Lawyers Should Use AI as a Tool (ChatGPT and Google Bard are quite impressive!) – Not To Misrepresent the Law
As the Age of AI dawns, there is concern that it will take over much of the work we all do and make us less important and perhaps useless. We even fear it will be intelligent enough to overpower us all. But let us not overreact. AI has been and continues to be useful.
In our experience. ChatGPT and Google Bard provided accurate and useful answers to our legal questions. Even Google provided solid answers. They did not generally provide case law but sometimes, as the example discussed below illustrates, they found cases directly on point. For caselaw, Westlaw, Lexis and Google Scholar remain the best source.
But would some lawyers try to use AI to deceive their opponents and the courts? Apparently so.
Misuses of AI
This seems astonishing. It is hard to imagine that a lawyer would file papers that a computer had generated using ChatGPT or Google Bard, and never even bother to check the legal principles they wrote in papers filed with the Court. Well, it is happening.
A New York Times article recently tells of a lawsuit where a lawyer had used ChatGPT to prepare a legal brief. So, Have You Heard the One About the Lawyer Using A.I.? New York Times, May 29, 2023, Here’s What Happens When Your Lawyer Uses ChatGPT. “A lawyer representing a man who sued an airline relied on artificial intelligence to help prepare a court filing. It did not go well.” May 27, 2023, by Benjamin Weiser
According to the article, when a defendant in a lawsuit asked a Manhattan federal judge to toss out a case, the lawyers on the other relied on contrary case law. But as the article explains:
“There was just one hitch: No one — [not the defendant’s] lawyers, not even the judge himself — could find the decisions or the quotations cited and summarized in the brief.
That was because ChatGPT had invented everything.”
Michael Cohen Joins In
“Donald Trump’s former fixer had sought an early end to court supervision after his 2018 campaign finance conviction. He enlisted the help of Google Bard.
The fictitious citations were used by the lawyer in a motion submitted to a federal judge, Jesse M. Furman. Mr. Cohen, who pleaded guilty in 2018 to campaign finance violations and served time in prison, had asked the judge for an early end to the court’s supervision of his case now that he is out of prison and has complied with the conditions of his release.
The ensuing chain of misunderstandings and mistakes ended with Mr. Cohen asking the judge to exercise ‘discretion and mercy.’”
Long before there was any thought of something called artificial intelligence or more specifically Chat GPT, or Google Bard, lawyers have been have been accused of misstating or distorting the law and sometimes even provided phony case citations. This will be the new excuse for any misstatement of the law. The lawyer can say he used Chat GPT, and it made a mistake. Of course, that will not matter because that is never going to be an excuse for filing false statements of the law.
Our experience with Chat and Google
To evaluate the credibility of Chat GPT, we asked it to find us one of the cases cited in the airline and it responded that the case was bogus and may have been used in a legal proceeding in another matter. So, it looks like Chat GPT was re-educated. We also tried to figure out how you could get Chat GPT to produce false citations.
We were unable to get it to provide much case law, but we believe that if you asked it to provide a brief on a subject and then cases to support the issue, Chat GPT or Google Bard might very well produce false citations. In other words, it may produce a brief stating what the user advises is the law and then fill in the case citations based on what the AI has been told by the person using it.
But let’s not be down on artificial intelligence, ChatGPT, Google Bard. Bing or Google. Here is an example, a client asked a law firm whether insurance policies cover the cost of complying with an injunction. There is caselaw but it is complex and requires careful analysis of the cases one finds in Westlaw.
Evaluating the AI Search Engines
Out of curiosity, we punched in the issue of whether insurance policies cover the cost of complying with an injunction into Chat Bing. Guess what? In a split second it produced the precise cases on this subject which were also found using Westlaw.
More recently, we asked the same question of Google (not Google Bard). This the response:
This measure is the appropriate remedy in any situation in which, if the defendant is not ordered to cease performing an action, the plaintiff will be unable to be properly compensated. Liability insurance policies typically provide coverage only for “damages,” not for injunctive relief.
We then asked Google Bard. It gave five answers which varied to provide more insight, The responses were “nuanced”…
Whether insurance policies cover the cost of complying with an injunction is a complex question with no definitive answer, as it depends on several factors, including:
- The specific wording of the insurance policy:
- Certain policies, such as some commercial liability policies, may explicitly exclude coverage for the costs of complying with injunctions. Others may define “damages” in a way that could encompass compliance costs. Carefully reviewing your policy’s language is crucial.
- The type of injunction:
- Some courts have distinguished between mandatory injunctions issued in response to legal claims seeking monetary damages and those seeking purely equitable relief, like correcting a wrong or preventing future harm. Coverage for complying with the latter may be less likely.
- The underlying claim:
- The nature of the lawsuit that led to the injunction can also influence coverage. For instance, environmental cleanup costs ordered by a court might not be covered under a general liability policy, while legal fees incurred to comply with an antitrust injunction might be.
This is impressive. (Try the search yourself and you will see the full range of its answer.)
Finding “Obscure” Phrases
Let us try another example. Chat Bing is widely available now. A lawyer wanted to write something about a statement in an article that was never published in a journal. It was simply a paper that was known internally in the industry where it was written. The three words were “taking the suddenness out of accident.” Chat Bing could not find those words or that phrase and said that it did not exist as far as it knew and asked if another question might be more appropriate.
Making the same search through Google, which does not purport to be a very sophisticated AI device but certainly one that all of us rely on all the time, produced a different result. Google immediately found that phrase in a legal document which referred to the unpublished paper and the phrase, “taking the suddenness out of accident.”
Do not Fear AI – Master It
So, what does all this tell us? Stop worrying about AI or Chat GPT and start thinking about human beings and how they should manage these incredibly useful and sophisticated devices. They are not going to take over the world. And they certainly are not going to destroy it. We can do that ourselves.