In a high-profile incident within a federal case, a New York lawyer is facing potential sanctions after citing fabricated cases generated by ChatGPT in a legal brief, as reported by various news outlets.

The incident occurred within a personal injury lawsuit filed by Roberto Mata against Colombian airline Avianca, currently pending in the Southern District of New York. Steven A. Schwartz from Levidow, Levidow & Oberman, one of the plaintiff’s attorneys, admitted to consulting ChatGPT to augment his legal research while preparing a response to Avianca’s motion to dismiss.

However, in an order issued in early May, Judge P. Kevin Castel noted that “six of the submitted cases appear to be fictitious judicial decisions, complete with fabricated quotes and internal citations.” This unprecedented situation raised serious concerns.

In a subsequent affidavit filed in May, Schwartz explained that not only did ChatGPT provide the legal sources, but it also vouched for the reliability of the opinions and citations that the court now questions.

For instance, a document attached to Schwartz’s affidavit reveals that he specifically asked ChatGPT whether one of the six cases questioned by the judge was genuine, to which the chatbot falsely confirmed its authenticity.

Furthermore, Schwartz inquired about the legitimacy of the other provided cases, and ChatGPT incorrectly asserted that they were real and accessible in reputable legal databases such as LexisNexis and Westlaw.

In his affidavit, Schwartz acknowledged that the source for the legal opinions he relied upon has proven to be unreliable. As a consequence of Schwartz’s errors, The New York Times published a front-page story, and a hearing has been scheduled by the judge to consider potential sanctions.

Legal professionals emphasize that this incident should serve as a cautionary tale for attorneys utilizing AI-powered technology, emphasizing the importance of independently verifying AI-generated insights. However, it should not lead to a complete abandonment of AI’s potential within the legal industry.”

Verification Imperative: Lessons Learned from a Legal Misstep

Legal experts are pointing out a critical oversight made by Schwartz that led to his recent predicament. The failure to verify the authenticity of the cases generated by ChatGPT during his research has been identified as a fundamental mistake.

One notable voice, Kay Pang, an experienced in-house counsel at VMware, emphasized the importance of following a cardinal rule: “Verify, verify, verify!” In a LinkedIn post, Pang highlighted the lawyer’s oversight in not conducting proper due diligence.

Nicola Shaver, the CEO and co-founder of Legaltech Hub, echoed this sentiment on LinkedIn, emphasizing the necessity of independent verification rather than solely relying on ChatGPT’s responses.

During a Lawtrades webinar, Ashley Pantuliano, Associate General Counsel at OpenAI, acknowledged ChatGPT’s value as a starting point for legal research. However, Pantuliano issued a cautionary note, alerting attorneys to the potential for inaccurate information generated by the tool. She stressed the need for attorneys to remain diligent and exercise caution when utilizing ChatGPT.

In an affidavit submitted in the Avianca case, Schwartz candidly admitted his lack of prior experience using ChatGPT for legal research. Unaware of the possibility that the tool’s contents could be false, he expressed deep regret for relying on artificial intelligence to supplement his legal research. Schwartz vowed to never repeat this mistake and pledged to rigorously verify the authenticity of any future AI-generated content.

Despite multiple attempts to reach Schwartz for comment, no response was received as of Tuesday morning. By incorporating these valuable insights, legal professionals can learn from Schwartz’s misstep and prioritize the essential practice of verification in their own research endeavors.

Embracing AI in Legal Work: A Perspective for Tech Innovators and Researchers

In the rapidly evolving landscape of legal technology, recent discussions surrounding the limitations of AI have raised concerns among legal professionals. However, it is crucial for technology innovators, researchers, and other professionals interested in Generative AI to understand that not all AI tools are created equal. Drawing from a recent article by Alex Su, Head of Community Development at Ironclad, it becomes evident that dismissing AI entirely would be a mistake. Instead, it is important to recognize the potential benefits and limitations of AI in the legal field.

Su clarifies that ChatGPT, although an impressive AI model should not be equated with all AI-powered legal tools. He emphasizes that there are reputable companies with a track record of successfully serving legal customers through AI-powered legal tech tools. While acknowledging that no generative AI product can claim 100% reliability, Su highlights that these vendors are likely to provide users with accurate warnings and candid discussions about their accuracy rates, surpassing those of ChatGPT, particularly in law-related use cases.

For legal professionals, embracing AI requires a “learning mindset.” AI can be a valuable tool for streamlining legal processes, particularly in generating initial drafts or providing a preliminary analysis. By recognizing AI’s role as a “first pass or first draft tool,” lawyers can leverage its capabilities to enhance their efficiency and focus on higher-value tasks that require human expertise.

It is essential not to disregard AI altogether. As Su points out, legal professionals will inevitably encounter AI technology in their work, regardless of their reservations. Rather than discarding AI, professionals should strive to understand its limitations and the areas in which it can bring significant value. Embracing AI as a complementary tool in the legal profession will lead to better-informed decision-making, improved productivity, and increased efficiency.

In conclusion, technology innovators, researchers, and professionals interested in Generative AI must adopt a balanced perspective when considering AI in the legal field. While acknowledging the limitations and potential pitfalls, it is crucial to recognize the value AI can bring as a supporting tool in legal work. By approaching AI with an open mind and understanding its role as a preliminary aid, legal professionals can harness its potential to drive positive change within the industry.