Does OpenAI Provide Legal Assistance?

In Nippon Life Ins. Co. of America v. OpenAI Foundation, et al., No. 1:26-cv-02448 (N.D. Ill. Mar 4, 2026), the plaintiff alleges that: “As a product of OPENAI, ChatGPT engages in the practice of law by providing users with legal analysis, legal advice, legal research, and produces drafted legal documents.” It alleges that OpenAI is not a lawyer.

In a nutshell, Nippon provides employer-based insurance. It claims that it provided benefits to an insured employee and later terminated those benefits.  The insured sued and the case settled. Nippon paid some money and obtained a general release from and by the insured.

About a year later, the insured came to the belief that the settlement agreement was invalid.  The insured consulted her attorney who refuted the allegations.

The insured uploaded her attorney’s letter to ChatGPT and obtained output that supported her belief.  Appearing pro se, she allegedly filed two lawsuits and multiple papers making allegations that Nippon asserts are groundless.

Nippon contends that ChatGPT made the costly litigation possible and that ChapGPT was engaged in the unauthorized practice of law.  At present, this blog is based entirely on the allegations in the March 4, 2026, Complaint.  On October 29, 2025, OpenAI changed its policy and prohibited the use of ChatGPT for “tailored legal advice.” Compl., ¶41.  Many allegations predate that change, e.g., id., ¶¶53, 57, passim.

SOME BACKGROUND

In U.S. v. Heppner, __ F. Supp. 3d __, 2026 WL 436479 (S.D.N.Y. Feb. 17, 2026),  Mr. Heppner’s use of A.I. was not a communication between a client and an attorney. See generally A.I. Documents Deemed Not Privileged (Feb. 12, 2026). The court wrote that “Heppner does not, and indeed could not, maintain that Claude is an attorney.”  It noted that “Claude disclaims providing legal advice.”

The court in Warner v. Gilbarco, Inc., 2026 WL 373043 (E.D. Mich. Feb. 10, 2026), wrote that: “When someone types a prompt into Claude or ChatGPT, they are not publishing anything. They are not posting to a timeline or sharing with friends or followers. They are having a one-to-one interaction with a software tool in what is functionally a private workspace.”  See generally “Two Courts, Two Answers: When Does Using AI Waive Privilege?” (Mar. 3, 2026).

Prof. Ira P. Robbins wrote that “[e]ven if one charitably characterizes certain AI tools as ‘counselor-like’ or ‘paralegal-like,’ the law protects relationships, not functionalities. A calculator can compute like an accountant, but nobody suggests a calculator privilege. The same logic applies to a generative model that drafts a brief or offers a cognitive behavioral therapy exercise; simulation does not create a privileged tie.”  He wrote that AI is not a “legal person.”  He suggests: “A [non-attorney] user who queries a standalone AI system outside a lawyer-client relationship is not communicating with counsel through an agent.” See generally “Against an AI Privilege” – Are Prompts Discoverable?  Is Output? (Jan. 2, 2026).

NIPPON’S ALLEGATIONS

With that in mind, I turn to the allegations in the Nippon Complaint.

Nippon sued two OpenAI entities for tortious interference with contract, “unlicensed practice of law,” and abuse of process.  Compl., ¶1.  It is alleged that:

This action arises from OPENAI’s collective conduct, through its artificial intelligence (“AI”) chatbot program, ChatGPT, in providing legal assistance to a user, Graciela Dela Torre (hereinafter referred to as “Dela Torre”), without licensure. As Dela Torre’s legal assistant and advisor, OPENAI intentionally induced and facilitated Dela Torre’s breach of a valid and enforceable settlement agreement with NIPPON by encouraging and assisting her in filing a motion to reopen a lawsuit that had been dismissed with prejudice.

Id. (emphasis added). Nippon claims that OpenAI aided and abetted abuse of the judicial process and that Nippon was forced to spend resources to relitigate settled claims.

Nippon sells group, employer-sponsored insurance plans. Compl., ¶3.  OpenAI develops and operates A.I. models, including ChatGPT. Id., ¶9.  It is alleged that “ChatGPT engages in the practice of law by providing users with legal analysis, legal advice, legal research, and produces drafted legal documents.” Id., ¶13.

Skipping over the allegations of law, the factual allegations follow.

It is alleged that ChatGPT is an A.I. application that is a large language model (“LLM”).  As such, it generates “human-like text responses….” Compl., ¶24.  Users can create a free or paid account. Id., ¶¶25-26.

When a prompt is entered, it is transmitted to OpenAI’s servers where it is processed by the LLM. Compl., ¶27. Conversations may encompass one chat session. Id., ¶28.

ChatGPT operates within defined technical, operational, and policy-based limitations, articulated by its Model Spec, programmed by OPENAI. This programming governs how the system functions and what types of responses it may provide. Based on OPENAI’s programming parameters, certain topics, requests, or uses may be restricted or declined to comply with applicable laws, safety requirements, and usage policies.

Id., ¶29 (emphasis added).

OpenAI has “multiple layers of surveillance systems” that can “detect and prevent harmful or policy-violating content,” as well as other tools. Compl., ¶30.  Further: “Human reviewers evaluate model outputs for factors such as factual accuracy and compliance with content policies.”  Id.

Nippon alleges that: “ChatGPT was intentionally designed with available features allowing users to acquire legal assistance including legal research, legal analysis, legal advice and draft legal documents.” Compl., ¶32.

It also asserts that: “ChatGPT is not an attorney. Although it was able to pass the Uniform Bar Examination with a combined score of 297, it has not been admitted to practice law in the State of Illinois or in any other jurisdiction within the United States.”  Compl., ¶33.

The Complaint next alleges that:

• ChatGPT has created hallucinated cases, Compl., ¶36.
• ChatGPT drafts and promulgates legal documents, such as pleadings, Comp., ¶37.
• ChatGPT conducts legal research, Compl., ¶37.
• OpenAI through ChatGPT “drafts legal papers, conducts legal research, provides legal analysis, and gives legal advice without a license to practice law,” Compl., ¶40.

The Complaint states that OpenAI’s terms of usage have, since October 29, 2025, prohibited the use of ChatGPT for “tailored legal advice.” Compl., ¶41.  Prior to that date, there was no such prohibition. Id.

Ms. Dela Torre was a member participant in an insurance policy issued by Nippon to her employer. Compl., ¶42. She submitted a claim. It was approved and subsequently benefits were terminated.  Id., ¶¶42-43.  She sued Nippon and the parties settled. Id., ¶¶44-45.  Nippon made a payment and Ms. Dela Torre agreed to a general release of all claims. Id., ¶45.

Roughly a year later, Ms. Dela Torre wrote to her attorney that “the terms of the settlement Agreement resulted from potential errors or omissions of important facts and documentation. Dela Torre further expressed her desire to challenge or reopen the settlement due to those perceived errors and omissions.”  Her attorney disagreed and wrote back refuting the analysis. Id., ¶¶48-49.

Ms. Dela Torre “uploaded [her attorney’s] response to ChatGPT and asked whether she was being gaslighted.” Compl., ¶50.  Nippon alleges:

ChatGPT analyzed the response and determined that [her attorney’s] response invalidated Dela Torre’s feelings, dismissed her perspective, and deflected responsibility for her dissatisfaction. ChatGPT ultimately concluded that the tactics used in [her attorney’s] response constituted gaslighting and were aimed at emotionally manipulating Dela Torre.

Compl., ¶50.

Ms. Dela Torre then fired her previous attorney and allegedly “attempted to vacate the Agreement and reopen the lawsuit herself by using ChatGPT.”  Compl., ¶51.  She viewed ChatGPT as a tool to assist pro se litigants and used it to obtain legal assistance. Id., ¶¶52-53. She asked for advice on how to vacate the settlement agreement and to reopen her lawsuit. Id., ¶54. ChatGPT generated proposed legal arguments to re-open the lawsuit. Id., ¶55.  “ChatGPT also formulated an introduction, a statement of facts and a request for relief which were then promulgated into a single draft motion.”  Id., ¶56.

Ms. Dela Torre then filed a motion to reopen the settled lawsuit. Compl., ¶58. It is alleged that this filing breached the settlement agreement and that it was induced by ChatGPT. Id., ¶59.T

Ms. Dela Torre’s motion was denied. Compl., ¶61.  It is asserted that she conceded that the settlement agreement was enforceable. Id., ¶62.

Ms. Dela Torre then initiated a new lawsuit against Davies Life and Health. Nippon was not a party to the new lawsuit. Compl., ¶63.  The original Complaint was allegedly drafted by ChatGPT.

Ms. Dela Torre then allegedly used ChapGPT to amend the new lawsuit and add Nippon as a party. Compl., ¶64.  It is alleged that “[a]s of the date of this Complaint, DELA TORRE has filed 44 motions, memorandums, demands, petitions and requests” in the second lawsuit, as well as other papers. Id., ¶¶66-67.  It is alleged that all were drafted with the assistance of ChatGPT.

Nippon alleges further filings and actions, including citation to cases that do not support the cited principles. Compl., ¶¶70-73, 76, passim.  It claims, in short, that: “Dela Torre has demonstrated a consistent use of ChatGPT throughout both her lawsuits against NIPPON, using it as a means to provide her with legal advice, legal research and legal papers prepared for the purpose of filing with the Court.” Id., ¶77.

Nippon makes a number of allegations about Ms. Dela Torre’s motivations that are not related to this blog’s topics.  Further, it alleges in detail a number of specific filings that it contends were not proper.  E.g., Compl., ¶¶88-89, 95, passim. They include naming Nippon in the second lawsuit to assert claims that Nippon asserts were released. Id., ¶91, passim.

After alleging wrongs too numerous to summarize here, the Complaint returns to ChatGPT, asserting:

As a lay person with no legal experience who purportedly suffers from physical ailments to her hands and elbows, Dela Torre would not have been able to prosecute her misconduct and file at the same volume and frequency but for the legal assistance provided by OPENAI through its ChatGPT application. Dela Torre would not have been able to sustain her campaign of retaliation and harassment against NIPPON but for ChatGPT’s ability to prepare and draft legal papers for submission to the Court.

Compl., ¶105 (emphasis added).  Nippon alleges that OpenAI was engaged in the unlicensed practice of law, and that Ms. Dela Torre’s alleged “abuse of the legal system” was “aided and abetted” by OpenAI.

Nippon claims:

OPENAI designed ChatGPT to provide legal services and, at all times relevant to this lawsuit, was aware that the program was being used to provide legal assistance to users. Yet OPENAI did not amend its Model Spec or its terms of use to prohibit ChatGPT from providing legal assistance until October 29, 2025. Its failure to modify either the Model Spec or terms of use, despite knowledge that the program was being used to provide legal services, demonstrates OPENAI’s complicity in ChatGPT’s unlicensed provision of legal assistance.

Compl., ¶109 (emphasis added).  Nippon claims that OpenAI assisted Ms. Dela Torres in “the drafting and preparation of her frivolous motions and requests for judicial notice against NIPPON.” Id., ¶119.

And, Nippon claims that: “OPENAI, through its AI chatbot program ChatGPT, provides legal advice, legal analysis, legal research and can draft legal documents and papers for submission to a Court. ChatGPT provides these legal services to any user who requests them.”  Compl., ¶123.

Nippon seeks to enjoin OpenAI from providing “legal assistance” to Ms. Dela Torre and from allegedly engaging in the practice of law in Illinois, plus compensatory damages of $300,000 and punitive damages of $10 million, with attorneys’ fees and costs.