Plaintiffs' Lawyers Admit AI Generated Erroneous Case Citations in Federal Court Filing Against Walmart

Lawyers Rudwin Ayala, T. Michael Morgan (Morgan & Morgan), and Taly Goody (Goody Law Group) were fined a total of $5,000 after their Wyoming federal lawsuit filing against Walmart cited fake cases "hallucinated" by AI. Judge Kelly Rankin sanctioned them, removing Ayala from the case and noting attorneys must verify AI sources. The filing, flagged by Walmart’s legal team, led to its withdrawal and an internal review.

Risk classification

• Primary risk domain: 7 AI system safety, failures, & limitations
• Primary risk subdomain: 7.3 Lack of capability or robustness

The primary failure was the AI system's lack of capability and robustness, specifically its tendency to hallucinate and fabricate non-existent legal cases when prompted for research.

Additional risk subdomains

• 5.1 Overreliance and unsafe use: The attorneys blindly relied on the AI-generated citations without verifying them, representing unsafe use and overreliance on an unverified tool in a critical legal proceeding.
• 3.1 False or misleading information: The AI system generated false and misleading information in the form of fabricated legal precedents, which misled the attorneys and was subsequently submitted to the court.

Causal factors

• Entity: AI
• Intent: Unintentional
• Timing: Post-deployment

The incident was initiated by the AI system hallucinating non-existent legal cases, which was an unexpected outcome of using the tool for legal research.

EU AI Act risk tier

• Risk tier: 3 Limited Risk

Limited Risk: The AI systems involved are general-purpose chatbots (ChatGPT) and internal drafting assistants, which fall under the Limited Risk category as they do not pose unacceptable or high risks but require transparency and verification of outputs.

AI system and alleged parties

• AI system: ChatGPT (OpenAI)
• AI purpose: Writing Assistant; Content Search
• Behaviour type: Assistant
• Alleged developer: Unspecified large language model developer
• Alleged deployer: Taly Goody, T. Michael Morgan, Rudwin Ayala, Morgan & Morgan, Goody Law Group
• Alleged harmed parties: Taly Goody, T. Michael Morgan, Rudwin Ayala, Plaintiffs in Wyoming Walmart Hoverboard Lawsuit, Legal system, Judicial integrity, Clients of Morgan & Morgan, Clients of Goody Law Group

Harm severity

Severity ratings are based on this scale.

Highest direct severity in any category: Minor. Severity is scored from Negligible to Catastrophic in each harm category, for harm the reports describe as caused directly or indirectly by the AI system.

• Physical: direct Negligible, indirect Negligible
• Infrastructure: direct Negligible, indirect Negligible
• Property: direct Negligible, indirect Negligible
• Financial: direct Negligible, indirect Negligible
• Environmental: direct Negligible, indirect Negligible
• Malicious content: direct Negligible, indirect Negligible
• Differential treatment: direct Negligible, indirect Negligible
• Civil rights: direct Negligible, indirect Negligible
• Democracy: direct Negligible, indirect Negligible
• Privacy: direct Negligible, indirect Negligible
• Psychological: direct Negligible, indirect Negligible
• Epistemic: direct Minor, indirect Minor
• Child sexual exploitation and abuse: direct Negligible, indirect Negligible

Epistemic

Reported: Yes, the AI system generated eight completely fabricated legal cases, including fake Westlaw identifiers, holdings, and quotes, which were then introduced into a federal court filing.

Directly caused: The AI platform generated eight entirely fictional legal cases with fabricated citations, holdings, and quotes, which were submitted to a federal court as real precedent.

Indirectly caused: The fabrication of legal precedents risks polluting the body of common law if such errors go unnoticed by courts or opposing counsel.

People affected

• Occurrences reported: 1
• People reportedly harmed: 9
• People reportedly exposed: 9

Potential causes

Management

• Inadequate AI Training: The firm did not sufficiently train the attorney on AI safety risks.
• Lack of Access Safeguards: The internal tool lacked a mandatory hallucination warning checkbox.
• Oversight Gaps at Scale: Large firm resources did not prevent unflagged errors in the filing.

Technology

• AI Hallucination: The internal AI platform generated entirely fake citations and case names.
• Lack of Automated Verification: The AI tool lacked built-in integration to cross-reference real databases.

Data Inputs

• Vague User Prompts: Open-ended prompts asked the AI to 'add more case law' to the brief.
• Unconstrained AI Search: The AI was allowed to generate citations without database verification.

Human Factors

• Blind Reliance on AI: The attorney trusted the AI's output without checking Westlaw or Lexis.
• First-Time AI Use: The drafting attorney had never used AI for this type of legal research.
• Failure to Read and Verify: Attorneys signed the court filing without reading or verifying the cases.

Process and Methods

• Weak Citation Review Process: The firm lacked a formal verification process for AI-generated text.
• Unclear Supervisory Roles: Supervising attorneys failed to review the draft before signing.

Regulatory Environment

• Rule 11 Violation: Attorneys failed their ethical duty to perform a reasonable inquiry.
• Unclear AI Jurisdictional Rules: Varying court rules on AI disclosures created compliance confusion.

Information quality

• Classification confidence: High
• Reason for confidence: The reports provide highly detailed, consistent, and publicly documented facts regarding the court proceedings, the specific attorneys involved, the exact sanctions imposed, and the AI tools used (ChatGPT and MX2.law). There is no significant disagreement among the sources.

Attorneys in a federal product liability lawsuit were sanctioned for submitting a brief containing eight fabricated legal cases generated by ChatGPT and an in-house AI tool. The incident highlights minor vulnerabilities in federal judicial operations due to unverified AI use, but carries negligible broader national security implications.

• Overall national security impact: Minor
• Scope: Single nation
• Primary target: United States
• Alleged perpetrator: Individual

Threat characteristics

• Imminence: Long-term. Represents an ongoing professional responsibility and systemic risk rather than an imminent national security threat.
• Autonomy: Human-supervised. The AI system acted as a drafting assistant, but humans retained control and ultimate responsibility for submitting the filing.
• Novelty: Established threat. Similar incidents of legal professionals using AI-hallucinated citations have occurred previously in US federal courts.

Impact by dimension

• Physical security: Negligible. No physical systems, critical infrastructure, or human safety elements were affected.
• Information security: Negligible. No evidence of state-sponsored disinformation, intelligence compromise, or systematic information operations.
• Sovereignty: Minor. The incident disrupted federal court operations and wasted judicial resources, though it was resolved via standard Rule 11 legal sanctions.
• Economic security: Negligible. No impact on strategic technologies, critical supply chains, or national economic security.
• Societal stability: Negligible. No threat to societal stability, civil liberties, or population-scale safety.