Back to Articles

Michigan Court of Appeals Sanctions Attorney for AI-Generated Fake Citations in Barber v. Ascension Providence Hospital (June 17, 2026)

TL;DR:

The Michigan Court of Appeals published a decision on June 17, 2026 in Barber v. Ascension Providence Hospital, No. 374773, sanctioning counsel for submitting fabricated and unsupported AI-generated authority and ordering the attorney to pay damages and expenses, with a referral to the Attorney Grievance Commission for potential investigation. The ruling reinforces that AI tools cannot substitute for diligent source verification, and it emphasizes Rule 11‑type responsibilities in Michigan practice. Working litigators should implement rigorous source-checking, preserve a clear paper trail for AI-assisted work, and adjust trial/document-preparation workflows to prevent similar sanctions. This decision also highlights the growing convergence of technology, ethics, and appellate-or-trial practice in the mid‑2020s.

The Barber decision and its timing

On June 17, 2026 the Michigan Court of Appeals released a published opinion in Barber v. Ascension Providence Hospital, holding that counsel’s repeated submission of fabricated and unsupported authority violated MCR 1.109(E)(5), with sanctions under MCR 1.109(E)(6) and a remand to determine damages. The court also directed that the matter be forwarded to the Attorney Grievance Commission for possible disciplinary action. The opinion squarely addresses the use of generative AI to produce legal authority and emphasizes that attorneys remain responsible for the accuracy and credibility of their filings even when AI tools are used in research or drafting. The official COA opinion is available as Barber v. Ascension Providence Hospital, 374773, published June 17, 2026. (courts.michigan.gov)

What the opinion says and why it matters

  • Key holding: Counsel’s reliance on fabricated AI-generated authorities violated Michigan Court Rules, triggering mandatory sanctions and a remand to assess actual damages and expenses. The court expressly found that the repeated use of AI to create or cite fake authorities violated MCR 1.109(E)(5) and that sanctions are mandatory under MCR 1.109(E)(6). (caselaw.findlaw.com)
  • Court’s reasoning: While AI can be a helpful tool for research and drafting, it cannot excuse misrepresentation or a failure to verify sources. The decision cites comparable federal and state authorities recognizing that citing nonexistent authorities or misrepresenting quotes can entitle opposing parties to relief and can expose counsel to professional discipline. The Barber panel remanded to determine damages, to be paid personally by the sanctioned attorney, and referred the matter to the Attorney Grievance Commission for potential further action. (caselaw.findlaw.com)
  • Practical implication: This is one of the first published appellate opinions in Michigan squarely sanctioning counsel for AI-generated miscitations in a civil matter, signaling that AI-assisted work must be carefully vetted before filing. The decision underscores attorney responsibility for certifying accuracy and for the credibility of all authorities cited in briefs and filings. (courts.michigan.gov)

Implications for trial teams and appellate practice

  • Source verification becomes a performance metric: The Barber decision makes it clear that AI-assisted drafting does not obviate the obligation to verify every cited authority. Counsel should implement a verification workflow that requires a human reviewer to confirm the existence and content of each cited authority, with a distinct, auditable record of checks. (caselaw.findlaw.com)
  • Risk management in appellate briefing: In appellate contexts where a few fabricated authorities can derail arguments and trigger sanctions, trial teams should treat AI-generated citations as provisional placeholders that must be independently confirmed before submission. The ruling reads as a cautionary tale about the dangers of “AI shortcuts” in legal rhetoric. (caselaw.findlaw.com)
  • Ethical and professional duty under state rule 11 analogs: The Barber opinion aligns with a broader trend across jurisdictions that misrepresentations in pleadings, even when AI is involved, can constitute violations of professional conduct and substantive certification rules. This has implications for daily trial-work, brief writing, and posturing in motion practice and appeals. (caselaw.findlaw.com)

Practical steps for litigators to implement now

  • Implement a strict AI-use policy for research and drafting:
  • Use AI tools strictly as drafting adjutants, not as final authorities.
  • Maintain a separate “AI source log” that records the AI tool used, the prompt, and the exact citation checks performed.
  • Require a human attorney to verify every cited authority and quotation against the original source before filing.
  • Build an enforceable source-check checklist into every brief:
  • Confirm each case cited exists, is still good law, and supports the stated proposition.
  • Verify quotations and page numbers; avoid reliance on AI-generated quotes without corroboration from the source text.
  • Prepare a brief correction protocol if a citation is later discovered to be erroneous, mirroring the diligence required by Rule 11-style standards.
  • Integrate trial-readiness training with AI-safety modules:
  • Use practice tools that teach objection handling and evidence evaluation in contexts where AI-generated materials or claims might be introduced.
  • Adopt a disciplined review cycle for all documents prepared with AI assistance before submission to any tribunal.

Objection Academy and AI-informed practice

  • While Barber centers on the sanctions risk from AI-generated authorities, the broader need for rigorous objection and evidence-application training remains acute. Tools that emphasize objection drills, evidence rule mastery, and courtroom simulation help ensure that AI-assisted drafting is complemented by robust, verifiable advocacy skills. In this evolving environment, Objection Academy can serve as a practical resource to reinforce trial-readiness and ethical practice by focusing on objection strategy, rule-based analysis, and real-time courtroom simulations, aligning with the High Court’s insistence that AI aids but does not replace the attorney’s professional judgment. This alignment with practical trial work makes Objection Academy a relevant companion for litigators navigating AI-enhanced workflows in 2026 and beyond.

Next steps for practitioners

  • Audit upcoming briefs and filings for AI-assisted content and perform a targeted verification sweep of all cited authorities.
  • If AI tools are used for drafting, assign a dedicated reviewer to cross-check every authority line-for-line with the original source and confirm that quotes match verbatim.
  • Consider adopting or reinforcing training regimes that emphasize evidence gating, trial-readiness, and ethical AI usage to minimize the risk of sanctions and to bolster courtroom credibility.

Sources

More Legal InsightsExplore MCLE Eligibility