How AI is reshaping lawyers' professional liability risk
By Al Roberts
Artificial intelligence (AI) is quickly becoming a standard tool in legal practice.
Roughly four out of five legal professionals now use AI, with adoption spanning firms of all sizes, according to the 2025 Clio Legal Trends Report. Attorneys are leveraging generative AI to conduct research, draft documents, summarize case law and improve efficiency.
However, as adoption accelerates, so do the professional liability risks.
For insurance agents and brokers serving law firms, understanding how AI is changing lawyers professional liability (LPL) exposures is becoming increasingly important.
First warning signs
While the insurance industry hasn't yet seen a significant volume of AI-driven legal malpractice claims, there have been disciplinary proceedings and regulatory scrutiny tied to attorneys' misuse of AI tools. One of the most widely cited early examples occurred in 2023 when attorneys submitted a court filing containing multiple fictitious case citations generated by ChatGPT. The court sanctioned the attorneys and the incident quickly became a cautionary tale throughout the legal profession.
The case highlighted a fundamental reality: AI can assist lawyers, but it cannot replace their professional judgment.
Many generative AI platforms are prone to "hallucinations," or responses that appear authoritative but contain inaccurate, skewed or fabricated information. Legal research generative AI tools produced incorrect information from 17% to 34% of the time, according to a Stanford University study. If an attorney relies on those results without independent verification, the consequences can be significant.
For insurance providers, this creates a familiar challenge. At its core, a lawyer who relies on inaccurate AI-generated information may still be committing a professional error or omission. In other words, the technology is new, but the underlying exposure is not.
Competence remains governing standard
One of the most important concepts in evaluating AI-related legal risk is attorney competence. Under the American Bar Association's (ABA) Model Rule 1.1, lawyers have a duty to provide competent representation. That obligation does not disappear simply because a task is delegated to an AI platform. If attorneys choose to use AI tools, they must understand their limitations and verify the accuracy of the information produced. Therefore, the risk is not necessarily simply using AI; it's using AI carelessly.
An attorney who blindly accepts AI-generated research, citations or legal analysis without review may be exposing both the client and the firm to unnecessary risk. From a professional liability perspective, that exposure can look very similar to other forms of negligent legal work.
Most lawyers professional liability policies include some level of coverage for disciplinary proceedings brought by state bar associations or state supreme courts. From a coverage standpoint, an attorney's reliance on an AI-generated hallucination may still be treated as an error or omission, much like a mistake made using traditional legal research tools. It is best to review specific coverages provided by the insurance policy.
Equally important to hallucinations are confidentiality concerns. According to ABA Model Rule 1.6, attorneys have an ethical obligation to protect client information. If lawyers enter confidential, privileged or personally identifiable information into AI platforms that lack enterprise-grade protections, they may be exposing sensitive data to third parties. This raises important questions regarding client consent, disclosure and data security.
Wild, wild west
ABA Model Rule 1.4 says lawyers have a duty to communicate with clients — but does that apply to disclosing AI usage?
Many jurisdictions are still developing guidance on how attorneys should use AI. State bar associations, courts and regulators are approaching it differently. Some jurisdictions encourage disclosure when AI is used, while others remain largely silent on the issue. Standards continue to evolve, making it difficult to establish universal best practices.
This "wild west" environment creates uncertainty for law firms and insurance providers alike. Legal firms often find themselves operating in a rapidly evolving regulatory environment, and for insurance professionals, the lack of uniformity presents an underwriting challenge.
At the same time, AI continues to improve. Attorneys are becoming more sophisticated users, and firms are increasingly implementing internal policies governing AI adoption. Over time, those controls may help reduce risk. In the meantime, attorneys must know what their local jurisdictions require regarding disclosure of using AI. Caution should be taken when drafting filings or submitting information to a court that contains AI-generated information.
For now, however, insurance providers and agents should recognize that law firms may have dramatically different approaches to AI governance, even when they practice in similar areas of law.
Key questions brokers should be asking their clients
As AI becomes more prevalent, agents and brokers should consider expanding their conversations with law firm clients. Key questions to ask can include the following:
Does the firm have a formal AI use policy?
● What AI tools are attorneys and staff permitted to use?
● Are AI-generated work products independently reviewed and verified?
● How does the firm protect confidential client information when using AI?
● Has the firm provided training on ethical and professional obligations related to AI?
● Does AI usage impact fees and billable hours?
● Does the firm's professional liability policy address AI-related exposures?
● Does the firm notify clients of AI usage or ask for their consent to use AI tools?
These conversations can help identify potential vulnerabilities before they develop into claims.
Emerging trend: Pro se litigation
Another emerging consideration is the growing accessibility of AI to non-lawyers. Individuals representing themselves in legal matters, also known as pro se litigants, may increasingly use AI to draft pleadings, analyze legal issues and pursue claims. Although it remains unclear how significant this trend will become, AI could make self-represented individuals more aggressive and more persistent in pursuing disputes, including claims against attorneys.
Whether those claims ultimately succeed is a separate question. Regardless, AI has the potential to change not only how lawyers practice law, but also how legal disputes are initiated and pursued.
Proceed with optimism
In the near future, AI may become as commonplace in legal practice as online legal research platforms are today. The firms that benefit most will be those that view AI as a tool to enhance professional judgment rather than replace it.
For insurance providers, agents and brokers, the opportunity lies in helping law firm clients adopt AI responsibly. The most effective risk management strategy is not avoiding the technology altogether, but rather encouraging attorneys to continue to exercise the competence, diligence, and professional judgment that clients and courts expect. As AI becomes more embedded in legal practice, those fundamentals will matter more than ever.
Al Roberts is Senior Vice President of Underwriting, Professional Liability, at Berkley Select, which he joined in 2021. His career of more than 25 years spans the private practice of law as well as working at an MGA and multiple carriers specializing in professional liability. Roberts holds a JD from the University of Illinois Chicago School of Law and a BS from Northern Illinois University. Al can be reached at [email protected].
Opinions expressed here are the author's own.
Link to article:
Originally published in: How AI is reshaping lawyers' professional liability risk
Disclaimer:
Berkley Select is a member company of W. R. Berkley Corporation, a Fortune 500 Company. The views expressed here are those of the author and do not necessarily represent the views of Berkley Select or W. R. Berkley Corporation.