Bunce v. Visual Technology Innovations (1)
- Party
- Lawyer
- AI Tool
- ChatGPT
Hallucinated Content
Fabricated
- Case Law
Non-existent case cited in the Withdrawal Motion; the court could not locate it in any research tool and deemed it an AI hallucination.
- Case Law
Non-existent case cited in the Appeal Motion; the court could not locate it and deemed it an AI hallucination.
Misrepresented
- Case Law
Misrepresented for a proposition it does not support; the Court explained Degen concerns the rule of disentitlement, not the cited point.
Outdated Advice
- Overturned Case Law
Relied upon despite being reversed and vacated; subsequent order expressly vacated the sanctions decision.
- Overturned Case Law
Cited for immediate appealability under the collateral-order doctrine, but that proposition was impliedly overruled by Cunningham and Comuso; no longer good law for the point.
Outcome
Notes
AI UseCounsel admitted using ChatGPT to draft two motions (Motion to Withdraw and Motion for Leave to Appeal), without verifying the cases or researching the AI tool’s reliability.Hallucination Details2 Fake cases:McNally v. Eyeglass World, LLC, 897 F. Supp. 2d 1067 (D. Nev. 2012) — nonexistentBehm v. Lockheed Martin Corp., 460 F.3d 860 (7th Cir. 2006) — nonexistentMisused cases:Degen v. United States, cited for irrelevant propositionDow Chemical Canada Inc. v. HRD Corp., cited despite later vacaturEavenson, Auchmuty Greenwald v. Holtzman, cited despite being overruled by Third Circuit precedentRuling/SanctionThe Court sanctioned Counsel $2,500 payable to the court and ordered him to complete at least one hour of CLE on AI and legal ethics. The opinion emphasized that deterrence applied both specifically to Counsel and generally to the profession.Key Judicial ReasoningRule 11(b)(2) mandates reasonable inquiry into all legal contentions. No AI tool displaces the attorney’s personal duty. Novelty of AI tools is not a defense.
Protect Your Firm Before the Next Sanction
This case is one of 921 documented AI hallucination incidents in legal proceedings. The AI Ethics Training for Attorneys covers the specific failure modes, the Model Rules at risk, and the verification protocols that prevent them — built directly from the sanctions record.