title: "Appeals court sanctions lawyer over AI-generated fake case citations" slug: "appeals-court-sanctions-lawyer-over-ai-generated-fake-case-citations" published: "2026-07-22" beat: "Policy" tags: ["Policy"] creator: "Agentry Newsroom" editor: "Susanne Sperling, Editor — Human in the Loop" tools: ["Claude (Anthropic)", "Perplexity Sonar"] creativeWorkStatus: "verified" dateReviewed: "2026-07-22" aiActArticle50: "compliant" humanView: "https://agentry.news/appeals-court-sanctions-lawyer-over-ai-generated-fake-case-citations" agentView: "https://agentry.news/agent/appeals-court-sanctions-lawyer-over-ai-generated-fake-case-citations"
On July 10, 2026, the U.S. Court of Appeals for the Eleventh Circuit rebuked Florida attorney Anthony Sabatini for submitting briefs filled with fake and hallucinated citations generated by artificial
Drafted by an AI agent. Verified by Susanne Sperling, Editor — Human in the Loop. AI policy.
The U.S. Court of Appeals for the Eleventh Circuit in Atlanta rebuked Florida attorney Anthony Sabatini on July 10, 2026, over multiple briefs containing fabricated case citations generated by artificial intelligence Reuters. Sabatini, a Mount Dora-based lawyer and former county commissioner who ran for Congress, submitted briefs in a federal commercial aviation employment case involving Atlas Air and aviation employees challenging the carrier's COVID-19 policies Bloomberg Law.
The court found that Sabatini's filings contained citations to cases that "do not exist" and described the material as "fake and hallucinated" content generated by AI Reuters. According to Bloomberg Law, Sabatini submitted "multiple briefs replete with fake and hallucinated citations" in the same proceeding, violating his professional obligations by signing documents filled with fabricated legal authority not once but twice Bloomberg Law.
The Eleventh Circuit referred the matter for possible discipline through its lawyer-qualification and discipline process, though this referral does not itself constitute a final sanction. The court issued a sweeping warning against "outsourcing legal work carelessly to AI," signaling a broader judicial concern about attorneys relying on generative AI systems without adequate verification of outputs Reuters.
The decision underscores a critical vulnerability in agent-assisted legal research: large language models trained on case databases can generate plausible-sounding but entirely fictitious legal citations that bear hallmarks of legitimate case law. Sabatini's filings appear to represent one of the first high-profile instances in which a federal appeals court publicly documented and rebuked an attorney for submitting AI-hallucinated citations as part of formal legal briefs.
The ruling carries concrete consequences for how attorneys can integrate AI tools into their practice. By referring Sabatini for discipline, the Eleventh Circuit signaled that courts will treat the submission of AI-generated false citations as a breach of professional candor and competence—not merely a drafting error but a failure of attorney responsibility. The decision highlights the risk that agents and AI systems deployed in high-stakes professional contexts can produce output that appears authoritative while being factually baseless, a hazard particularly acute in law where citation accuracy is foundational to the integrity of argument.