(c) 2010-2024 Jon L Gelman, All Rights Reserved.

Tuesday, September 5, 2023

Implementing AI for Workers’ Compensation Law Firms and Insurance Companies

Artificial intelligence [AI] opens new frontiers for workers’ compensation law firms and insurance companies. Last week, Google announced new applications that will vastly expand how workers’ compensation claims can be serviced, managed, and supported.

Google Cloud Vertex AI is a managed machine language [ML] platform that helps businesses build, deploy, and scale ML models. Workers compensation attorneys and insurance companies can use it to:


Vertex AI can train machine learning models to classify workers’ compensation claims into different categories, such as injury type, severity, and liability. The automation process will free up attorneys and insurance adjusters to focus on the more complex issues of cases.


Vertex AI can analyze large amounts of claims data to identify patterns and trends. This information can be used to improve risk management and prevent future claims. With this trending data, attorneys can focus on adapting skill sets to handle emerging waves of potential litigation, including adapting pre-discovery processes. This may include locating knowledgeable medical experts and gathering evidence for trial.


AI will be capable of personalizing the claims experience for claimants. Google Vertex AI can be used to create personalized communication and support for claimants. This can help to improve the claimant experience and reduce the risk of disputes. Clients, through the use of AI, could receive responses to inquiries and could be provided with status reports on a real-time basis.


The automation and classification of claims could be trained to identify collateral third-party claims and assist in ascertaining other supplemental benefits available through collateral resources and benefit programs.

A law firm could utilize a machine learning model to classify claims into different categories, such as slip and fall, car accident, product liability, Social Security Disability Insurance, or Multi-District Litigation [MDL]. This will increase the accuracy in identifying other benefits and save time and resources for the law firms.


An insurance company could use AI to identify workers’ compensation claims data patterns. The company has found that specific industries, such as construction and manufacturing or a particular chemical producer, have a higher rate of claims than others. This information can be used to improve risk management and prevent future claims. 


AI could personalize the claims experience for clients. The law firm could establish a chatbot that uses machine learning to answer client's questions and provide information as to the status of the case. This will reduce the number of calls to the law firm and increase the client satisfaction rate and law firm efficiency.

As the technology develops, AI will likely be used in even more innovative ways to improve the workers’ compensation claims/litigation process.


The cloud-based system can be trained on insurance companies and law firms' private models (data). However, specific requirements must be met to meet regulatory requirements and tailor to each jurisdiction.

The private large language models (law firm or insurance company) must be converted into a format that the cloud-based AI can understand. This can be achieved using a variety of tools and techniques.

The private models must be hosted in a secure environment that meets the security requirements of the insurance companies and law firms.

The private models must be appropriately anonymized to protect the privacy of individuals. This can be done using various techniques, such as removing personally identifiable information from the data.

Once these requirements are met, the cloud-based AI system (First Party, Third-party, and “Model Garden” language models, can be used to train new models based on the private models of insurance companies or law firms. These newly created private models must be segregated in a secure environment and exclusively used by the specific law firm or insurance company. The language models should not be integrated or used to train other language models. The end product would be a private language model for the specific insurance company or law firm.

Here are some of the benefits of using cloud-based AI to train on private models: the ability to leverage the expertise of the insurance company or law firm, reduce the costs to create and utilize models throughout the firm or company, and accelerate innovation and collaboration.


Regulatory bodies will probably eventually weigh in on the use of AI in the legal and insurance framework of workers’ compensation.  Those would include the privacy of data, the mode of dissemination, and the scope and extent of their use. 

The anticipated legal and regulatory frameworks should guide law firms and insurance companies considering using cloud-based AI model language generation. These issues should not deter the advance of innovative AI in the workers’ compensation process. The advancement of AI will undoubtedly provide more efficiency and cost savings, resulting in monumental benefits for the compensation system.


Warehouse Workers' Injuries Are Increasing as Employers Use Artificial Intelligence


Jon L. Gelman of Wayne, NJ, is the author of NJ Workers’ Compensation Law (Thomson-Reuters) and co-author of the national treatise Modern Workers’ Compensation Law (Thomson-Reuters). For over five decades, the Law Offices of Jon L Gelman 1.973.696.7900 have represented injured workers and their families who have suffered occupational accidents and illnesses.

Blog: Workers ' Compensation

LinkedIn: JonGelman

LinkedIn Group: Injured Workers Law & Advocacy Group

Author: "Workers' Compensation Law" West-Thomson-Reuters

Recommended Citation: Gelman, Jon L.,   Implementing AI for Workers’ Compensation Law Firms and Insurance Companies, Workers' Compensation Blog (9/5/2023),

© 2001-2023 Jon L Gelman. All rights reserved.

Attorney Advertising

Prior results do not guarantee a similar outcome.


Download Adobe Reader