When AI Hallucinates, Your Business Isn’t Protected

Multiple cases have made national news because lawyers used Artificial Intelligence (AI) tools to generate their briefs and didn’t verify their citations. The reason these cases were buzzworthy was that the AI tools that were used apparently “hallucinated” and created fake citations to cases that did not exist. In the U.S. District Court for the Central District of California, this happened to law firms, K&L Gates LLP and Ellis George LLP, who admitted the attorneys used AI tools, including Google Gemini, CoCounsel, and Westlaw Precision.

With AI developments moving faster than ever, employers are confronted with difficult choices about how to take advantage of these tools and still avoid potential pitfalls. AI tools can often create a faster work product, synthesize data, and potentially automate manual processes to reduce operational costs.

On the other hand, information generated from AI is not always accurate and complete, may be biased, and removes the nuances of human decision-making and perspective. While AI tools often provide a tempting shortcut, employers should ask, is good enough right for my business?

Whether you are using AI tools to select candidates, generate work product, or answer your questions, here are additional things to keep in mind:

• Verify the Information: Double-check that the information provided by AI is accurate, especially when it comes to employment law compliance. AI tools need to be “fed” the right questions and relevant information, and even then, they do not always produce an accurate answer. You may click on a citation only to find out that it doesn’t match the answer the AI tool has generated.
• Avoid “Black & White” Thinking: In our experience, AI tools often provide “black and white” answers, even when there is no statute, case law, or agency guidance to support the answer. When the AI tool is prompted for a specific citation to support its guidance, it may then “admit” that one does not exist. When it comes to HR compliance and employee relations issues, for many situations, there is no single “right or wrong” answer, but instead multiple strategies and approaches.
• Incorporate Human Review to Mitigate Bias and Discrimination: California’s Civil Rights Department issued regulations effective October 1, 2025, aimed at addressing potential discrimination when employers use AI tools to make employment-related decisions. When using AI tools in selecting candidates for hiring, promotions, etc., incorporate human review to help mitigate claims of bias. Employers must also hold on to AI-related records for four years. Refer to CEA’s article here for more information.
• Draft a Clear AI Policy for Your Staff: Employers should also consider whether—and to what extent—they want employees using AI to carry out their job duties. This can raise questions regarding confidentiality issues, protecting proprietary information, the quality and accuracy of employee work product, among other concerns. It is best to address these issues in a clear, written handbook policy.
• Reach out to a Subject Matter Expert: California Employers Association is available to employers when you need to bounce ideas or questions off an HR subject matter expert, ensure your handbook policies are compliant, and train your staff in an engaging environment. Members may reach out to us via phone, email, and live chat! Call us at 800.399.5331.