The quality and organization of the data behind artificial intelligence tools can make the difference between useful results and costly mistakes.
That is particularly important when it comes to employment law compliance, where inaccurate information or inconsistent language can expose employers to significant legal and financial risks.
Speaking at the HR Tech 2026 conference in Las Vegas, Kimball Parker, founder of AI-powered employment law compliance platform SixFifty, said organizations can dramatically improve AI results by first investing in structured, reliable data.
“One incredible thing about AI is that it does pretty well with unorganized data,” he said. “But if you do the hard work of organizing your data, it will produce stunning results.”
Parker’s experience building SixFifty offers a case study in how combining curated legal information with generative AI can help employers navigate one of HR’s most complicated responsibilities.
The Growing Complexity of Employment Law
Parker founded SixFifty in 2019 with the backing of law firm Wilson Sonsini. The company was created to make complicated areas of law more accessible and actionable for businesses through technology. It was later acquired by Paychex and integrated into the payroll platform, while continuing to operate as a standalone product and integrate with major human capital management systems.
The company’s focus on employment law emerged during the COVID-19 pandemic, when the rapid expansion of remote work created new compliance challenges.
“Compliance takes time, resources, and legal expertise — and most teams are already stretched thin,” Parker said. “As remote work expanded and employment laws continued to evolve, we saw firsthand how challenging it was for fast-growing companies to stay compliant with the state-specific requirements where all their employees worked.”
Before the pandemic, SixFifty had employees in two states. Within months, its workforce was distributed across 10 states, creating unfamiliar legal obligations.
“We didn’t know the employment laws in all of those states,” Parker said. “That was a big problem. I was stunned by the complexity.”
Parker identified three reasons employment law is particularly difficult for businesses.
First, it covers more issues than virtually any other area of law, spanning the entire employment lifecycle, from job postings and offer letters to workplace policies and employee separations.
Second, requirements vary not only by state but also by city and county. An employer with workers in California and Texas, for example, must navigate substantially different legal frameworks. Even neighboring cities can impose different obligations.
Third, employment laws change frequently. State legislatures, city councils, and county governments regularly introduce new requirements, creating hundreds of changes that employers may need to track each year.
The financial exposure can be significant. Parker cited survey findings indicating that 20 percent of businesses were involved in an employment dispute in 2025, with an average payout of $180,000 when a business loses such a dispute.
“For most small businesses, that will end your business,” he said.
Building a Structured Legal Database
Recognizing that employment law was both complicated and consequential, SixFifty set out to build a comprehensive database of U.S. employment law.
Although laws vary considerably across jurisdictions, Parker said they follow a relatively logical structure. That makes it possible to organize, categorize and summarize legal requirements in a consistent way.
The company spent three years, beginning in 2021, cataloging employment laws across the country, completing the project in 2024. Parker said the result is the largest database of employment law data in the country.
That investment became particularly valuable as generative AI tools emerged, he said.
Parker cautioned that relying on general-purpose AI tools for legal work presents serious accuracy challenges. Employment agreements, policies, and other legal documents require precision and consistency. Even small differences in wording can change the meaning of a legal provision.
“If you go to ChatGPT and ask for an employment agreement for Illinois, and then wait five minutes and ask for it again, you will get two different documents,” he said. “They will be directionally similar, but not the same. That does not work for a legal document. Every word matters.”
He also pointed to recent research from Stanford University indicating that consumer AI tools make errors 58 percent to 82 percent of the time when answering legal questions.
Even when AI systems are instructed to rely exclusively on statutes and original case law, the error rate remains between 17 percent and 34 percent, according to the research Parker cited.
The problem, he explained, is that interpreting and summarizing legal sources requires substantial legal expertise. General-purpose AI models are also trained on internet content that may be inaccurate or outdated.
By contrast, Parker said, layering AI over a curated, tagged, and organized database of employment law summaries can reduce the error rate to below 5 percent.
“Structured data will turn an AI tool from good to excellent,” he said.
Implications for HR
For HR departments, the potential payoff extends beyond reducing compliance errors. AI can make sophisticated processes more accessible to organizations that lack the resources to maintain large internal legal teams or engage outside counsel for every question.
A technology that simplifies complex legal requirements can help smaller businesses and resource-constrained HR teams better understand their obligations.
Ultimately, Parker said, the goal is to broaden access to reliable legal guidance. “The law is a public good. And every company in the country is entitled to solid, reliable, precise legal help.”
Was this resource helpful?
Validate your HR expertise
Earning your SHRM-CP credential makes you a recognized expert and leader in the HR field.
Related Content
Transform HR with mobile technology to streamline workflows, enhance communication, and boost employee engagement across distributed teams.
A new study found that talent development practitioners believe performance-based assessments are more effective at measuring employee learning outcomes than knowledge tests.
As AI makes deeper inroads into organizations, employee concerns about the technology continue to grow. Here's how HR leaders can address the fear of AI among their own staff, as well as in the wider organization.