Employers now have an additional thing to worry about: “pro se” (or simply put, plaintiffs representing themselves) employment lawsuits have nearly doubled in the last few years, mainly as a result of platforms like ChatGPT. ChatGPT and other language learning models (LLMs) have made it extremely easy for individuals to launch lawsuits against their employers. They can easily draft lengthy complaints and draft responses to requests made by defense teams or the courts in record time. Additionally, these pro se lawsuits have become harder to fight, given the deference the courts give pro se plaintiffs. In the age of rapidly advancing AI and LLMs, what measures can employers take to protect their companies from potentially frivolous and costly lawsuits?

Review proprietary AI systems

Does your company have a proprietary AI chatbot that employees can use? If so, you may need to review the information these chatbots are giving your employees. One thing that we have learned as AI use has become more prevalent is that LLMs very rarely antagonize a user, meaning they largely reaffirm and confirm a user’s beliefs. Unfortunately, this has resulted in tragic outcomes in certain instances. In other cases, it has resulted in folks being convinced they have a claim against a company or their employer. This has triggered an influx of new pro se lawsuits that are often supported by documentation that has been almost entirely drafted by some type of AI.

Now, you might be thinking, “If an AI system is proprietary to a company, it would never provide information that may harm the company, right?” It’s not that simple. LLMs are not actually built to verify facts or provide objective truth. Instead, they are designed to provide “plausible-sounding answers.”1 Let’s consider this scenario: An employee asks a workplace’s chatbot if they have a legal claim against the company because a manager put them on a performance improvement plan. The employee tells the chatbot that this feels like retaliation because some time ago, the employee had complained about having to work late. The chatbot, in line with its design, tells the employee that, yes, they do in fact have a retaliation claim and can file a lawsuit. Armed with this knowledge, given to them by the company’s own chatbot, the employee goes on to file a lawsuit against their employer for retaliation, submitting a lengthy claim. The company’s general counsel then spends countless hours reading through the claim and providing discovery for the case, only to find out that it was their own technology that had caused this lawsuit to begin with. Does this mean the employee will win the case? No, but it could potentially make their case stronger and result in costly legal fees for the employer.

This is all to say: it is imperative that employers are consistently monitoring their AI chatbots or any type of proprietary AI that employees have open access to and ensure it is not introducing legal risk.

General counsel should look out for flags that a claim is AI-generated

Companies with general counsel should closely review lawsuits for any signs of AI-generated content. Some signs a claim may be AI-generated are the length of the claim, the citations provided in the claim, and even the speed at which requests are responded to.2 This has significantly increased the work for general counsel as responses that used to take days or weeks are now taking mere minutes, giving them less time to properly prepare a defense.

Some signs that a claim is AI-generated are quite obvious, such as erroneous or completely fabricated citations. In California, an attorney was fined $10,000 after a court discovered an appeal they had filed was full of fake quotations. Of the 23 quotes that were included, 21 of them were entirely made up by artificial intelligence.3 These are conspicuous signs that general counsel can look out for and can develop a strategy to combat.

A few signs that might not immediately indicate a claim is AI-generated but are important to consider are the length of the lawsuit, the speed at which filings are submitted, and the cadence of the language in the claim. Super lengthy claims with overly verbose definitions of laws are typically AI-generated. Additionally, if a lengthy claim or response was filed within minutes of a response being requested, that’s typically a warning sign that the plaintiff is using AI. The language that is being used in a claim can also reveal whether a claim was generated by AI. In my experience, the cadence and voice employed by LLMs and AI are typically overly confident, presenting statements as facts without providing supporting evidence. If a claim seems to have overly confident, bordering on aggressive, language, it is likely it has been generated by AI.

Pro se plaintiffs typically don’t have favorable outcomes

I know that some employees might see this trend of using LLMs and artificial intelligence to construct legal claims as the great equalizer, as they no longer need to spend thousands on legal representation. Historically, however, pro se plaintiffs have had unfavorable outcomes, with some statistics showing only 3% of pro se plaintiffs winning final judgments.4 Obviously, pro se plaintiffs don’t have the benefit of large legal teams that know the ins and outs of the legal process, which shrinks their likelihood of success. Additionally, pro se plaintiffs don’t have the legal experience to formulate a strong legal strategy, something artificial intelligence also can’t help them with. Overall, AI can only really help with boilerplate language and paperwork, and even then, it doesn’t do it very well, as its tendency is to focus on definitions of the law over sound legal arguments.

Conclusion

Although pro se litigation against employers has doubled in the last few years, employers have tools at their disposal to combat these claims. Employers can work with general or outside counsel to develop strategies to identify AI-generated claims that may be dismissed by the courts. Additionally, employers should closely monitor any type of proprietary AI that is easily accessible to employees and ensure it does not introduce legal risk.

If you require assistance navigating the complex AI landscape, please do not hesitate to contact us.

California recently enacted Assembly Bill 692, which went into effect on January 1, 2026. This new law significantly limits the use of “stay-or-pay” provisions and training repayment agreement provisions (TRAPs) in employment contracts. For employers, this means it’s time to take a closer look at any agreements that require repayment if an employee leaves the organization.

If your organization has historically required employees to repay training costs, relocation expenses, sign-on bonuses, or similar payments under certain circumstances, AB 692 may directly impact those arrangements. Below are some of the key provisions employers should understand:

• Targets “stay-or-pay” / training repayment agreement provisions (TRAPs) tied to continued employment or a work relationship with a specific employer.

• Makes it unlawful to require a worker, as a condition of employment or work, to agree to pay a “debt” if the relationship ends (e.g., training, education, relocation, immigration costs, certain bonuses).

• Prohibits provisions that allow an employer, training provider, or debt collector to start or resume collection, or end forbearance on a debt, when employment ends.

• Bans contract terms that impose any penalty, fee, or cost on a worker because the employment or work relationship ends (including quit fees, liquidated damages, lost profits, replacement/retraining fees).

• Defines covered individuals broadly as “workers,” which includes employees and prospective employees, and potentially other categories depending on interpretation.

• Declares covered repayment/penalty provisions void and unenforceable in contracts covered by the statute.

Although AB 692 restricts some stay-or-pay provisions, employers are still allowed to seek repayment in the following cases:

• Governmental loan repayment or forgiveness programs administered by federal, state, or local agencies are allowed.

• Repayment of tuition for a transferable education credential is permitted, subject to conditions (e.g., no repayment if separated except for misconduct).

• Certain apprenticeship program arrangements approved by the Division of Apprenticeship Standards are carved out.

• Certain upfront, discretionary sign-on bonuses or similar payments at the outset of employment can include prorated, interest-free repayment terms in a separate agreement if strict safeguards are met (e.g., max two-year retention period, no repayment if terminated except for misconduct, clear disclosure and 5 business days to consult counsel).

In addition to limiting certain contract provisions, AB 692 also establishes meaningful enforcement mechanisms and remedies. Employers who fail to comply, or who continue using agreements that violate the law, may face significant exposure. Key enforcement provisions include:

• AB 692 creates a private right of action for workers and, in some cases, worker representatives, including representative actions on behalf of similarly situated individuals.

• AB 692 provides for actual damages or a statutory minimum of $5,000 per affected worker, whichever is greater.

• AB 692 authorizes injunctive relief and recovery of reasonable attorneys’ fees and costs for prevailing workers.

Enforcement of AB 692 is not retroactive and applies only to contracts and employment agreements entered into on or after January 1, 2026. That said, now is the time for employers to proactively review offer letters, training agreements, bonus agreements, and related documents to ensure compliance and avoid unnecessary penalties.

If you need assistance reviewing your employment agreements and contracts to ensure compliance, please feel free to reach out to The HR Girl.

Effective January 1, 2026, pursuant to SB 294, all California employers must provide a standalone written notice of workers’ rights to all new hires and distribute this notice to current employees by February 1, 2026 and at time of hire for employees hired after February 1, 2026. California’s Department of Industrial Relations (DIR) released a template notice last month in English and Spanish. To ensure compliance with this new law, California employers must:

• Provide a standalone written notice of workers’ rights to all new hires and current employees. The Labor Commissioner has provided a template that employers can use. This notice must be distributed beginning on February 1, 2026 and every year thereafter.

• Provide employees an opportunity to indicate a designated emergency contact that employers must contact in the event an employee is arrested or detained either at the worksite or offsite during work hours, if the employer has actual knowledge of the arrest or detention. Employers must provide an opportunity to designate an emergency contact by March 30, 2026 for existing employees and at time of hire for employees hired after March 30, 2026.

• Comply with the non-retaliation and non-discrimination clauses set forth by this bill which prohibits employers from discharging, threatening to discharge, demoting, suspending, or in any manner discriminating or retaliating against an employee for exercising or attempting to exercise their rights under the bill, as provided.

• Train management and HR staff on new requirements.

• Update on-boarding documents to include the designation of an emergency contact.

Employers that fail to meet the requirements of SB 294 could face hefty fines of up to $10,000 per employee. As such, it is imperative that all California employers prepare thoroughly.

If you need assistance navigating the new requirements set forth by SB 294, please feel free to reach out to The HR Girl – we are happy to assist!

If you log into LinkedIn, you are bound to find a story about someone who has been on the job market for over a year. They have submitted thousands of applications, with little to no response. Or worse: they've been strung along an incredibly arduous interview process only to be rejected by an automated email after their 7th interview (I am not joking, I've actually heard of people who have gone through 7 rounds of interviews only to be rejected). So, why is the job market so bad? The answer is complex.

The AI Conundrum

Among the stories of people submitting thousands of applications only to be ghosted by recruiters, the common belief is that AI is the reason they are being rejected. This is only half right. Many people believe that the “AI algorithm” automatically rejects applicants based on missing keywords, erroneous formatting, or even just due to mere bias. In Mobley v. Workday, Inc. (2024), Derek Mobley alleges Workday’s algorithm-based screening tool had a disparate impact on him and other applicants based on race, age, and disability. Even though Workday is not considered the employer, or even a prospective employer, Mobley alleges that “Workday may nonetheless be held liable as an “agent.”” Mobley is not the only plaintiff in this case, as it has now become a class-action lawsuit, with a potential class of hundreds of millions of participants. So yes, algorithm-based applicant screening tools have been accused of discrimination, but that’s not really the whole picture.

In my view, AI tools have made it increasingly easier for job seekers to apply for jobs. Whether you believe this is a good thing or a bad thing is subjective, but the objective truth is that this has led to a saturation of applicants in roles. Most, if not all, HR professionals and recruiters are still reviewing each individual resume that is submitted. That means most recruitment and HR professionals are reviewing thousands of resumes for roles that are posted. Personally, I have had roles that have had over 500 applicants overnight because a role I was recruiting for was posted on Indeed, and all an applicant needed to do was submit a resume. That’s after just one night, after a few days, the applicant count had risen to the thousands. And I was an HR department of one, reviewing all of these resumes. The capacity of HR and recruitment teams to sift and thoroughly review each resume that comes in has been stretched thin, and with a decrease in open roles for job seekers, this issue may only get worse.

The new roles that just aren’t coming

What I haven’t mentioned, and the clear elephant in the room, is that the job market is bad because the economy is bad. Why is the economy bad? Well, I’ll leave that to the (honest) economists to explain, but one obvious reason is that labor, operational, and materials costs have skyrocketed as a result of newly imposed tariffs. This has led to a huge spike in layoffs, with consulting firms like Challenger, Gray, & Christmas stating over 1.1 million Americans have been laid off this year alone.1

Unfortunately, job creation numbers have not kept up with the number of layoffs. In their latest job creation report, which was delayed due to the government shutdown*, the Labor Department reported that 64,000 jobs had been created in November, but 105,000 jobs had been “lost” in October.2

Speaking of job numbers, I think it’s important to explain where these numbers come from, as we often see them in news articles and job reports without much insight into how they are derived. Most employment data comes from the Current Population Survey, which is handled by the Bureau of Labor Statistics (BLS). The survey provides data on key employment indicators, including unemployment rates, labor force participation rates, and employment-population ratios.3 After reviewing the data on the current population survey for 2025, I noticed that there are no numbers reported for October, with a simple note that states “* Data unavailable due to the 2025 lapse in appropriations.” November then shows a 59.6% employment-population ratio, which is not significantly lower than any other month or year between 2015-2025.

The bottom line is that there aren’t enough roles available for the number of people entering the job market following massive, nationwide layoffs. This is why most HR and recruitment teams are seeing thousands of applicants for their open roles. Thus, the competition for these roles has increased, while the capacity of hiring teams to review all applicants thoroughly and thoughtfully has decreased. It’s a compounding issue that is only being made worse by one other factor: interviewing and hiring processes have become longer and more complex.

The labyrinth of interviews

A few weeks ago, I was scrolling on TikTok while eating lunch, as a normal Millennial would do, and I came across a video of a girl describing her most recent interview experience.4 She goes on to explain what her interview process had been like: 6 interviews and a surprise assessment that consisted of an entire content strategy and a myriad of other things for active projects the company was currently working on. This immediately raised a red flag with her. She inquired about the assessment and also expressed her hesitation about completing an assessment on an active campaign. She requested the use of an NDA so that her work would not be used later on, and then, she was ghosted. They never responded to her request nor gave her an update on the role. Six interview rounds later, she was left with no job offer and no explanation on why they were not moving forward with her.

This experience, unfortunately, is not unique. Many job seekers are seeing increased interview rounds, pre-employment assessments, and time-to-hire. In 2025, the average time-to-hire was between 45-60 days, with executive and higher-level positions taking up to 120 days to hire.5 Four months to hire for one position may be common for government agencies, but it is not common for private companies, and also not advisable, as it can lead to increased recruitment costs.

The question is: why are employers so hesitant to hire? Many employers blame their low confidence in candidate quality on “The Great Resignation” that occurred during the COVID era. They are hesitant to hire a candidate who may resign before completing their probationary period, and are hesitant to invest in employees who aren’t looking to stay long term. This may be valid, but in my opinion, the “Great Resignation” brought equilibrium to a job market that had largely swayed in favor of employers and had kept wages and compensation packages stagnant.

Another reason we’re seeing such sluggish hiring is the lack of true entry-level jobs or companies that are willing to provide on-the-job training. Many candidates note that entry-level roles now require two or more years of experience in a particular area and offer little to no on-the-job training. From the employer’s perspective, hiring someone with more experience typically means less, or shorter, training is required. This also typically means onboarding and training costs are reduced.

Is there a solution?

I want to lean into hopefulness and believe that there is a solution to the issues we are currently having in the job market. Here are a few recommendations, based on my experience and the experience of many other HR and recruitment professionals, that will ease some of the pitfalls we are seeing in today’s employment world:

1. You cannot apply a “set it and forget it” approach to AI when it comes to hiring. If you are using any type of artificial intelligence or algorithmic technology in your recruitment processes, you must monitor it closely and conduct continuous bias audits. Don’t assume technology will always function and always have a backup for when it fails.

2. Shorten those interview processes (unless it’s for highly skilled labor). That project manager or executive assistant role does not need seven rounds of interviews and 3 pre-employment assessments. Be intentional and direct with your interview questions and cap the interview process to a maximum of 3 interviews.

3. Provide on-the-job training for entry-level roles and lower the experience requirements. As a standard practice, entry-level roles should have some type of on-the-job training and should not require more than five years of experience for a candidate to be considered.

4. Be open and honest during the hiring process and try not to ghost candidates. Look, I get it. Sometimes we get very busy and forget to email a candidate back who has been contacting us about the status of the position. However, it is important to respect the time and effort a candidate has given throughout the recruitment process and be proactive in communicating the status of their candidacy. Also, this can avoid any potential risk of the candidate feeling that they were discriminated against. (Reminder that the CA Fair Employment and Housing Act prohibits discrimination in the hiring process.)

If you need help building out your hiring and recruitment processes, feel free to reach out to us!

As an HR Consultant, I have encountered a myriad of new services and software that are lauding how AI will make workflows more efficient and halve the time employees need to spend on tasks. This may sound enticing, but with emerging state and federal regulations on AI, companies may actually be exposing themselves to greater legal risk if they don't deploy AI responsibly.

AI HR is infiltrating all HR-related services

AI is being adopted and deployed faster than employment laws are being developed. This often leaves employers in limbo on how to use AI in a compliant way. The issue, however, is that AI is being used in almost every area that deals with employment compliance and law. Whether it’s employment law or training and development, AI has become a prominent selling feature for many HR-related services. Take, for example, Ethena, a compliance training platform, that is promoting an “AI Training Builder”, where companies can upload policy documents and their AI will create training modules for employees. As an HR professional who has had to create training decks before, this tool sounds like an absolute God-send. I worry, however, that this AI tool may not fully meet the training requirements for each state. Additionally, this tool relies on documents to be uploaded, which leaves it open to human error. Nonetheless, this tool promises to reduce the amount of time HR professionals need to spend on creating training materials, and that is a huge selling point for most in this industry.

It doesn’t stop there, though. Applicant tracking systems, such as LinkedIn and Indeed, have also deployed AI tools that claim to make the job search process easier for both job seekers and job posters. AI in recruitment has been scrutinized recently, however, because it can have underlying bias in the tools that are used to review job applicants (looking at you Workday). This points to the remaining necessity for a human to monitor AI tools and ensure they are operating in a compliant manner.

Additionally, an entire industry of certifications and training has risen in response to the increased use of AI in the HR space. Companies like AIHR are selling “Artificial Intelligence for HR Certifications” for over $1,000! Certifications look great on a resume, but before spending thousands on certifications, both employees and employers should evaluate how much value these certifications will actually bring. Given that regulations and limitations on AI use in the workplace are fluctuating, information that was pertinent one year may become obsolete the next.

State and federal governments are still playing catch-up on AI regulation

AI regulation is lagging behind development. However, more and more states are passing legislation that regulates the use of AI in the workplace, specifically in hiring, separation, and promotion decisions.

• California: Effective October 1, 2025, employers are prohibited from using AI tools to discriminate on the basis of FEHA-protected classes. Employers are also required to keep records of all data input into an AI system and the decisions and/or recommendations it made for 4 years.

• New Jersey: In January 2025, New Jersey’s Attorney General Platkin and the New Jersey Division of Civil Rights (DCR) launched a Civil Rights Technology Initiative, which provided guidance on the use of AI in helping screen applicants and employee performance management. The Guidance stated that the existing New Jersey Law Against Discrimination (LAD) applies to discrimination resulting from the use of AI as it does to any other discriminatory conduct.

• Colorado: Effective June 30, 2026, employers with more than 50 employees that develop or deploy “high-risk AI” must comply with the Colorado AI Act. The Colorado AI Act sets forth documentation and disclosure, incident reporting, risk management, impact assessment, and notification requirements for both developers and deployers of high-risk AI. Failure to comply with the new act could result in civil penalties of up to $20,000 per violation.

• Illinois: An amendment to the Illinois Human Rights Act, effective January 1, 2026, adds definitions for “Artificial Intelligence” and “Generative Artificial Intelligence”. It also makes it a civil rights violation for an employer to use AI in employment decisions, including recruitment, hiring, promotion, reinstatement, selection for training and apprenticeship, discharge, discipline, and tenure, or terms and conditions of employment that may discriminate against employees on the basis of protected classes. It also prohibits using zip codes as a proxy for protected classes, which is a new and unique protected class.

Why AI still poses risks for workplaces

The AI world is the wild west. With cloudy state and federal regulations, employers should thoroughly evaluate AI programs before implementing them on a company-wide level. Although AI HR programs boast higher levels of efficiency, employees using AI still need to be wary of using AI for work-related tasks, as it can introduce risks regarding confidentiality and compliance. In my experience, many businesses are using a “set it and forget it” approach when it comes to AI, which is a huge misstep. AI programs need to be regularly tested and monitored to ensure they are not operating in a non-compliant way.

Conclusion

AI looks like it is here to stay for the foreseeable future. If you choose to implement AI HR software or make it a cornerstone of your people strategy, make sure it is being regularly bias-tested, monitored, and updated to remain compliant with state and federal AI regulations.

If you need assistance navigating the complex AI HR space, feel free to reach out to The HR Girl for more guidance!

1. Workplace "Know Your Rights" Act (SB 294): Employers of all sizes must provide a written notice to current employees that explain workers’ protections against unfair immigration-related practices by February 1, 2026.

   • The Department of Labor (DOL) is scheduled to release a template notice by January 1, 2026.

   • Employers must also allow employees to choose an emergency contact that can be contacted in the event an employee is arrested or detained at work no later than March 30, 2026.

   • Noncompliance could result in hefty fines of up to $10,000 per employee.

2. 30-Day Window to Provide Data Breach Notifications (SB 446): Effective January 1, 2026, organizations must inform affected parties of data breaches within 30 days of discovery or notification that a breach has occurred.

   • Violations of this new law could result in large fines and civil lawsuits.

   • Employers must notify the CA Attorney General within 15 calendar days of notifying consumers of a breach if a breach involves more than 500 California Residents.

3. Job Posting Requirements and Equal Pay Act (SB 642): Effective January 1, 2026, the definition of “pay scale” will be: a good faith estimate of the salary or hourly wage range that the employer reasonably expects to pay for the position upon hire.

   • This act also extends the statute of limitations for which a worker can bring a civil action to recover wages when they were paid less than another worker of a different sex or race / ethnicity for comparable work to three years.

4. Mandatory Penalties for Pay Data Reporting Failures (SB 464): Effective January 1, 2026, civil penalties for failing to comply with the state’s pay data reporting requirements will become mandatory.

5. Changes to Cal-WARN Notices (SB 617): Signed into law by Governor Newsom on October 1, 2025, this new law makes significant changes to the Cal-WARN notices and requires employers to provide the following:

   • Whether the employer plans to coordinate services, such as a rapid response orientation, and through which entity;

   • Email and telephone number of the local workforce development board;

   • Description of the rapid response activities offered by the local workforce development board;

   • Description of CalFresh, the state’s food assistance program;

   • CalFresh benefits helpline;

   • Link to CalFresh website; and

   • Email and telephone number of the employer.

   • Employers must also notify the state that these notices have been complied with.

6. Ban on “Stay-or-Pay” Provisions (AB 692): It is unlawful to include terms in a contract that impose fees or penalties, or require employees to repay debts or otherwise be sent to debt collection, in contracts entered into on or after January 1, 2026. Employers that offer sign-on bonuses will need to review the language in offer letters to ensure it remains compliant with AB 692.

As noted, California has a slew of new employment laws that may affect your business. If you're not sure whether these new employment laws apply to your business, feel free to reach out to The HR Girl for assistance!