of their existing harassment prevention training. That component is expected to give supervisors and employees practical guidance on recognizing, reporting, and confronting workplace speech that vilifies, humiliates, or incites hatred against people based on the Fair Employment and Housing Act’s (FEHA) protected characteristics, including race, religious creed, color, national origin, ancestry, physical or mental disability, reproductive health decision making, medical condition, genetic information, marital status, sex, gender, gender identity, gender expression, age, sexual orientation, and veteran or military status.

Notably, the bill does not increase the existing training time requirements or require a standalone course. Employers may incorporate the new content into their existing harassment prevention programs, and training will satisfy the new requirement if it already includes information about both the prohibition against unlawful harassment and the prevention of abusive conduct based on protected characteristics. As a result, some employers’ current training may already cover part or all of the new requirement.

The Undefined ‘Hate Speech’ Problem

AB 1803 does not define “hate speech.” The law does not distinguish between harassment based on a protected characteristic, which is unlawful under FEHA, and “hate speech,” which is not otherwise addressed in FEHA, and which may or may not be different behavior.

The Assembly Committee on Labor and Employment’s analysis of the bill noted that committee staff was unaware of any existing definition of hate speech under California law and observed that hate speech itself is not illegal but can violate employment law when it rises to an actionable level of workplace harassment or discrimination. Lawmakers flagged this gap as the bill moved through committee, but it was not resolved before the bill reached the governor’s desk.

For employers, this creates some ambiguity heading into 2028. The lack of a fixed statutory definition allows flexibility in how training content is framed, but it also leaves open questions about how “recognizing and confronting” hate speech will be interpreted in practice, both by trainers and, eventually, by courts and the Civil Rights Department.

Key Takeaways

AB 1803 does not overhaul California’s harassment prevention training scheme, but it does add a substantive new component that covered employers will need to build into their existing programs beginning January 1, 2028. Employers should begin reviewing current training content, coordinating with training vendors, and confirming that reporting and investigation practices can handle complaints involving hate speech, ahead of the effective date.

Ogletree Deakins’ California offices will continue to monitor developments with respect to AB 1803 and related legislation affecting California’s workplaces and will provide updates on the firm’s California and Workplace Violence Prevention blogs as additional information becomes available.

Follow and Subscribe
LinkedIn | Instagram | Webinars | Podcasts

Quick Hits

• Beginning in January 2027, the IDOL will no longer allow employers to use “prefers not to identify” for reported employees’ race/ethnicity.
• The IDOL notes an academic study that suggests ways to examine EPRC filings and possibly take stronger action based on statistical differences.
• The IDOL has updated its process for handling overdue/unmade EPRC filings from three (3) steps to two (2) steps, meaning filers have less time to remedy filing issues and avoid penalties.
• The IDOL’s July 2026 notice makes it clear to filers that EPRC filing requirements are not tied to EEO-1 filing requirements, meaning that EPRC requirements will continue even if EEO-1 filings end.

The July email notes that, while there have been over 9,000 EPRC submissions to the IDOL since 2022, most employers interact with the EPRC process only every two years, and this notice helps ensure employers stay informed.

Significant Demographic Reporting Changes

The July email includes a discussion of the IDOL’s October 2025 addition of “Middle Eastern or North African” as a race/ethnicity reporting option for EPRC filers, implemented by modifying the CSV upload file used to provide EPRC data. While this race/ethnicity category has been added to both Illinois and California pay reporting, it has not been adopted by the U.S. Equal Employment Opportunity Commission (EEOC) for EEO-1 reporting.

In July 2026, the IDOL again updated the upload CSV file to remove the previously available race/ethnicity and sex option of “prefers not to identify.” Illinois employers could previously use this option to report employees who declined to provide their race/ethnicity and/or gender. Despite this mid-2026 change, the IDOL will still allow EPRC filers to continue to use the prior version of the CSV upload file containing the “prefers not to identify” field until January 2027, provided that they have previously downloaded the old version of the CSV upload file. Beginning in January 2027, the EPRC filing system will no longer accept any submissions containing “prefers not to identify.” The IDOL advises employers, through answers to frequently asked questions (FAQs), that they may upload an explanatory document providing additional information about employees that they wish to share.

For employees who decline to provide their race/ethnicity and/or gender, the IDOL advises that the employer may use employment records or observer identification to determine the missing self-identification (self-ID) information. The IDOL advises that records should be kept separately from an employee’s “basic personnel file or other records available to those responsible for personnel decisions.” In describing the self-ID process for EPRC reporting, the IDOL states that it follows the essence of federal guidance for self-ID, noting that employers should offer employees the opportunity to self-ID and provide a statement concerning the voluntary nature of this opportunity. The EPRC FAQs include a sample communication that can be shared with employees.

While still pending, the EEOC’s effort to end EEO-1 reporting could impact employers’ collections of race/ethnicity and gender data from all employees. If this happens, employers may need to consider a more narrowly focused collection of self-ID information for employees who will be reported in their EPRC filings.

Illinois Is Reviewing EPRC Data and Considering Next Steps, Including Self-Audit Tools

The IDOL partnered with the University of Illinois Urbana-Champaign Project for Middle Class Renewal to analyze the filed 2021–2023 EPRC data. This partnership included the production of a report (“UIUC Project Report”), which presented findings on pay disparities and recommended best practices for the IDOL and employers.

EPRC FAQ 30 provides more information on the UIUC Project, including access to the full report, which contains best-practice recommendations and voluntary HR-oriented practices published on pages 12–14. This report’s key findings include discussions of gender, racial, and ethnic pay gaps. In discussing the identified gender pay gaps, the report states that the analyzed EPRC pay data “reveals a consistent and measurable wage gap” between men and women working at “large private-sector firms” in Illinois. While the size of the pay gap varies with several factors, the report states that women in Illinois earn less on average than men “even when working in the same job categories for the same employers.” The analyses also revealed “larger and consistent wage gaps” between White workers and workers of color across large Illinois private-sector employers. More specifically, the report notes that “clear pay inequities remain” for Black and Hispanic workers in “lower-paid roles.”

The report recommends using a threshold-based flagging system to identify employers with unusually large wage gaps by comparing percentage differences relative to the dominant group, which is typically composed of whites and males.

The flagging system includes three (3) tiers: (1) informational for employers with average pay gaps of 5 percent to 15 percent; (2) moderate risk for employers with gaps exceeding 15 percent but below 25 percent, which would be recommended for further review; and (3) high risk for employers with wage gaps exceeding 25 percent, which would warrant immediate attention. As EPRC data collection improves, the report suggests that the system can be updated to include additional thresholds, such as the top 5 percent of employers with the widest gap within certain job classifications or geographic regions.

EPRC FAQ 30 advises employers on the report’s recommendations, which they can use to evaluate their company’s pay equity compliance. This FAQ states that the IDOL is considering different options for employer self-audit tools, which the IDOL hopes to make “available soon.”

Additional Administrative Updates

In addition to these major developments, the IDOL notified July email recipients of minor updates, including changes to the process for dealing with overdue or unmade EPRC filings. These changes reduce the process from three steps to two, giving filers less time to remedy filing issues and avoid penalties. The new process discusses penalties beginning with the first step, unlike the prior process.

The IDOL also made it clear that the EPRC requirement has been completely decoupled from the EEO-1 filing requirement, meaning that the EPRC requirement will continue if EEO-1 reporting ends.

The July email also notes improvements within the EPRC portal. This includes allowing users to obtain copies of certificates and providing an improved notice of the thirty-day deadline to make revisions to a rejected submission before the filing must be refiled, including a new filing fee. There was also a notice that a new EPRC reporting tool will be rolled out to users on January 1, 2027.

Key Takeaways

• Beginning no later than January 1, 2027, all employees, including those in EPRC reports, must have their gender and race/ethnicity established, even if they declined to provide self-ID information. For such employees, the IDOL directs employers to use employment records or observer identification to populate the missing self-ID information.
• The IDOL has reviewed the first three years of filed EPRC reports as part of the UIUC Project Report and is considering next steps. This report found specific evidence of pay gaps on the basis of both gender and race/ethnicity. The IDOL indicates that it is considering self-audit tools based on these findings and “hopes” to make them available soon. While the report’s threshold-based flagging system is not directly addressed by the IDOL in the July email or the EPRC FAQs, it provides a framework for the IDOL to easily evaluate employer submissions and consider them for additional follow-up.
• While it will be important to see if the IDOL takes concrete steps with respect to this threshold flagging system, employers who file EPRC reports may want to consider analyzing their data using the proposed system to see how their data performs as well as to determine explanations for identified differences.
  • Employers may also want to consider more in-depth analyses, including a privileged pay audit in cases of large pay gaps shown by this process.

The July email shows that the IDOL continues to focus on changes to increase the validity of the filed EPRC data. It also suggests a shift from ensuring employers report as required to a deeper review of the filed data with an eye toward possible enforcement action. Employers may want to keep a careful eye on their EPRC reporting obligations moving forward.

Ogletree Deakins’ Chicago office, Government Contracting and Compliance Practice Group, Pay Equity Practice Group, and Workforce Analytics and Compliance Practice Group will continue to monitor developments with respect to the Illinois EPRC application and review process and will provide updates on the firm’s Government Contracting and Compliance, Illinois, Pay Equity, and Workforce Analytics and Compliance blogs as additional information becomes available.

Follow and Subscribe
LinkedIn | Instagram | Webinars | Podcasts

Quick Hits

• The October 2026 Visa Bulletin reflects the start of FY 2027 immigrant visa number availability, with the State Department noting advancement across various immigrant visa categories.
• USCIS has confirmed that employment-based adjustment-of-status (“green card”) applicants may use the Dates for Filing chart in October 2026, opening filing eligibility to a substantially larger pool of applicants than the Final Action Dates chart alone would allow.
• EB-2 India and EB-5 India Unreserved, which were unavailable in September 2026, again have October 2026 final action dates.
• Some employment-based “Rest of World” categories retrogressed in October 2026, and the State Department stated that the retrogressions are intended to keep visa issuances within FY 2027 quarterly and annual limits.

U.S. Citizenship and Immigration Services (USCIS) has separately confirmed that employment-based adjustment-of-status applicants may file in October using the Dates for Filing chart in the October 2026 Visa Bulletin rather than the more restrictive Final Action Dates chart, meaning the wider dates for filing windows described below determine October filing eligibility.

Dates for Filing Control for October 2026 filing eligibility

USCIS has confirmed applicants may use this chart for October employment-based adjustment filings. That makes filing eligibility and final-approval timing two separate tracks this month: dates for filing controls whether Form I-485 can be filed now (plus related Employment Authorization Document (EAD)/advance parole eligibility); final action dates still controls final approval.

• EB-1: China and India both July 1, 2024 (up seven months from December 1, 2023).
• EB-2: Rest of World (ROW)/Mexico/Philippines March 15, 2026; China January 1, 2023 (~one year advance); India January 15, 2015 (unchanged).
• EB-3: ROW/Mexico August 1, 2024; China April 1, 2024; India January 15, 2015; Philippines January 1, 2024.
• EB-5 Unreserved: ROW/Mexico/Philippines current; China March 1, 2021; India May 1, 2024.

Employment-based Category	All Chargeability Areas Except Those Listed	China-mainland born	India	Mexico	Philippines
EB-1	Current	July 1, 2024	July 1, 2024	Current	Current
EB-2	March 15, 2026	January 1, 2023	January 15, 2015	March 15, 2026	March 15, 2026
EB-3	August 1, 2024	April 1, 2024	January 15, 2015	August 1, 2024	January 1, 2024
Other Workers	June 1, 2022	October 1, 2020	January 15, 2015	June 1, 2022	June 1, 2022
EB-5 Unreserved	Current	March 1, 2021	May 1, 2024	Current	Current
EB-5 Set Aside Categories: Rural, High Unemployment, Infrastructure	Current	Current	Current	Current	Current

Source: U.S. Department of State, October 2026 Visa Bulletin

Notably, EB-2 and EB-3 ROW/Mexico filing dates (March 15, 2026, and August 1, 2024) sit well ahead of their retrogressed final action dates, so most affected applicants can still file in October even though final approval will wait. Employees whose priority dates clear the dates for filing cutoff but not the final action date may now file and may be eligible for interim benefits (EAD, advance parole) while final action is tracked separately.

Final Action Dates (Approval/Visa Issuance)

• EB-1: Current for ROW, Mexico, Philippines. China unchanged at July 1, 2023. India advances to February 1, 2023 (from October 15, 2022).
• EB-2: India returns at November 1, 2013 (from unavailable). China advances to October 1, 2021. ROW/Mexico/Philippines retrogress to January 1, 2025.
• EB-3: ROW/Mexico retrogress to May 15, 2024. China advances one week to January 8, 2022. India unchanged at January 1, 2014. Philippines advances to August 15, 2023.
• EB-3 Other Workers: ROW/Mexico retrogress to January 1, 2022. China advances to October 1, 2019. India unchanged. Philippines advances to January 1, 2022.
• EB-4/Religious Workers: December 15, 2022, across all countries. H.R. 6500 (signed September 2, 2026) extends the Religious Workers category through December 11, 2026.
• EB-5 Unreserved: China unchanged at December 1, 2016. India returns at December 1, 2023 (from unavailable).
• EB-5 Set Aside Categories: Remain current.

Employment-based Category	All Chargeability Areas Except Those Listed	China-mainland born	India	Mexico	Philippines
1st	Current	July 1, 2023	February 1, 2023	Current	Current
2nd	January 1, 2025	October 1, 2021	November 1, 2013	January 1, 2025	January 1, 2025
3rd	May 15, 2024	January 8, 2022	January 1, 2014	May 15, 2024	August 15, 2023
Other Workers	January 1, 2022	October 1, 2019	January 1, 2014	January 1, 2022	January 1, 2022
EB-5 Unreserved	Current	December 1, 2016	December 1, 2023	Current	Current
EB-5 Set Aside Categories: Rural, High Unemployment, Infrastructure	Current	Current	Current	Current	Current

Source: U.S. Department of State, October 2026 Visa Bulletin

Because USCIS has authorized dates for filing, the retrogressions above mainly affect when a case can be finally approved, not whether it can be filed in October. Since chart selection is reassessed monthly, this authorization is not guaranteed to carry into November.

Next Steps

With USCIS’s confirmation that the Dates for Filing chart governs October 2026 employment-based adjustment filings, the gap between the two charts may be relevant to employers and foreign national employees whose priority dates fall between the dates for filing cutoff and the final action date. The return of final action dates for EB-2 India and EB-5 India Unreserved may also affect applicants who were unable to move forward during September’s unavailability. Because USCIS reassesses chart selection each month, it is not yet known whether the October authorization will carry into November 2026. Despite this month’s movement, significant backlogs remain in the EB-1, EB-2, EB-3, and EB-5 categories for applicants chargeable to China and India, and those backlogs may continue to affect long-term retention and work-authorization considerations.

Ogletree Deakins’ Immigration Practice Group will continue to monitor developments and will post updates on the Immigration blog as additional information becomes available.

Follow and Subscribe
LinkedIn | Instagram | Webinars | Podcasts

• Twenty-eight states require employers to provide either paid or unpaid leave for voting.
• Some states have notice requirements related to employees’ right to take time off to vote.
• Election Day is November 3, 2026, but many states have in-person early voting that starts mid-October.

Quick Hits

Twenty-eight states and Washington, D.C., require employers to provide time off for employees to vote. Many states also require time off for employees who serve as election officials or voting machine technicians. Some states also require employees to request time off in advance.

For example, Colorado’s law, which took effect on June 1, 2026, mandates up to two hours of paid leave to vote on any day the voter service and polling centers are open, not just on Election Day. In Maryland, employers must provide employees who claim to be registered voters up to two hours of paid leave to vote on Election Day, but only if the employee does not have two consecutive nonworking hours to vote while the polls are open. California’s law prohibits employers from requesting that an employee bring a mail ballot to work or vote with a mail ballot at work. 

Employer obligations related to voting do not stop at providing time off. Some states have notice provisions that require employers to display a poster or online information describing employees’ right to take time off to vote.

Like other leave laws, the wide variation in state voting leave laws may make compliance complicated for multistate employers. Using state-specific supplements or addenda to employee handbooks may be helpful for multistate employers to address unique state and local laws. While having state-specific voting leave policies is not generally required, this approach may be the most efficient way to help employees and managers navigate these situations. Additionally, employers may want to consider training managers on how to respond to employee requests for time off in these scenarios, regardless of whether jurisdiction-specific policies are used.

From a practical perspective, maintaining adequate staffing can obviously be a concern for certain industries or corporate units that operate 24/7, including many healthcare facilities, hospitality businesses, manufacturing facilities, convenience stores, and gas stations. To permit time off to vote while the polls are open, employers may consider options like shift adjustments, shift swapping, or extra staffing on Election Day.

Next Steps

Employers may want to review their employee handbooks and written policies to ensure they comply with applicable state and local laws on voting leave. They may also wish to train managers to properly handle requests for time off to vote.

Ogletree Deakins’ Leaves of Absence and Multistate Advice and Counseling Practice Groups will continue to monitor developments and will post updates on the Leaves of Absence, Multistate Compliance, and State Developments blogs as additional information becomes available.

In addition, the Ogletree Deakins Client Portal covers legal developments in state and major locality voting and election leave laws. Premium-level subscribers have access to comprehensive law summaries and policies; Snapshots and Updates are complimentary for all registered client users. For more information on the Client Portal or a Client Portal subscription, please email clientportal@ogletree.com.

Lucas J. Asper is a shareholder in Ogletree Deakins’ Greenville office.

This article was co-authored by Leah J. Shepherd, who is a writer in Ogletree Deakins’ Washington, D.C., office.

Follow and Subscribe
LinkedIn | Instagram | Webinars | Podcasts