In his essay “AI Disorder in the Courtroom,” Oliver Bateman raises several concerns about Artificial Intelligence (AI) in both the legal profession and various elements of courtroom procedure or evidence. Technology is rapidly evolving and inevitably disrupts existing norms and processes; however, such disruption should not be presumed to result in disorder. While such disruption can bring challenges, focusing too much on potential problems or rushing to make decisions while the technology is still novel may prevent the greater benefits that the technological disruption could bring to the legal profession and legal system.
AI is far from the first novel and potentially disruptive technology that the legal system and the legal profession have had to encounter. Similarly, it is far from the first time that concerns about the potential impact of new technology on the legal system and beyond have occurred.
I’d suggest two key elements regarding the potential benefits of AI in the legal profession as well as how to counter the potential risks are largely missing from Bateman’s essay—and much of the conversation around AI in the legal profession. Firstly, the essay lacks an equal treatment of the potential benefits of the technology for citizens, practitioners, and the system as well as the risks it might propose. Second is the important question of how the concerns about AI are already addressed, and what these concerns are compared to the potential risk of harm in the current system. In short, AI may merely reflect or amplify existing concerns or disorder. This relates both to the potential democratizing effects of AI in the legal profession as well as its use in the courtroom and criminal justice process.
We should not see AI as a silver bullet. We should allow, however, for the development of norms rather than a rush to presume the worst-case scenarios in any profession or application. We also must be cautious not to let our early fears dictate the evolution of the technology.
The Potential Benefits of AI to the Legal Profession
Like many professions, the legal profession is faced with the disruptive effects of AI. The disruptive capabilities of AI should not automatically be presumed to be negative to either clients or lawyers. While Bateman refers to the “Clarence Darrow problem,” this neglects to fully recognize the current gap in access to legal services as well as the potential democratizing effect to empower the average citizen in their interactions with the legal system. In fact, positive uses of AI can improve access to legal services and the ability of lawyers to better serve their clients.[1]
One way AI may be able to improve the legal system is in helping fill the current gaps in legal services that exist between those who need legal services and the amount of access to legal services individuals can pursue due to a variety of factors. Numerous people lack legal documents—like the power of attorney or a will. They may have questions about business formation or choose not to assert their legal rights in cases where they do not have the resources or see obtaining legal services as burdensome. Particularly among low-income Americans, nearly half of those choosing not to seek legal services cited cost as a reason. More than half doubted their ability to find a lawyer they could afford should they need one.[2] AI should not be viewed as a complete replacement for lawyers and the legal profession even in these types of cases; however, it could provide those currently without access to legal services with better knowledge of when they might need representation or with some degree of legal assistance.
AI should not be seen as a substitute for the full services of an attorney, nor will it likely be able to handle complicated and nuanced cases. But it could provide some guidance or stop-gap in many scenarios where the alternative might be a lack of legal services or a poor self-help attempt. AI-enabled tools could help people understand when a situation might actually merit legal action. They could assist those currently awaiting more formal legal services in understanding what to retain or what information their attorney may need, or in guiding through basic questions around common legal needs like writing a will, forming a business, or challenging a parking ticket.
This should not be confused with a complete “robo-lawyer” replacement but instead viewed as a way to improve access to the many basic legal needs that go underserved. In this way, AI could provide an incredibly democratizing aspect to legal services and empower more people to understand and access the legal system. It might even create more work for lawyers in some cases as people recognize how such services might be helpful or feel more empowered in asking legal questions or pursuing legal claims.
This democratization could also benefit attorneys in their current ability to serve clients and provide greater services. For example, the request for free or reduced services via legal aid often comes with significant wait times or limited options regarding the types of services provided. Many public defenders’ offices also face significant strains due to demand, including in situations where the case burdens have been such that it may be difficult to provide adequate counsel.[3] Using AI would allow lawyers to focus on the complexity and nuances of transactions and cases or find precedents they might have otherwise missed. This would be particularly useful in most many firms, where there are a small number of attorneys, and they may not have the same full research access or assistance as a large firm.[4]
There are, of course, ways AI could also improve the legal system itself. This includes basic customer service tasks or helping attorneys or individuals connect with the appropriate legal resources, improving scheduling, and automating tasks like copy-editing opinions.
All of these elements would be, to some degree, disruptive on the current legal system and may even call into question existing norms around what qualifies as practice of law. However, this disruption would not necessarily create disorder. Instead, it could create greater empowerment, agency, and efficiency within the legal system.
AI in The Courtroom Has Dealt with New Technologies and Many of the Concerns are Old Questions
How technology should be allowed in the courtroom is not a question that has emerged only with AI. We can think of any number of technologies—from photocopies to video to the internet—that have raised questions around evidence. In many cases, the law is struggling with the role these technologies have in the courtroom at the same time. Just as outside of the legal system, these concerns are not always novel to AI but may just be a new application of existing concerns or require clarity in existing norms or rules.
Some of the basic concerns around the use of technology to engage in creating forgeries or fake evidence and hallucinating false precedents are likely already covered by existing rules of professional conduct, rules of evidence, or even applicable laws around fraud or false allegations. If there are concerns that AI could create a loophole rather than rush to create AI specific rules or laws, we should first examine if the existing issues could be updated to clarify that using AI does not remove existing obligations.
In the limited cases where courts have encountered such issues, existing rules have been able to deal with the concerns. For example, attorneys have been fined for AI-generated court documents[5] and held responsible for briefs where AI hallucinated non-existent court cases.[6] This is without any changes to existing rules and shows that if the harms occur from the application and not the underlying technology, we will largely be able to adapt.
Many of the concerns about AI in the court system are not problems of AI, but larger problems of society. There are legitimate concerns about the potential biases in AI used in sentencing or detention, and they certainly should be addressed.[7] Many of these concerns, however, are reflections of our existing issues in society and are not unique to AI. If identified in AI, they may be more rapidly addressed in reprogramming than in the often decades-long processes of seeking to combat implicit and explicit bias in society.[8] Transparency in such algorithms and human oversight when individual rights are involved will be critical, but it should not be assumed that humans do not fall prey to such bias or prejudice themselves. Ideally, transparency would enable awareness and, when needed, changes in an algorithm while humans themselves remain a bit of a black box. It is often difficult to know if a judge’s decision may be based on when in the day he or she is making a decision, the past interactions with an attorney in a different case, or implicit bias about a particular group that might not be sufficiently caught in existing redress mechanisms.[9]
Bateman’s essay does highlight a particular example of where AI may raise some questions around how to communicate its use and impact when it is used in re-enactments or, in the example referenced, to create a victim impact. The reaction here is seen to have a greater emotional impact. As with many scenarios, including those around re-enactments in the pre-AI era, norms around disclosure of the technology to judge and jury will likely be useful to assist in weighing the persuasiveness of the AI-created statements or re-enactments. This will likely vary based on the specific use of AI in any particular situation. For example, the persuasiveness of a fully AI-generated statement or re-enactment would likely be different than one that translates a recording from another language or reads an existing statement. As a result, we should be cautious about one-size-fits-all approaches to a general-purpose technology and instead look at the particular application.
Conclusion
Rather than think of AI’s impact on the legal profession as bringing disorder, we should be a bit more optimistic about AI’s potential disruption and its ability to benefit citizens, practitioners, and the legal system. AI has a diverse array of applications both generally as well as in specific industries and situations as seen in the legal system. With that in mind, it is better to look at specific applications and concerns rather than a more general approach. While there are potential risks, many of these are pre-existing and not unique to this new technology. They must be weighed against both the potential benefits as well as the existing imperfect system. As such, we should be cautious about overestimating the risk or the inability to adapt to new technologies and consider the potential benefits.
Jennifer Huddleston is a senior fellow in technology policy at the Cato Institute.
[1] See Jeff Neal, “The Legal Profession in 2024: AI,” Harvard Law Today, February 14, 2024.
[2] Legal Services Corporation, “The Justice Gap: The Unmet Civil Legal Needs of Low-Income Americans,” 2022.
[3] See American Civil Liberties Union, “ACLU Sues Over Public Defender Shortage and Resulting Wait List in New Orleans,” January 15, 2016.
[4] Brett Burney, “2024 Litigation and TAR TechReport,” American Bar Association, April 30, 2025.
[5] Jonathan Limehouse, “MyPillow CEO Mike Lindell’s Lawyers Fined for AI-Generated Court Filing,” USA Today, July 8, 2025.
[6] Maya Yang, “US Lawyer Sanctioned After Being Caught Using ChatGPT for Court Brief,” The Guardian, May 31, 2025.
[7] Richard Hua, “AI and Racial Bias in Legal Decision-Making: A Student Fellow Project,” Harvard Law School, March 19, 2025.
[8] Adam Thierer and Jennifer Huddleston, “Finding Our Humanity With AI,” U.S. News & World Report, January 2, 2018.
[9] See Kurt Kleiner, “Lunchtime Leniency: Judges’ Rulings Are Harsher When They Are Hungrier,” Scientific American Mind 27, no. 4 (September 2011): 7.
About The Author
