Sullivan & Cromwell Reportedly Filed Emergency Bankruptcy Motion Containing AI-Generated Legal Errors

Sullivan & Cromwell reportedly filed an emergency motion in the Chapter 15 bankruptcy case of Prince Global Holdings that contained numerous citation and legal inaccuracies. The firm later acknowledged that some errors were AI hallucinations and that its AI-use and citation-review policies had not been followed. After opposing counsel identified the problems, the firm reportedly undertook a formal remedial response.

Risk classification

• Primary risk domain: 3 Misinformation
• Primary risk subdomain: 3.1 False or misleading information

The AI system generated discrete false information in the form of fabricated legal citations and non-existent legal sources within a court filing.

Additional risk subdomains

• 5.1 Overreliance and unsafe use: The lawyers overrelied on the AI tool, failing to follow internal policies or perform a proper secondary review of the generated citations.

Causal factors

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

The risk was caused by the AI system generating unintended hallucinations and fabricated citations during post-deployment use by the law firm.

EU AI Act risk tier

• Risk tier: 3 Limited Risk

Limited Risk: The AI system was used to generate text content (court filing drafts), which falls under the category of AI-generated content requiring transparency, though it was used here as a drafting assistant.

AI system and alleged parties

• AI system: None named
• AI purpose: Writing Assistant; Technical Text Generation
• Behaviour type: Assistant
• Alleged developer: Generative AI developers
• Alleged deployer: Sullivan & Cromwell LLP, Lawyers, Law firms, Andrew G. Dietderich
• Alleged harmed parties: United States Bankruptcy Court for the Southern District of New York, Parties in In re Prince Global Holdings Limited, Insolvency practitioners, Court-appointed liquidators, Clients of lawyers, Boies Schiller Flexner LLP

Harm severity

Severity ratings are based on this scale.

Highest direct severity in any category: Negligible. 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 Negligible, indirect Negligible
• Child sexual exploitation and abuse: direct Negligible, indirect Negligible

People affected

• Occurrences reported: 1
• People reportedly exposed: 2

Potential causes

Management

• Inadequate Oversight of AI Use: Management failed to ensure compliance with existing AI policies.

Technology

• AI Hallucinations: The AI tool generated non-existent legal citations and misquoted law.

Human Factors

• Failure to Follow AI Policies: Attorneys used AI tools without following the firm's established AI policies.
• Inadequate Vetting of Output: Lawyers did not fully vet and verify the AI-generated research results.

Process and Methods

• Flawed Secondary Review: The secondary review process failed to catch the inaccurate citations.

Information quality

• Classification confidence: High
• Reason for confidence: The report clearly outlines the sequence of events: an unnamed AI tool was used for legal drafting, it generated fabricated citations, the firm failed to review it, and opposing counsel caught the error. There is no ambiguity about what occurred, though the specific AI model is not named.
• Ambiguities identified: The specific AI program used by the law firm is not identified in the report.
• Alternative interpretations: None. The firm explicitly acknowledged the AI hallucinations and apologized.

A prominent law firm submitted a federal bankruptcy court filing containing fabricated legal citations generated by an AI tool. While the incident briefly introduced false information into a government proceeding, it was quickly identified and corrected through standard judicial processes, resulting in negligible national security impact.

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

Threat characteristics

• Imminence: Long-term. The specific incident is resolved, but the potential for AI hallucinations to affect legal proceedings represents an ongoing systemic concern.
• Autonomy: Human-controlled. The AI tool was used as a drafting assistant, and human lawyers retained final control over the submission, despite failing to properly review the output.
• Novelty: Established threat. AI-generated hallucinations in legal filings are an established phenomenon with multiple documented precedents in recent years.

Impact by dimension

• Physical security: Negligible. The incident involved a legal filing and did not present any threats to physical systems, infrastructure, or human safety.
• Information security: Negligible. There is no indication of intentional information warfare, intelligence compromise, or state-sponsored disinformation campaigns.
• Sovereignty: Minor. The submission of fabricated AI-generated citations to a federal bankruptcy court temporarily affected judicial proceedings, but the impact was minor and managed through standard court procedures.
• Economic security: Negligible. No strategic technology theft, financial system attacks, or significant economic manipulation occurred as a result of this incident.
• Societal stability: Negligible. The incident did not threaten social cohesion, cause civil unrest, or result in systematic human rights violations.