The Volokh Conspiracy
Mostly law professors | Sometimes contrarian | Often libertarian | Always independent
Court Notes Apparently Nonexistent Precedents Cited in Brief, Suggests Brief May Have Been Partly AI-Written
From Ex parte Lee, decided last week by the Tenth District Texas Court of Appeals, in an opinion by Chief Justice Tom Gray, joined by Justices Matt Johnson and Steven Lee Smith:
Allen Michael Lee is charged with one count of aggravated sexual assault of a child and two counts of sexual assault of a child. Bail amounts were set at $400,000 in total for the three charges. He has not been able to make bail. He filed a pre-trial application for a writ of habeas corpus asking to either be released or have bail reduced to $15,000 total. After a hearing, the trial court denied Lee's application.
In one issue, Lee contends the trial court abused its discretion in denying Lee's request for a bail/bond reduction. Specifically, he contends the initial bail set was excessively high and that the trial court abused its discretion by denying his application without an explanation.
In presenting error to this Court, an appellant's brief must contain a clear and concise argument of the contentions made with appropriate citations to authorities and to the record. That has not occurred in this case. In the "Standard of Review" and "Applicable Law" sections of his brief, Lee cites to the general, applicable case law and statutes.
However, in his "Argument" section, where appropriate citations must be included, Lee cites to five cases to support the two sub-arguments to his issue. Only three of those five cases are published. None of the three published cases cited actually exist in the Southwest Reporter. Each citation provides the reader a jump-cite into the body of a different case that has nothing to do with the propositions cited by Lee. Two of the citations take the reader to cases from Missouri. As the State points out, even Texas cases with the same names as those cited by Lee do not correspond with the propositions relied upon. [A footnote lays out the details, quoting the State's brief. -EV]
Further, as again noted by the State, the brief is devoid of any citations to the record. These deficiencies, although brought to the Court's and to Lee's attention by the State in its brief to this Court, were neither contested nor corrected by Lee in any kind of reply, amended, or supplemental brief.
{Based upon a recent Texas Bar CLE, "Have the Robot Lawyers Finally Arrived? Practical Concerns and Ethical Dimension of ChatGPT," presented by John G. Browning of Spencer Fane LLP, it appears that at least the "Argument" portion of the brief may have been prepared by artificial intelligence (AI). To avoid this problem, Federal District Court Judge, Brantley Starr, requires the following certification for pleadings filed in cases pending in his court [see this post for the text, which I omit here -EV] ….
Because we have no information regarding why the briefing is illogical, and because we have addressed the issue raised on appeal, we resist the temptation to issue a show cause order as a New York federal district judge did in Mata v. Avianca, Inc., 2023 WL 3696209, 2023 U.S. Dist. Lexis 94323 (S.D.N.Y., May 4, 2023, order), or report the attorney to the State Bar of Texas for a potential investigation for a violation of the State Bar rules.}
Thus, Lee inadequately briefs his sole issue on appeal. The failure to adequately brief an issue presents nothing for us to review, and we are not required to make an appellant's arguments for him. Accordingly, because Lee inadequately briefs his sole issue, it presents nothing for our review and is overruled.
I tried to reach the author of Lee's brief, Craig Alan Greening of the Greening Law Group; I will add his response if I get it. Thanks to Sam Cole for the pointer.
Editor's Note: We invite comments and request that they be civil and on-topic. We do not moderate or assume any responsibility for comments, which are owned by the readers who post them. Comments do not represent the views of Reason.com or Reason Foundation. We reserve the right to delete any comment for any reason at any time. Comments may only be edited within 5 minutes of posting. Report abuses.
Please
to post comments
Seems like people should have figured out by now that using AIs to write pleadings doesn’t work. Perhaps it just identifies plaintiffs and attorneys who are too dumb to defend themselves or their clients in court.
I wonder if any probate attorneys are using it to draft wills and trusts. There's a whole potential field of ticking time bombs waiting to go off if they are.
How much risk for a will or trust would there be? Do they usually include something about which an AI could hallucinate?
For something simple AI would probably be ok though I still wouldn’t trust it. Its biggest drawback is not know which questions to ask and how to analyze moving parts. If it doesn’t think through all possible contingencies, for example, the will could be worthless or, worse, produce a result completely different from what was intended.
I suspect quite a few of these cases have already been ingested by the legal system, so to speak, and are just sitting out there like time bombs. And, of course, while the AI *can* hallucinate, it doesn't always, (Or rather, yeah, it's always 'hallucinating', but a lot of the time the hallucinations happen to agree with reality.) so a fair number of people who've used AIs even without due diligence will get away with it.
Maybe they could announce a sort of "AI brief amnesty", a reasonable period for anybody who has done this to own up and make things right, after which any new cases discovered will be severely sanctioned?
An Israeli colleague of mine caught ChatGPT making up journal article citations that didn't exist.
Simple solution: The courts should let the attorneys know that they will be fined $100,000 for each instance of a fake citation they use, and the various bar associations should consider that grounds for discipline.
That would stop the crap right now.
As I understand it, to win a legal malpractice claim, the plaintiff must show a breach of the "standard of care" along with demonstrating how this breach resulted in an injury to the client. In a case like this one, even assuming that the use of an AI program to (help) prepare the legal documents breached the "standard of care", it would appear to be difficult to show the end result would have been any different.
However, in the extreme case of an attorney letting an AI program draft all the legal documents, could the client have a consumer fraud claim against the attorney? That is, (s)he did not do what I paid her (him) to do, even if there was no legal malpractice?
I have always wondered this when I see an opinion where an appeal is dismissed for failure to develop an argument, failure to cite the record, provide appropriate citations, etc. The client may not have suffered any prejudice, since the best lawyer in the country may have also lost, but at the same time, the client also didn't get what he paid for.
it would appear to be difficult to show the end result would have been any different.
You put your finger on why legal malpractice cases are so hard. Even if the lawyer messed up, the client also has to show that the malpractice caused him loss — meaning but-for the malpractice he would have been in a better position, AND the malpractice was the proximate cause of the loss. This is sometimes referred to as a “case within a case,” meaning you have to try the underlying case. It’s often very tough to prove.
However, in the extreme case of an attorney letting an AI program draft all the legal documents, could the client have a consumer fraud claim against the attorney?
Improper billing can also be a damage claim in a malpractice case. If the lawyer bills you for time he did not put in, you have a claim. I can see someone arguing that paying a lawyer to do work that he does through AI is false billing. Comparable to him asking a paralegal to do the work and then just submitting it.
AFAIK, there have not been cases like that. Yet. AI is still young.
They're so hard because you have to convince lawyers to sanction lawyers on behalf of non-lawyers.
Malpractice suits are decided by juries.
Oh, I'm sorry, I thought there were judges involved, my bad.
Is the court empowered to give the client a lawyer who can write a brief in accordance with the rules?
The adversary system has its advantages, but saddling a client with an attorney who doesn't write a proper brief isn't one of those advantages.
The client didn't go and say "I want a lawyer who will forfeit my case."
How would we explain to a Martian, visiting America to study the justice system, how this sort of thing is OK?
Some states interpret their consumer protection statutes to exclude matters relating to the quality of professional services—“the learned profession” exception. Some matters, such as fraudulent billing, might still be covered by consumer protection laws if all the other elements of the statute are met.