Artificial intelligence (“AI”) and other emerging technologies are rapidly changing the workplace—and California lawmakers have responded. In 2026, Governor Newsom signed several significant bills regulating AI-assisted employment decisions and workplace surveillance, while vetoing another that would have imposed new AI-related notice requirements on public agencies.

For California public agencies, these new laws establish important limits on how automated systems may be used in employment decisions, where employees may be monitored, and what information AI-powered tools may collect or determine about employees.

Senate Bill (“SB”) 947: A Human Must Remain in the Loop

SB 947—referred to during the legislative process as the “No Robo Bosses Act”—regulates employers’ use of automated decision systems (“ADS”) in employment decisions. The law becomes operative July 1, 2027, and expressly applies to public employers.

An ADS generally includes a computational process derived from machine learning, statistical modeling, data analytics, or AI that produces a score, classification, recommendation, or other simplified output used to assist or replace human discretionary decision-making and that materially impacts individuals. Ordinary tools such as calculators, databases, antivirus software, firewalls, and identity and access-management tools are excluded.

SB 947 prohibits employers from using an ADS to violate or prevent compliance with labor, employment, occupational safety, or civil rights laws; determine an employee’s protected status under the Fair Employment and Housing Act; or predict and take adverse action against a worker for exercising legal rights.

Most significantly, an employer may not rely solely on an ADS when making a disciplinary or termination decision.

If an employer primarily relies on an ADS output, a human reviewer must corroborate the decision using relevant information, such as supervisory evaluations, personnel records, employee work product, peer reviews, or witness interviews. If the output cannot be corroborated, or the reviewer determines it is inaccurate, incomplete, or misleading, the employer may not use it to make the decision.

The law also requires a separate written notice when an employer primarily relies on an ADS in making a disciplinary or termination decision. Among other things, the notice must disclose that an ADS was primarily relied upon and human-reviewed, provide a means to contact a human for additional information, and inform the employee of the right to obtain a meaningful description of the employee’s data used by the system.

To ensure compliance and sound personnel practices, public agencies should not treat automated performance scores, attendance analytics, productivity systems, and similar technologies as substitutes for an appropriate disciplinary investigation. Agencies relying substantially on technology-generated findings should establish procedures documenting the human review, the information considered, and how the reviewer independently corroborated the result.

SB 947 also permits its requirements to be expressly waived through a collective bargaining agreement, but only if the agreement clearly and unambiguously waives the statute, addresses wages or earnings and other working conditions, and provides protection from algorithmic management.

Violations may be enforced by the Labor Commissioner or public prosecutors and are subject to a civil penalty of $500 per violation.

Assembly Bill (“AB”) 1883: New Limits on AI-Powered Workplace Surveillance

AB 1883 regulates certain uses of workplace surveillance technology. It takes effect January 1, 2027, and expressly applies to public employers, including charter cities.

The law broadly defines a “workplace surveillance tool” as a system, application, instrument, or device that collects or facilitates the collection of employee data, activities, communications, actions, biometrics, or behaviors by means other than direct human observation. Examples include video and audio surveillance, continuous time tracking, geolocation, electromagnetic tracking, and photoelectronic tracking.

AB 1883 targets two specific uses of AI. Employers may not use a workplace surveillance tool that uses AI to:

  • recognize, infer, or predict an employee’s emotional state; or
  • collect “neural data,” meaning information generated by measuring activity of an employee’s central or peripheral nervous system that is not inferred from non-neural information.

The law does not prohibit surveillance tools used for safety or other purposes that do not engage in these prohibited activities.

Notably, Governor Newsom urged the Legislature to revisit AB 1883 even as he signed it. His signing message cautioned that the law “lacks key definitions” that may create uncertainty about which technologies are covered and encouraged further clarification to avoid inadvertently restricting beneficial security and other legitimate technologies.

Public agencies should inventory employee-monitoring technologies—not merely products marketed as “AI”—and determine whether any use AI or machine learning to infer emotions or collect neural information. Particular attention should be given to biometric, wearable, safety-monitoring, and employee-analytics technologies whose functionality may extend beyond what agency personnel realize.

Violations may be enforced by the Labor Commissioner or a public prosecutor, with civil penalties of up to $500 per violation.

AB 1331: Keep Workplace Surveillance Out of the Bathroom

AB 1331 addresses another form of workplace surveillance: monitoring employees in workplace bathrooms. It takes effect January 1, 2027, and expressly applies to state and local public employers, including charter cities.

AB 1331 uses a similarly broad definition of “workplace surveillance tool,” but the final legislation is considerably narrower than earlier versions of the bill and focuses specifically on bathrooms.

Unless directed by a court order, an employer may not use a workplace surveillance tool to monitor or surveil employees in a workplace bathroom. Employees also generally have the right to leave behind workplace surveillance tools before entering a bathroom. Exceptions apply, including for employees required to remain available during meal or rest periods pursuant to federal law or existing state law, and for tools used only to record an employee’s one-time entry into and exit from a bathroom.

Employees may be required to carry such electronic badges, personal alarms, and similar devices used for identification, safety, or access to secured bathrooms in certain circumstances, provided the device does not detect or record audio or video and does not have physically embedded or attached AI.

For public agencies, compliance involves more than ensuring there are no cameras in restrooms. Agencies should review GPS-enabled radios, body-worn cameras, electronic badges, smartphones, wearable safety devices, and other equipment capable of generating location or activity data to determine whether they qualify as workplace surveillance tools and whether existing policies and settings comply with the new restrictions. Public safety agencies should pay particular attention to the exceptions for identification and safety devices.

Violations may result in civil penalties of up to $500 per violation.

AB 2656: One AI Mandate Left on the Cutting-Room Floor

Not every proposed workplace AI mandate became law. Governor Newsom vetoed AB 2656, which would have required covered public employers to provide recognized employee organizations at least 45 days’ written notice before developing, purchasing, or requiring generative AI to perform services within the scope of work of a represented classification.

The Governor’s veto message is significant for public-sector labor relations. Although agreeing that employees should be informed when employers introduce technologies that significantly affect them, Governor Newsom concluded that AB 2656 would create “redundant administrative layers” and potentially slow deployment of even innocuous technologies. More importantly, he stated that these issues are “best resolved through the collective bargaining process,” allowing employers and employees to tailor rules to their specific workplaces.

The veto means public agencies will not face a new, across-the-board statutory 45-day notice requirement for introducing generative AI to perform bargaining-unit work. It does not, however, eliminate existing labor-relations obligations. California public-sector collective bargaining law may independently require notice and an opportunity to bargain when new technology changes matters within the scope of representation or has reasonably foreseeable negotiable effects.

Other AI Developments Affecting Public Agencies

Governor Newsom also issued executive orders directing the State to prepare for AI’s potential effects on California’s workforce and strengthen standards for state procurement and use of AI. Among other things, the orders address AI-related workforce disruption, worker protections and retraining, and privacy, security, civil rights, and responsible-use standards for AI vendors doing business with the State.

A few other new, technology-related laws warrant mention:

  • SB 1159 clarifies that AI systems and other nonhuman entities are not “persons,” members of the public, or other participants entitled to exercise rights under several California laws governing public agencies, including the Brown Act, Public Records Act, Administrative Procedure Act, and CEQA.
  • AB 2392 establishes AI-related training and procurement requirements for California’s public higher education institutions.
  • SB 951 requires additional disclosures when layoffs covered by Cal/WARN result from AI or automation. Because Cal/WARN generally does not apply to California public agencies, however, these new requirements will not affect public agency employers.

What Public Agencies Should Do Now Before the New Rules Take Effect

California’s new laws do not prohibit public agencies from embracing AI and emerging workplace technologies, but they do require meaningful human involvement in consequential employment decisions and place new guardrails around surveillance of employees.

Here are some steps public agencies can take now:

  • Inventory AI-enabled and automated technologies;
  • Identify systems that collect, analyze, or draw conclusions from employee information;
  • Determine whether automated outputs influence discipline or termination;
  • Establish human-review and documentation protocols for SB 947; and
  • Evaluate bargaining obligations before implementing technology affecting represented employees.

Technology may change how work gets done, but it does not change the importance of sound judgment, fair processes, thoughtful labor relations, and personnel policies that keep pace with the evolving workplace. Public agencies should review and update their policies and procedures to ensure they address the use of AI, automated decision-making, and workplace surveillance consistent with these new requirements. Reach out to your trusted legal advisors for assistance.

On September 20, 2026, Governor Gavin Newsom vetoed Assembly Bill 1383 (“AB 1383”), which would have made some of the most significant changes to the Public Employees’ Pension Reform Act of 2013 (“PEPRA”) since its enactment. Among other changes, AB 1383 would have increased the pensionable compensation limit for PEPRA members, reduced the benefit age factor to age 55 for PEPRA safety members, and authorized an additional safety retirement formula that could have been adopted through collective bargaining.

In his veto message, Governor Newsom pointed directly to the fiscal consequences. He explained that AB 1383 would partially reverse PEPRA reforms and significantly increase costs for state and local governments. He also expressed concern that increasing benefits for safety employees could widen the retirement-benefit gap between safety and non-safety members, create pressure for additional benefit enhancements, and increase risk to public retirement systems. Recalling the fiscal pressures that led to PEPRA in the first place, the Governor cautioned against unwinding its reforms.

As a result of the Governor’s veto, AB 1383’s proposed pension changes will not take effect. For public employers concerned about rising pension costs, that is welcome news. But it does not mean pension costs are going away—or that agencies should lighten up on pension planning.

Don’t Put Away the Calculator Just Yet

Pension costs remain a significant and recurring component of public agency budgets. While AB 1383 will not add new costs, public agencies still have several tools available under existing law to manage their long-term retirement costs. These tools involve taking a strategic look at how compensation is structured, how pension costs are allocated, and how retirement costs are considered at the bargaining table.

Here are few good places to start.

Rethink the Role of Defined Contribution Plans.

    Many public agencies provide employer contributions to deferred compensation or other defined contribution plans as part of their overall compensation packages. Those contributions should not be viewed in isolation from pension strategy.

    Unlike salary and other forms of pensionable compensation, employer contributions to defined contribution plans do not increase pensionable compensation. PEPRA permits qualifying employer contributions to defined contribution plans for compensation above the statutory pensionable compensation limit, subject to applicable statutory and federal limits.

    That makes defined contribution plans worth considering as part of an agency’s overall compensation strategy, particularly for employees whose compensation approaches or exceeds the PEPRA cap. Directing a portion of future compensation increases into defined contribution plans rather than pensionable compensation can help slow the growth of pension costs while still providing employees with additional compensation for retirement.

    Agencies that already make employer contributions to these plans should periodically revisit both the amount and structure of those contributions as part of their broader compensation and bargaining strategy.

    Rebalance Pension Cost Sharing.

    Employees already pay a portion of their pension costs. Depending on an agency’s existing agreements and retirement structure, there may be opportunities to revisit how those costs are allocated.

    • Employer-Paid Member Contributions. For classic members, employers may pay some or all of the employee’s required member contribution, commonly referred to as Employer-Paid Member Contributions (“EPMC”). EPMC is prohibited for PEPRA members. Although many agencies have already negotiated reductions or elimination of EPMC for classic members, agencies that still provide it may wish to revisit the practice as part of future bargaining.
    • Employee Cost Sharing. Employers and employee organizations may also agree in writing for employees to pay a portion of the employer contribution. Cost-sharing arrangements can provide a meaningful mechanism for controlling employer pension expenditures.

    Because cost-sharing arrangements are negotiated, agencies should think beyond the immediate contribution rate. An agreement should address what happens if the cost-sharing provision later expires, terminates, or is withdrawn by one of the parties, including whether corresponding compensation adjustments should automatically be triggered, in order to preserve the parties’ underlying economic bargain.

    Agencies may also consider negotiated cost-sharing mechanisms that adjust as actuarially determined employer contribution rates change. The goal is not simply to shift today’s costs, but to develop a structure capable of responding to future costs as well.

    Structure Pensionable Compensation with Pension Costs in Mind.

    Pension costs are driven not only by contribution rates, but also by the reportable compensation on which those contributions—and ultimately retirement benefits—are calculated. Controlling reportable compensation is an important component of pension-cost management.

    Agencies should periodically review specialty pays, educational incentives, assignment differentials, and other premium pays to determine whether and how they are reportable to the retirement system. Even small structural decisions can compound over time. For example, an agency might negotiate specialty pay as a fixed dollar amount rather than a percentage of salary. A fixed amount does not automatically increase each time base salary increases, helping control the long-term growth of both compensation and associated pension costs.

    Lump sum payments may also provide flexibility in compensation negotiations. For PEPRA members, the statutory definition of pensionable compensation excludes several categories of compensation, including certain one-time or ad hoc payments. CalPERS reporting rules for classic members are different and more complex, and certain off-salary-schedule payments may be reportable only when specific regulatory requirements are satisfied.

    Keep Calm and Call Your Actuary.

    Changing pension benefits involves more than reaching agreement at the bargaining table.

    Government Code section 7507 establishes procedural requirements that apply before a public employer approves certain changes to pension or other postemployment benefits such as retiree health benefits. Among other things, the employer must obtain an actuarial analysis estimating the fiscal impact of the proposed change and publicly disclose and consider that information before adoption. Following approval, the agency’s chief executive officer must acknowledge in writing that they understand the current and future costs of the benefit change.

    Those requirements take time. Agencies contemplating pension changes should therefore build actuarial review and the section 7507 process into their bargaining timeline rather than waiting until a tentative agreement is ready for approval.

    A Pension Planning Checklist

    Before proposing, negotiating, or adopting pension changes, public employers should consider this due-diligence checklist:

    1. Know your workforce. Gather current demographic information and employer contribution rates for each bargaining unit.
    2. Know your numbers. Review the agency’s most recent actuarial valuation.
    3. Know your objective. Develop preliminary bargaining goals before proposals are exchanged.
    4. Model before you bargain. Obtain actuarial modeling early enough to inform negotiations—not merely approve the result.
    5. Look at the whole compensation package. Consider whether compensation adjustments can offset or mitigate long-term pension costs.
    6. Get everyone to the table early. Coordinate labor relations, finance, human resources, and retirement counsel before positions harden, and leave time for required actuarial analysis and presentation.
    7. Leave room for the unexpected. Budget conservatively until the long-term fiscal impact is understood.

    The Bottom Line: Pension Planning Never Retires.

    AB 1383’s veto gives public employers breathing room. It does not provide a pension-cost holiday. The proposed increases to the PEPRA compensation cap and safety retirement benefits will not take effect, but the larger fiscal challenge remains. Pension obligations are measured in decades, and decisions made during a single round of bargaining can affect agency budgets long after the MOU has expired.

    AB 1383 may be gone, but the pension-cost conversation is not. Public employers can continue to use cost-sharing arrangements, thoughtful compensation design, defined contribution plans, and actuarial planning to better understand and manage their long-term obligations. For assistance with pension strategy, compensation planning, or labor negotiations, reach out to your trusted legal advisors.

    An important part of the litigation practice is appellate law.  One side can win in the trial court – by a motion to dismiss, on summary judgment, or after a jury trial – only to have the result overturned on appeal.  The court of appeal can send the parties back for an entirely new trial, or in some circumstance, it can decide that the party who lost at trial should actually win the case altogether.  Also, the court of appeal can publish its decision, meaning that the decision will serve as binding law for future cases raising the same issues.  Thus, a published appellate decision can have far-reaching effects for the industry or administrative area involved.  In addition, published appellate decisions often draw media attention, thus further raising the stakes.

    In appeals, a party’s written briefing can serve as its sole opportunity to present arguments to the court and influence the court’s decision.  The parties present the appeal to a panel of three justices.  These three individuals decide the matter based only on the paper record from the trial court to determine if the court committed any errors that had a sufficient likelihood of affecting the result.  They do not hear any witness testimony, and they do not accept any additional evidence.  The court of appeal does often hold an oral argument, which is a hearing where the attorneys can argue the appeal in person.  But the hearings tend to be relatively short and are often taken up with the attorneys responding to questions from the justices (and responding to the questions may or may not serves as means for the attorneys to convey their key arguments).

    This all shows the importance of effective appellate briefs.  Below are six tips lawyers follow for preparing briefs on appeal.

    1. Be accurate: The appellate brief’s citations to the trial court record, and to applicable legal authorities must be exact.  Accuracy is a requirement for all legal briefs in any court, but for appeals, the stakes can be higher.  If the brief contains an accidental mis-statement, the other side can easily make accusations that the party that presented the brief has tried to mislead the court, create confusion, or lacks credibility.  The appellate court may agree with these contentions and respond accordingly.  Even if it does not, a lawyer’s need to respond to such contentions puts his or her side on the defensive.
    2. Be complete: It is important to make all available arguments that have a sufficient chance of success on appeal.  If the party’s first brief does not make a particular legal argument, the appellate court can consider it waived.  It will be difficult to make the argument for the first time at oral argument before the court of appeal, in subsequent briefing, or to a higher court like the California Supreme Court.
    3. Be clear and guide the court through the decision making sought: This applies both to sentence and paragraph structure and the overall organization of the brief.  Briefs should set forth, in a logical and clear way, the legal structure the court must assess, and how the facts presented in the record fit into that structure.  Briefs will be organized under separate point headings (different items in the table of contents) so as to make it absolutely clear which elements of law apply to which items of evidence.  What about addressing the other side’s arguments?  The brief can group the arguments at the end of the analysis section to which they relate and then restate and refute them in sequence, with typically one argument per paragraph.  This systematic approach constitutes the same approach the court takes preparing its decision, and can provide the court with an analysis it can more or less adopt if it sees fit.
    4.  Apply case themes:  In preparing a brief, attorneys often find that a particular fact, legal principle, or perspective will actually refute many of the other side’s arguments.  The attorneys will develop this into a case theme, something carefully crafted to be repeated in various ways throughout the briefing to keep it at the forefront of the justices’ perceptions.  Often, for consistency, it makes sense for the appeal brief  to include the same case themes as in the trial court.  On appeal, however, lawyers usually add themes that have a more technical dimension, meant to draw on the justices’ interest in accurately applying and developing the law rather than relying on themes based on more general concerns intended to persuade a jury.  Either way, a case theme on appeal can demonstrate to the court of appeal that it can resolve the whole matter by relying on one or two core principles or by making a few key rulings.
    5. Temper your invective: Lawyers sometimes fill their briefs with harsh, accusatory language against the other side or the other side’s lawyers.  They may label arguments made by their opposing counsel “ridiculous,” “bad faith,” “ignorant,” or the like.  But this type of invective is well-known to irritate courts, and even terms like “frivolous” or “bad faith” are thought to have the same effect if they sound perfunctory, and made without any effort actually to single out for the court arguments or conduct by the other side that are particularly outrageous.  Indeed, some appellate attorneys – in appropriate cases – choose to have their briefing not say anything particularly negative about the other side.  Instead, the briefing will simply explain cogently why, under applicable law and the evidence in the record, the other side cannot win the case.  This makes the brief appear more objective and appellate justices may find it easier to rule in favor of the side that takes a more measured tone.
    6. Tell the client’s story: Often both sides experienced the trial court litigation as an emotional saga that took a heavy toll.  It may turn out that the appeal, however, involves only a few more technical issues (e.g., jurisdiction, sufficiency of the evidence on monetary damages, evidentiary rulings on expert witnesses, etc.).  In such cases, the parties may well expect their lawyers nevertheless to write briefs that contain the whole narrative, an emphatic description of why the other side’s conduct was wrongful, and a impassioned explanation of why their side behaved properly and deserves vindication.  It does not help for briefs to include substantial matter irrelevant to issues on appeal.  At the same time, it is common for a brief to offer the court of appeal a context for the decision the court will make.  It is best to find a way to present this context in the appellate briefing, and tell the client’s story succinctly in the process.  This may mean that the brief will include notifications to the court that parts of the discussion serve as this kind of background.  Such matter can have some emotional impact, draw on a sense of fairness, and influence the court.

    We will continue to prepare updates on appellate law, and on litigation in general.

    It is important to note that this post was originally published December 10, 2019, but has now been updated to reflect today’s current information.

    Mental health crises are sensitive, and employees experiencing them should be treated with empathy and care. They also present employers with the challenge of coordinating workplace safety, disability and leave laws, and any related investigation or misconduct.

    If there is an immediate threat of harm, the agency should follow its emergency and workplace-safety protocols. Once the immediate danger has passed, the following principles can guide the agency’s response.

    When a Fitness for Duty Examination Is Appropriate

    When an employee exhibits signs of mental instability or impairment, it may be appropriate for the agency to require the employee to undergo a fitness for duty examination. Under the Americans with Disabilities Act (ADA), an employer may require a medical or psychological examination only when it is job-related and consistent with business necessity. This standard can be met when objective evidence creates a reasonable belief that a medical condition, including a mental health issue, impairs the employee’s performance of essential job functions or causes a direct threat to the employee or others in the workplace.

    For example, in Kao v. University of San Francisco (2014) 229 Cal.App.4th 437, the court upheld a fitness for duty examination where a professor frightened coworkers through episodes of rage, clenched fists, and a “wild cackling laugh.”

    An agency should not require a fitness for duty examination merely because an employee is difficult, emotional, unconventional, or inefficient. The justification for an examination is often stronger, however, for safety-sensitive positions involving firearms, emergency response, heavy equipment, or driving.

    Engage in the Interactive Process Early

    A mental or psychological disorder may qualify as a disability under the ADA or California’s Fair Employment and Housing Act (FEHA), thus triggering the employer’s obligation to engage in the interactive process to identify reasonable accommodations. The agency should offer to begin the interactive process with an employee when it becomes aware of a possible need for accommodation through the employee themselves, a third party, observation, or the employee’s exhaustion of other leave.

    Potential accommodations may include modified communication methods, adjustment of nonessential duties, reassignment to a vacant position, or a leave of absence. Reassignment to a different supervisor—even if the existing supervisor allegedly caused or exacerbated the mental health crisis—is generally not a reasonable accommodation. Employers retain authority over personnel assignments and reporting structure, although it may be appropriate to instruct the existing supervisor to change their methods.

    Leave related to a mental health condition should also be coordinated with the Family and Medical Leave Act (FMLA) and California Family Rights Act (CFRA). When an absence qualifies under these laws, the agency generally may designate the leaves to run concurrently after providing required notices. Exhaustion of FMLA or CFRA leave does not necessarily end FEHA or ADA obligations; additional leave may still be a reasonable accommodation absent undue hardship.

    Do Not Lose Sight of Workplace Investigations

    It is not uncommon for an employee experiencing a mental health crisis to be on the complaining or receiving end of allegations of harassment, discrimination, retaliation, or other policy violations.

    The agency’s obligation to investigate such complaints is not contingent upon the complainant’s mental health status. If an employee’s allegations would violate policy if true, the agency should not dismiss them because it suspects that the employee’s perceptions may be affected by a mental health condition. The agency should follow its normal processes for investigating complaints.

    If the complainant is on medical leave, the agency may provide the employee with the option to participate voluntarily in an interview with the investigator, or to hold the investigation in abeyance until the employee returns from leave. The agency may still preserve evidence, interview other witnesses, and take interim safety measures.

    If the employee on leave is the subject of the investigation, they may participate voluntarily in an interview while on leave, but the agency cannot compel them to do so. If the subject is a peace officer or firefighter, a delay in the ability to interview the respondent might require an agreement with the employee to toll the Public Safety Officers Procedural Bill of Rights Act (POBR)/Firefighter Bill of Rights Act (FBOR) one-year statute of limitations for discipline.

    Pause Before Disciplining Disability Related Conduct

    In Dark v. Curry County (9th Cir. 2006) 451 F.3d 1078, the U.S. Court of Appeals for the Ninth Circuit held that, with limited exceptions, disciplining an employee for conduct resulting from a disability is the functional equivalent of unlawfully disciplining them for their disability. In Dark, the county terminated an equipment operator with epilepsy after he experienced a seizure while driving a county vehicle. The court held that the county was obligated to engage in the interactive process, including considering whether leave or reassignment would mitigate the employee’s symptoms, before proceeding to discipline.

    The Dark court recognized two exceptions to this principle: (1) conduct that is the result of illegal drug use and alcoholism, and (2) egregious and criminal conduct.  

    The practical lesson of the Dark case is to slow down before imposing discipline when an employee’s conduct or performance issues may be disability related. If problems persist after reasonable accommodation efforts—or no effective reasonable accommodation would allow safe performance of essential functions—the agency may be able to proceed with appropriate corrective or disciplinary action.

    A Coordinated Response Is the Best Response

    Mental health crises rarely fit within a single legal framework. Public agencies should respond with compassion while grounding decisions in objective evidence, careful documentation, and consultation with legal counsel. That approach supports the employee while meeting the agency’s operational and legal responsibilities.

    Fire Jacket

    Governor Gavin Newsom signed Senate Bill (“SB”) 1024 into law on September 20, 2026. SB 1024 requires specified fire departments to provide up to 26 weeks of paid leave to an eligible active firefighting member who requests leave because they are disabled by pregnancy, childbirth, or a related medical condition.

    The duration of leave may depend on medical need and certification, up to a maximum of 26 weeks.  As a condition of granting or continuing the leave, a fire department may require written medical certification from the firefighter’s health care provider confirming the need for leave and stating its anticipated start date and duration.

    Firefighters may not be required to exhaust paid sick leave, vacation, or compensatory leave before or while taking SB 1024 leave. A fire department may, however, count certain benefits payable under another law, employer-provided paid-leave policy, or disability-insurance plan toward its obligation, provided the firefighter receives full pay during the leave without a reduction in vacation, sick, or other compensated-leave balances. The firefighter earns full retirement service credit during the leave and pays the required member contributions; required employer retirement-fund contributions are included in the required compensation.

    The leave must be paid at the firefighter’s regular rate of pay set forth in the applicable collective bargaining agreement and in accordance with the firefighter’s normal pay schedule. It runs concurrently with job-protected Pregnancy Disability Leave (“PDL”) and/or leave provided as a reasonable accommodation for pregnancy-related disability.

    All benefits continue to accrue during the leave as if the firefighter were actively working, including retirement contributions, seniority, promotional eligibility, and step increases. Group health coverage must also continue for the duration of the leave at the same level and under the same conditions that would have applied had the firefighter remained working. Upon return, the firefighter must be restored to their prior position or a position of equivalent rank, pay, schedule, station assignment, and promotional trajectory.

    For part-time firefighters who work a fixed number of hours per week, weekly pay must reflect the total number of hours they are normally scheduled to work. For part-time firefighters without fixed weekly hours, weekly pay must equal their average weekly pay during the six months immediately before the leave began.

    To qualify for SB 1024 paid leave, an active firefighting member must have at least 1,250 hours of service with the fire department during the 12 months before the leave begins. That threshold does not limit an employee’s right to unpaid, job-protected PDL.

    SB 1024 does not separately define what it means to be “disabled by pregnancy.” Existing PDL regulations define that phrase to include circumstances in which, in a health care provider’s opinion, an employee cannot perform one or more essential job functions because of pregnancy or cannot do so without undue risk. (Cal. Code Regs., tit. 2, § 11035, subd. (f).) The regulations also identify pregnancy- and childbirth-related conditions that may qualify, including prenatal or postnatal care, childbirth, recovery from childbirth, pregnancy loss, gestational diabetes, preeclampsia, and postpartum depression. It is not yet certain whether a court will apply that regulatory definition to SB 1024. However, SB 1024 uses the same phrase and expressly provides that its paid leave runs concurrently with PDL. Fire departments should therefore anticipate that the established PDL definition may inform the new statute’s disability requirement.

    SB 1024 does not preempt or limit collective bargaining agreements or state or local laws or policies that provide greater rights or benefits. It applies to specified public fire departments, including those operated by cities, counties, districts, the California State University, CAL FIRE, and county forestry or firefighting departments or units. A University of California fire department is requested, rather than required, to comply.

    LCW will continue to monitor developments concerning SB 1024, including any guidance or decisions addressing its disability requirement, and will keep clients informed of material updates.

    We are excited to continue our video series – Tips from the Table. In these videos, members of LCW’s Labor Relations and Collective Bargaining practice group will provide various tips that can be implemented at your bargaining tables. We hope that you will find these clips informative and helpful in your negotiations.

    A key witness agrees to talk, then disappears. Emails go unanswered. A promised list of available times never arrives. Meanwhile, the investigation clock keeps running, and the parties wait for an answer.

    A nonresponsive witness can complicate an investigation, but it does not automatically bring the process to a stop. The investigator’s job is to make reasonable efforts to obtain the information, protect the employee’s rights, gather the best evidence available, and explain any limits in the final findings.  The following are some recommendations both for investigators and for those agency employees charged with coordinating the investigation.

    1. Start by diagnosing the silence

    Before determining a witness “uncooperative,” try to find out whether the problem is practical, legal, or personal. The witness may be on leave, working a different schedule, worried about retaliation, unsure whether the interview is mandatory, or waiting for a union representative. A former employee or outside witness may simply have little incentive to respond.

    One clear, respectful message can often solve the problem. Identify the investigator, explain the subject in neutral terms, state that the person may have relevant information, offer reasonable scheduling options, and provide a deadline to respond. Avoid promising complete confidentiality or that the interview is “necessary” for the investigation to be completed. Instead, when asked, generally respond that information will need to be shared on a need-to-know, confidential basis, and  only to the extent reasonably necessary to investigate and respond to the complaint.

    2. Use a measured escalation process

    There is no universal number of contact attempts that makes an investigation “complete.” The right approach depends on the importance of the witness, the urgency of the allegations, agency policy, applicable labor agreements, and the reason for the delay. A practical sequence may look like this:

    A. Make the first request easy to answer. Offer several dates, a remote option, and a direct way to identify any accommodation, leave, or representation issue.

    B. Follow up in writing. If the witness does not respond, send a concise reminder with a reasonable deadline. Keep a record of each attempt, the method used, and any response.

    C. Clarify whether participation is expected. For a current employee, the agency may be able to issue a lawful work directive to attend and answer questions, depending on its policies, labor agreements, and the employee’s status. The investigator should coordinate with Human Resources or counsel before threatening discipline for noncooperation.

    D. Set and communicate the stopping point. Tell the witness when the investigator will proceed without the interview if there is no response. That creates a fair and final opportunity while keeping the matter moving.

    3. Check representation and public-safety rules before compelling answers

    A person who begins as a witness may become a possible subject as facts develop. For represented public employees, questioning may trigger a right to union representation when the employee reasonably believes the interview could lead to discipline and asks for a representative. If the interview changes direction, pause and reassess instead of pushing ahead under the original “witness” label.

    Additional rules apply to peace officers and firefighters when an interview could lead to punitive action. The Firefighters Procedural Bill of Rights Act and the Public Safety Officers Procedural Bill of Rights Act both address matters such as notice, timing, recording, representation, and the consequences of refusing to answer job-related questions. Agencies should confirm the correct procedure before issuing a directive or characterizing a refusal as insubordination.

    4. Build the record without the witness

    If reasonable efforts do not secure the interview, the investigator can turn back to the evidence map. The investigator can ask what the witness was expected to prove or disprove, then look for other sources: emails, texts, chat messages, calendar invitations, access logs, video, photographs, policies, work records, prior statements, or other people who observed the event or its aftermath.

    The unavailable witness may be important without being indispensable. Workplace findings are generally based on whether the evidence shows that an allegation is more likely than not to have occurred. A fair conclusion can rest on the full record, including circumstantial evidence and credibility factors. Silence by itself, however, should not be treated as proof that the witness supports one side or has something to hide.

    5. Explain the limitation, and finish the investigation

    The report should state, neutrally and accurately, the efforts made to contact the witness, any reason given for not participating, and whether the missing information limited the analysis. It should distinguish between “unsubstantiated” and “unable to determine” when agency policy uses those terms, and it should avoid speculation about what the witness would have said.

    An investigation does not have to be perfect to be fair. It should be prompt, thorough, impartial, and reasonable under the circumstances. When the record is sufficient, the investigator should make findings and close the matter. When the missing interview truly prevents a reliable finding, the report should say so and identify any follow-up that may be appropriate if new evidence later becomes available.

    6. Plan for the next silent witness

    Agencies can reduce delay by adopting investigation procedures that explain who must cooperate, how interview requests will be delivered, when representation may be requested, how leave and accommodation issues will be handled, and how noncooperation will be documented. Clear rules make it easier to act consistently when a difficult situation arises.

    We are excited to share our newest video from our series – Wage & Hour Issues in the Workplace. In these videos, members of LCW’s Wage & Hour practice group will provide various tips that can be implemented in your workplace. We hope that you will find these clips informative and helpful!

    Our archive of Wage and Hour Issues in the Workplace videos is available here: https://www.calpublicagencylaboremploymentblog.com/category/wage-and-hour-2/.

    Focused latin hispanic business woman working on laptop computer reading financial document trader report in office. Thinking accountant entrepreneur manager businesswoman doing paperwork using pc

    On August 6, 2026, the IRS issued Fact Sheet FS-2026-13, providing its most detailed guidance yet about the One Big Beautiful Bill Act’s (OBBBA) tax deduction for qualified overtime compensation. The new Fact Sheet identifies frequently asked questions and provides answers related to the deduction. For public agencies, the message is clear: 2026 payroll systems, overtime classifications, and W-2 reporting deserve a closer look now—not next January when reporting is due.

    What is Qualified Overtime Compensation (QOC) for the Tax Deduction?

    QOC is the amount of overtime compensation required under Section 207 of the FLSA that is in excess of a non-exempt employee’s regular rate of pay. Generally, QOC that is subject to the deduction is the “one-half” (0.5) portion of the time and one-half (1.5) paid to a non-exempt employee for overtime hours worked under the federal Fair Labor Standards Act (FLSA). For more information on the basics, see LCW’s prior blog post about the deduction.

    The Key Question: Is the Overtime Required by the FLSA?

    This is where public agencies need to pay particular attention. The IRS makes clear that the deduction applies only to overtime compensation required under Section 7 of the FLSA. Overtime paid solely because of a memorandum of understanding, a collective bargaining agreement, or an agency’s own policy does not qualify.

    Comp Time Off Creates Another Wrinkle

    Public agencies also should take note of the IRS’s guidance concerning compensatory time off (CTO). As described in LCW’s prior blog post about the OBBBA and CTO, the FLSA permits public agencies to compensate employees for FLSA overtime in the form of CTO, subject to an agreement and certain conditions and limitations. When an employee chooses to accrue CTO in lieu of overtime pay in cash, each hour of FLSA overtime worked is credited with one and one-half hours of CTO in the employee’s CTO bank. The employee may then use their CTO hours as paid time off in the future or may cash out CTO hours that go unused at the regular rate of pay upon separation (or at other times based on the terms in an MOU or other agreement).

    The IRS specifies that CTO is only reported when hours are used or cashed out, not when CTO hours are earned and banked for future use. A CTO payment can be provided when an employee takes time off using their accrued CTO hours or otherwise cashes out unused CTO hours. The IRS provides specific formulas for determining the qualifying amount.

    IRS Notice 2025-69 provides the following example illustrating how to take a CTO payment and calculate the reportable QOC amount:

    Example 6. Individual D works for a State or local government agency that gives compensatory time at a rate of one and one-half hours for each overtime hour worked under 29 USC 207(o). In 2025, Individual D was paid wages of $4,500 with respect to compensatory time off taken in accordance with section 207(o). For purposes of determining the amount of qualified overtime compensation received in tax year 2025, Individual D may include $1,500, one-third of these wages for purposes of determining qualified overtime compensation under section 225(c).

    The following is an example of how a CTO cash out payment is reported as QOC:

    CTO Cash Out Example: Mary works for a local government agency that gives compensatory time at a rate of one and one-half hours for each overtime hour worked under 29 USC 207(o). In the tax year, Mary did not take any paid time off using CTO, but pursuant to an applicable memorandum of understanding, Mary was able to cash out $6,000 of unused CTO hours. Mary may include $2,000, which is one-third of the CTO cash out payment, for purposes of determining qualified overtime compensation under section 225(c).

    For agencies with employees who have significant comp-time balances, this could create an important timing issue for payroll and year-end W-2 reporting.

    W-2 Reporting Is No Longer Optional

    Perhaps the most immediate operational change is the enforcement of the reporting requirement.  The transition relief from 2025 that made it optional for employers to report QOC is no more. Beginning with tax year 2026, employers must report QOC for non-exempt employees on Form W-2 using Box 12, Code TT.

    Importantly, the amount reported is the total QOC paid—not necessarily the amount the employee ultimately may deduct. Employers are required to report the full amount of QOC the employee received regardless of the deduction cap.  For example, an employee could have $30,000 of QOC reported in Code TT even though the OBBBA caps the amount of the overtime deduction at $12,500 ($25,000 for joint filers) and limits the deduction if the employee’s modified adjusted gross income (“MAGI”) is over $150,000 ($300,000 for joint filers). The deduction is phased out by $100 for every $1,000 of MAGI above $150,000 ($300,000 for joint filers).  See LCW’s article about the deduction cap for more information.

    Report Qualified Overtime Compensation Correctly, or Be Prepared to File Corrections

    One of the biggest takeaways from the new IRS guidance is that if the amount of QOC reported on the W-2 is wrong, employees are going to rely on employers to correct it using Form W-2c. Employees rely on their employers to report accurate amounts since they use the W-2 when calculating and claiming their deduction on Schedule 1-A (Form 1040).  The IRS provides guidance for the following scenarios:

    • Employer Discovers an Error: If an employer discovers an error made in Form W-2, Box 12, Code TT, the employer must file a Form W-2c, Corrected Wage and Tax Statement and furnish the Form W-2c to the employee as soon as possible.  An employer that files or furnishes an incorrect W2 may be subject to penalties. 
    • Employer Overstates the Amount: If an employer overstates the amount of QOC on the W-2, the employee cannot claim the overstated amount.  The employee is only entitled to the QOC amount they actually received during the taxable year.  For example, if an employee actually received $5,000 in QOC but the employer erroneously reported $10,000 on the W-2, the employee may only include $5,000 for their deduction.
    • Employer Understates the Amount: If the employee believes the employer omitted or understated the amount of QOC on the W-2, the employee must request a Form W-2c (corrected form) from their employer that properly reports the accurate amount.  This is because any amount not reported on the W-2 may not be considered for purposes of the deduction.
    • Employer Refuses to Correct the Understated Amount: Although employers are required to correct inaccurate W-2 information, if an employer is unwilling or unable to provide an employee with a Form W-2c with the higher corrected QOC amount, the employee is only entitled to use the lower/understated QOC amount reported on the W-2 to determine their deduction. This may lead to complaints raised by the employee or an employee association. For example, if an employer paid an employee $10,000 of QOC but only reported $5,000 on the W-2, and then refuses to furnish a corrected W-2c upon the employee’s request, the employee is only allowed to include $5,000 of QOC for their deduction.
    • Employee Cannot Self-Correct the Understated Amount: If the employer is unwilling or unable to provide the employee with a Form W-2c to correct the amount, the employee cannot take matters into their own hands by using a substitute W-2 (Form 4852). The OBBBA only allows deductions based the amount of QOC received during the taxable year and reported on the W-2. Form 4852 does not satisfy that requirement since it is not a form furnished by the employer. In this type of situation, the employee only entitled to use the lower/understated QOC amount reported on the W-2 to determine their deduction.

    What Should Public Agencies Do Now?

    1. Audit payroll coding. Make sure the system can distinguish FLSA-required overtime from additional overtime paid under an MOU or agency policy.
    2. Prepare for 2026 W-2 reporting. Payroll vendors and internal systems should be configured to separately track QOC for Code TT reporting.
    3. Be prepared in case employees request corrections to the QOC amounts. Have a plan to check QOC calculations and be responsive to employee requests.

    The IRS FAQs are guidance, not regulations, and the IRS expressly cautions that the FAQs may be revised and will not be relied on or used by the IRS to resolve a case. Still, they provide an important roadmap for employers as the OBBBA overtime deduction reporting gets underway.

    Governmental 457(b) plans are a familiar part of the benefits package at many public agencies. Employees can defer a portion of their compensation and into the plan for retirement, often as a supplement to a defined benefit pension. It is tempting to think of a governmental 457(b) plan as the public sector version of a 401(k).

    That is a useful shorthand, but the comparison has its limits. Governmental 457(b) plans operate under Internal Revenue Code section 457 and are subject to rules that differ in important respects from those governing 401(k) plans and other retirement plans. Some of those differences create opportunities for employees, while others can create administrative problems when an agency, employee, or payroll department assumes the rules work the same way.

    Here are five differences public agencies should keep in mind.

    1. The Contribution Limit Stands on Its Own

    The annual contribution limit is a good example of how familiar numbers can obscure different rules. For 2026, the basic contribution limit for a governmental 457(b) plan is $24,500, or 100 percent of the participant’s includible compensation if less. That happens to be the same dollar amount as the 2026 elective deferral limit applicable to 401(k) and 403(b) plans.

    Despite the identical dollar limits, the 457(b) limit is generally separate from the limit that applies to 401(k) and 403(b) plans. An employee who contributes to a 403(b) plan, for example, may also contribute up to the applicable annual limit to a governmental 457(b) plan. Subject to the terms and limits of the respective plans, an employee with access to both may contribute up to the applicable annual limit to each. That can make a 457(b) particularly valuable for public employees seeking to increase their retirement savings.

    The rule is different, however, for contributions made to the 457(b) plan itself. Employer contributions generally count toward the same annual 457(b) limit as the employee’s salary deferrals. For example, an agency that contributes $2,000 to an employee’s 457(b) account generally reduces by $2,000 the amount the employee may defer from compensation for that year.

    2. Employees Approaching Retirement May Have Another Catch Up Option

    The age based catch up rules changed significantly under the SECURE 2.0 Act, passed by Congress in December 2022 . Prior to the SECURE 2.0 Act, any participant age 50 or older was able to contribute an additional amount to their retirement, with the limit set at a uniform dollar amount for participants. In 2026, participants age 50 or older may be eligible to contribute an additional $8,000. Participants who attain ages 60, 61, 62 or 63 during the year may qualify for an even larger catch up contribution of $11,250.

    A governmental 457(b) plan may also provide a different catch up opportunity under Internal Revenue Code section 457(b)(3). During the final three taxable years before reaching the plan’s normal retirement age, an eligible participant may be able to make additional contributions based on amounts the participants could have deferred, but did not defer, in prior years of participation. This is sometimes described as allowing an employee to contribute up to twice the normal annual limit. While that can be true, it is not an automatic doubling of the contribution limit. The amount available depends on the participant’s contribution history. Suppose an employee is two years from the plan’s normal retirement age and wants to maximize contributions before retiring. The employee contributed well below the applicable limit during several earlier years of participation. Depending on the plan terms and the employee’s contribution history, some of that unused capacity may now be available under the special catch up. By contrast, an employee who consistently contributed the maximum may have no unused capacity available for the special catch up.

    The special rule also creates an administrative issue that requires more than determining the employee’s age. The employee cannot use the special three-year catch up and the age based catch up in the same year. If both are potentially available, the participant may contribute under the provision that permits the greater contribution. The IRS has specifically identified administration of this special catch up as an area in which 457(b) plan errors occur. For agencies that offer it, accurate historical contribution records matter.

    3. Age 59½ Does Not Have the Same Significance

    One of the more notable differences between a governmental 457(b) plan and a 401(k) plan appears when an employee takes a distribution.

    Generally, a distribution from a governmental 457(b) plan is not subject to the 10 percent additional federal tax on early distributions that generally applies to distributions from other retirement plans before age 59½. For example, a public employee who separates from employment and takes an otherwise permissible 457(b) distribution before reaching age 59½ generally does not incur the additional tax solely because of age.

    There is an important exception for amounts that were rolled into the 457(b) plan from certain other retirement plans or accounts. If a participant previously rolled assets from a qualified retirement plan, 403(b) plan, or IRA into the governmental 457(b) plan, distributions attributable to those rollover amounts may still be subject to the 10 percent additional tax.

    Employees should also be careful when considering a rollover out of the 457(b) plan. Moving governmental 457(b) assets into an IRA or another retirement plan can affect how a later distribution is treated. If the assets are rolled into an IRA or an employer plan that is not a governmental 457(b) plan, a later distribution before age 59½ may be subject to the 10 percent additional tax unless an exception applies. An employee who expects to need access to retirement funds before age 59½ may therefore want to understand the potential tax consequences before completing a rollover.

    4. “457(b)” Does Not Always Mean Governmental 457(b)

    Public agencies occasionally encounter guidance that refers broadly to “457 plans.” That description can obscure an important distinction.

    Section 457(b) plans may be maintained by state and local governments, but they may also be maintained by certain tax exempt organizations. Governmental and nongovernmental 457(b) plans are subject to important differences.

    For governmental plans, Internal Revenue Code section 457(g) requires plan assets and income to be held in trust for the exclusive benefit of participants and their beneficiaries. The requirement may also be satisfied through qualifying custodial accounts or annuity contracts. By contrast, assets set aside to fund a nongovernmental 457(b) plan generally remain the property of the tax-exempt employer and are subject to the claims of the employer’s general creditors. 

    This distinction is especially important when researching plan questions or responding to employee inquiries. Internet searches for “457(b) distribution rules” or “457(b) rollover rules” can produce information about both governmental and nongovernmental plans. Agencies should confirm that the guidance they rely on actually addresses the type of plan they sponsor.

    5. Outsourcing Administration Does Not Outsource Every Responsibility

    A public agency may have an experienced deferred compensation provider administering its 457(b) plan. Employees may enroll through the provider, select investments on the provider’s website, request distributions from the provider, and call the provider when they have questions. From an employee’s perspective, the provider may appear to “run” the entire plan.

    However, the agency remains involved in important aspects of plan administration. The agency still has to implement employee deferral elections through payroll and timely transmit contributions. Employment and compensation information may need to be provided to the recordkeeper. The agency and provider also need to understand who is responsible for determining eligibility for particular plan features, and the plan must ultimately operate in accordance with its written terms and Internal Revenue Code section 457.

    Consider an employee using the special three year catch up. The recordkeeper may maintain the account, but calculating the employee’s permitted contribution can require historical information about prior eligibility and contributions. If the recordkeeper does not have accurate information, the contribution limit can be calculated incorrectly.

    Contribution elections present another potential source of error. For example, an employee changes a contribution election, but the change is entered incorrectly and payroll contributions eventually exceed the employee’s applicable annual limit. The IRS updated its guidance concerning excess governmental 457(b) deferrals in July 2026. When a governmental 457(b) plan exceeds the applicable plan limitation, the excess deferral, together with allocable net income, must be distributed as soon as administratively practicable after the plan determines that there is an excess deferral.

    These situations are good reasons for agencies to periodically examine how responsibility is divided among benefits personnel, payroll, and the plan provider. An agency should know who monitors contribution limits, who determines eligibility for special catch up contributions, what information the provider receives from payroll, and how the parties communicate when an error is identified.

    A service agreement can assign administrative tasks. It cannot eliminate the need for accurate information, clear allocation of responsibilities, and effective coordination .

    Know Which Rules Apply to Your Plan

    These differences are not merely technical. Some can provide meaningful advantages to public employees. The separate 457(b) contribution limit may allow an employee participating in another type of retirement plan to save substantially more. The special three-year catch-up can help some employees increase savings as they approach retirement. And the different treatment of governmental 457(b) distributions under the 10 percent additional tax may be particularly important for employees who leave public service before age 59½.

    At the same time, not every governmental 457(b) plan offers every feature permitted by federal law. The Internal Revenue Code establishes the framework, but the agency’s plan document determines which optional features are actually available. An agency should therefore review its own plan terms before advising an employee that a particular catch up, distribution option, Roth feature, or other provision is available.

    Recent changes make that review particularly timely. SECURE 2.0 increased catch-up contributions for participants who attain ages 60 through 63 during the year. Beginning in 2026, participants whose prior year wages from the employer sponsoring the plan exceed the applicable threshold generally must make age based catch up contributions on a Roth basis. Those changes may require coordination among the agency, payroll, and the agency’s 457(b) provider.

    Governmental 457(b) plans can look deceptively familiar. Understanding the rules that make them different from other retirement plans, and the administrative responsibilities those rules create, can help public agencies administer their plans more effectively and provide employees with more accurate information concerning an important part of their retirement benefits.