AI In The Hands of Pro Se Litigants
What Lawyers Need to Consider In Regard to Pro Se Litigants Using AI
Early in my career I worked at a firm where we represented some governmental entities such as Maricopa County, Arizona. As a young associate I got to respond to motions filed by incarcerated persons who were not represented by an attorney—pro per or pro se1 litigants. I dislike litigating cases against a pro se opponent. There have only been a few occasions where I had to. Not only is it difficult, unnecessarily time-consuming, and frustrating since they often can’t be swayed by strong arguments and sound logic, but in addition, I feel bad for them. It seems unfair that one’s access to justice should be tied so closely to one’s ability to pay for legal representation.
I tried to treat pro se litigants fairly and not take advantage. Going up against a pro se litigant, however, could be unfair to my client who had to pay me to respond to pointless, frivolous motions. It was almost certain that my client would not be able to recover attorneys’ fees from the prolific pro se litigant, determined to make up in number of filings what they lacked in legal knowledge. To preclude losing on the doctrine of waiver of an argument, we as lawyers must respond to all arguments and motions filed by pro se opponents. One time I found myself responding to a motion titled, “Motion Why the Judge Is An Idiot.” Yep, that was the exact title. My pro se opponent was angry that the judge had ruled in my favor in a commercial lease case. Could I understand why he was upset at losing? Sure. But did he imagine that was going to benefit him in the eyes of the judge, the trier of fact? Surely not.
With the easy availability of generative AI tools like ChatGPT, Gemini, WatsonX and many others, pro se litigants finally have access to tools that will help them in their efforts to pursue their legal objectives. Or do they? AI tools are most effectively used in legal settings by those with legal knowledge, so will AI in the hands of legal novices be more harmful or helpful? Here are some ways that AI tools will benefit pro se litigants, and some ways that they will likely detriment pro se litigants.
AI’s Positive Impacts
There are several potential benefits of pro se litigants using AI tools. Chatbots could help create concise, understandable legal documents such as small claims complaints. But beyond the bounds of simple chatbot assistance, there are other, more varied benefits. Some are clearcut – others are more nuanced.
1. Enhanced Understanding: GenAI tools can help pro se litigants better understand legal terminology, procedures, and their own cases. For example, if a person is served with a complaint but doesn’t understand it, they could feed it into an AI tool, and receive an understandable explanation. They could even have the explanation crafted at a grade 6 reading level or ask for it to be phrased in poetry in the style of Elizabeth Barrett Browning or in the writing style of Stephen King, or the “voice” of Elvis Presley. They could ask for examples and further explanation of anything that was still unclear. This one benefit alone could reduce a substantial barrier to entry for those intimidated or bewildered by the legal system. Similarly they can feed their opponents’ motions into their AI tool and ask it to explain the arguments to them, ask for counter arguments, and even a counter-motion. Again, a huge boon to the untrained in law.
2. Higher Quality Decision-Making: A better understanding of how the law applies to the circumstances in their situation could help pro se litigants make more informed decisions, potentially leading to better outcomes. AI might persuade one to decide against filing a lawsuit, or might help another to add in pertinent claims that the person was not aware of. For example, in a fictional case I was asking for help with, ChatGPT 4.o added the claim of “breach of fiduciary duty” to my breach of contract claim.2
Ostensibly, any credible AI tool would have helped my unrepresented opponent come up with a better argument or motion than the one entitled “Motion Why the Judge is an Idiot.” Or perhaps at the very least AI could have made the argument more substantial and grounded.
3. Document Preparation: AI tools can assist in drafting legal documents, filling out forms, and even generating pleadings, which can be overwhelming and intimidating for those without legal training. When I asked ChatGPT 4.o to help me draft a complaint against my fictional neighbor to get my fictional lawnmower back, with a bit of prompting it prepared the complaint, a summons, gave me the address and phone number of the court where I should file it, and told me about how much it would cost to file it. It also gave me instructions on having it served. Overall, there is a great deal of benefit to pro se litigants in using AI for document preparation.3
Negative Impacts
The positive impacts of AI tools for those not able to afford an attorney are significant. But the negative impacts of untrained persons using AI as a substitute for a lawyer also exist, and are often under-recognized, or at least under-acknowledged.
1. Over-Reliance: It is easy to imagine how pro se litigants might over-rely on AI tools without understanding the nuances of their legal situations, thereby leading to suboptimal decisions. For example, not being aware of deadlines, such as the time within which to file a response to a motion, or the requirement to join a necessary and indispensable party, or available legal tools such as discovery requests. These are things that an AI tool could inform the layperson about, but not if the layperson doesn’t know enough to ask about them.
2. Complex Cases / Issues: While AI can help with straightforward matters, complex cases requiring deep legal expertise--or even moderately complex cases--will still be challenging for pro se litigants to handle effectively even with AI assistance. Again, AI in law is most effectively used by those with legal knowledge. AI tools might also miss more nuanced arguments or subtleties in caselaw.
3. Unawareness of Risks: Pro se litigants might misinterpret AI-generated advice or fail to recognize the limitations of these tools, potentially harming their cases. As referred to above, they might not be aware of deadlines, calendaring requirements, or understand evidentiary burdens. In other words, they won’t know what they don’t know. Recently a lawyer was asking in a Reddit forum for advice about how to respond to a pro se litigant’s motion packed-full of bad citations. More than one colleague suggested that the lawyer point out the inaccuracies in a motion for sanctions. AI tools caution that they can make mistakes, but a pro se litigant might not be aware of the problem of fake citations, and likely wouldn’t know how to check the citations for accuracy and applicability. If the judge grants a Rule 11 motion or some other motion for sanctions a pro se litigant could be detrimental by their use of AI.
4. Ethical and Procedural Issues: Improper use of AI tools, such as inadvertently submitting documents generated by AI without appropriate review, could lead to ethical or procedural issues in court. Aside from just incorrect legal citations, a pro se litigant using AI could unintentionally disclose information that should be protected such as a location, a social security number, or some other thing that amounts to an ethical if not legal violation. Or a pro se litigant using AI could unknowingly make admissions that hurt their case. And pro se litigants can be detrimental by being sanctioned for false or inaccurate citations. Pro se litigants might also put themselves at risk of being manipulated or taken advantage of by unscrupulous or malevolent lawyers who exploit deadlines, complex procedural issues, nuanced court cases, or unfair settlement terms, believing that AI will protect them from these possibilities.
Pro se litigants who are tech savvy might obtain an illicit benefit from AI tools. It is not outside the realm of possibility that a pro se litigant would have access to AI while in the courtroom. For example, having a friend or family member surreptitiously use an AI tool to record witness testimony then ask it to point out holes in the testimony, or provide a list of questions to cross-examine a witness. This use would simultaneously put the attorney opposing them at an unfair disadvantage, since lawyers are not be allowed to do the same in a courtroom.
Conclusion
While GenAI tools hold a lot of promise for pro se litigants, as Mark Twain is attributed with saying, "It ain't what you don't know that gets you into trouble. It's what you know for sure that just ain't so." Although Mark Twain might not have originally said those exact words, the quote effectively conveys the idea that overconfidence in false information can be more dangerous than acknowledging one's ignorance. It is crucial to recognize that pro se litigants, while benefitting from their use of AI tools, will not understand its limitations and the importance of human oversight. Although AI tools will typically warn users that they should seek outside legal help, or talk to a legal professional, those mild statements are unlikely to provide sufficient caution to an angry, scared litigant who cannot afford legal counsel. The legal profession has work to do in adapting to the opportunities and perils that AI tools bring to the pro se litigant.
"Pro se" meaning "for oneself" or "on one's own behalf" in Latin; and "In propria persona" often abbreviated as "pro per" meaning "in one's own person" in Latin. They will be used interchangeably to indicate a person or entity in a legal proceeding without an attorney or legal representative.