With AI hallucinations increasingly showing up in legal filings, law firms face a new professional responsibility challenge: staying on the cutting edge of technology without surrendering human judgment and client trust. Wisner Baum warns that AI can support legal work, but unverified research, fake citations, and weak internal governance can expose firms to sanctions, malpractice claims, and reputational harm.
LOS ANGELES, July 14, 2026 /PRNewswire/ — When a Mississippi federal judge recently sanctioned attorneys on both sides of a breach-of-contract case for “failing to prevent AI hallucinations in legal filings,” the consequences went beyond monetary penalties. According to Bloomberg Law, the court disqualified the attorneys and barred them from appearing before that court for two years.
In legal matters, AI errors carry high stakes, yet many law firms now use this technology for research, drafting, and litigation support. Wisner Baum LLP warns that lawyers are entering a new era of accountability. AI mistakes in court filings are not just software failures. They are failures of attorney responsibility.
The Washington Post reported that courts have documented at least 95 U.S. incidents involving AI-generated hallucinations in legal filings since June 2023. In July 2025, the Associated Press reported that a federal judge sanctioned three Butler Snow attorneys defending Alabama’s prison system after filings included references to fake ChatGPT-generated cases. The judge removed the lawyers from the case, ordered disclosure of the sanctions to clients, opposing counsel, and judges, and referred the matter to the Alabama State Bar.
AI Cannot Replace Human Judgment in Litigation
The message from courts is becoming clear: judges will not tolerate AI hallucinations in court filings. They are sanctioning lawyers who fail to verify AI-generated work.
“AI can be a useful tool, but it cannot replace the lawyer’s duty of candor, competence, and independent judgment,” said Crawford Appleby, partner and member of the AI Committee at Wisner Baum LLP. “The attorneys whose names are on the filing are responsible for what is in it. If a brief cites a fake case, misstates the law, or misrepresents what a real case stands for, the problem is not that a machine made a mistake. The problem is that the legal team allowed that mistake to reach the court.”
The risk is broader than fake citations. In 2024, the American Bar Association issued Formal Opinion 512, warning that lawyers using generative AI must consider duties related to competence, confidentiality, communication, supervision, meritorious claims, candor to tribunals, and reasonable fees. Meanwhile, a Stanford-led study found leading legal AI research tools hallucinated between 17% and 33% of the time under tested conditions.
For law firms, those rates create a governance challenge. Legal AI products may be marketed as litigation-ready, but they should not be used without attorney supervision and diligence. AI may invent cases, misstate precedents, or incorrectly apply legal principles. In litigation, every one of those errors matters.
AI Is Here to Stay
Appleby doesn’t see the issue as a pro-AI or anti-AI debate. The more important question is whether firms have policies, supervision, verification steps, and professional discipline to use AI responsibly.
“Firms should assume AI is already being used somewhere in the workflow,” Appleby said. “The response is not to panic or ban it. AI is here to stay. Instead, usage must be addressed with policy, approved tools, training, supervision, and mandatory verification before anything reaches a court, client, or opposing party.”
Wisner Baum, a national plaintiffs’ law firm known for complex, high-stakes litigation, sees the issue as especially important in cases involving injured people, scientific evidence, expert testimony, corporate conduct, and public accountability. But in all cases, accuracy is not just administrative housekeeping. Accuracy can affect strategy, client trust, judicial integrity, and claims involving real harm.
The firm has adopted an internal AI-use policy limiting attorneys to approved generative AI tools and requiring human verification through trusted legal sources before AI-assisted work is used in litigation. An internal team has developed in-house tools tailored to specific legal tasks, including identifying hallucinated authorities in briefs before they are filed.
Be the Tortoise, Not the Hare
Appleby has practical reminders for law firms using AI:
- Don’t rely on a court’s AI disclosure rules: All legal and factual representations need to be verified so that nothing presented to the court is false. Appleby cautions: “Disclosure rules alone do not solve the problem. What matters most is that firms and courts have screening systems in place and that lawyers treat verification as mandatory, not optional.”
- Sanctions don’t limit exposure: Even when courts do not impose sanctions, AI-related mistakes may still create malpractice exposure. If inaccurate AI-generated research causes lawyers to miss controlling authority, submit defective arguments, or otherwise harm a client’s case, professional liability questions may follow.
- Limit and verify: For now, use AI for targeted research rather than to draft entire briefs, and check every cited authority against trusted legal databases such as Westlaw, Lexis, or CourtListener. Manually confirm that every cited case supports the legal proposition asserted.
“The cure for AI hallucinations is humanity,” Appleby said. “The bottom line: Lawyers should always be looking for new technologies that can make them more efficient and provide advances. But they should proceed with caution in adopting and implementing AI tools. Slow and steady wins the race. Be the tortoise, not the hare.”
About Wisner Baum
Wisner Baum is a nationally recognized plaintiffs’ law firm with offices across the United States. For decades, the firm has represented individuals harmed by corporate misconduct, dangerous products, environmental exposure, and systemic failures. Wisner Baum is known for its trial experience, investigative rigor, and leadership in complex, high-impact litigation involving some of the world’s largest companies. For more information, visit www.wisnerbaum.com.
References
- American Bar Association Standing Committee on Ethics and Professional Responsibility. (2024, July 29). Formal Opinion 512: Generative artificial intelligence tools. American Bar Association, americanbar.org/news/abanews/aba-news-archives/2024/07/aba-issues-first-ethics-guidance-ai-tools/.
- Associated Press. (2025, July 24). Judge sanctions lawyers defending Alabama’s prison system for using fake ChatGPT cases in filings. AP News, apnews.com/article/lawyers-judge-ai-prison-alabama-c6a64736cb488cf6379624403d3757ca.
- Kay, J. (2025, November 13). Fake AI citations produce fines for California, Alabama lawyers. Bloomberg Law, news.bloomberglaw.com/litigation/fake-ai-citations-produce-fines-for-california-alabama-lawyers
- Magesh, V., Surani, F., Dahl, M., Suzgun, M., Manning, C. D., & Ho, D. E. (2024). Hallucination-free? Assessing the reliability of leading AI legal research tools. arXiv, dho.stanford.edu/wp-content/uploads/Legal_RAG_Hallucinations.pdf.
- Seiden, D. (2026, June 9). Lawyers sanctioned over AI use on both sides in federal case. Bloomberg Law, news.bloomberglaw.com/litigation/lawyers-on-both-sides-in-mississippi-case-punished-for-ai-errors.
- Wu, D. (2025, June 3). Lawyers using AI keep citing fake cases in court. Judges aren’t happy. The Washington Post, washingtonpost.com/nation/2025/06/03/attorneys-court-ai-hallucinations-judges/.
Media Contact:
Karla Jo Helms
JOTO PR
727-777-4629
jotopr.com
View original content to download multimedia:https://www.prnewswire.com/news-releases/ai-in-the-courtroom-is-becoming-a-malpractice-risk-for-law-firms-wisner-baum-warns-302824760.html
SOURCE Wisner Baum
