by Dr. Lance B. Eliot
For a free podcast of this article, visit this link https://ai-law.libsyn.com/website or find our AI & Law podcast series on Spotify, iTunes, iHeartRadio, plus on other audio services. For the latest trends about AI & Law, visit our website www.ai-law.legal
Key briefing points about this article:
- There is much debate about the possibility of AI serving as judges in our courts
- Some argue it should never be permitted, others say it will be inevitable and unstoppable
- One argument being made is that perhaps AI judges would be solely used in the lower courts
- The logic for use at the lower courts includes scalability and consistency
- Countering points are that this drains humanity from the lower courts and is a slippery slope
Do you lay awake at night due to the notion that someday we will have AI-based autonomous judges in our courtrooms, looming over humanity and deciding the fate of humans that come before their grand majesty?
So that you don’t needlessly lose an excess of sleep, please be aware that this kind of AI is still only a gleam in the eye of AI developers. A long and bumpy road has to be first crossed to get there. That being said, it is important to realize there is semi-autonomous AI that might reach that goal line sooner than we think.
In the case of AI that is semi-autonomous and likely to be nothing more than a supplemental aid to a human judge, there is little argument that such high-tech would be a handy sidekick. Perhaps such AI could tidy up the loose ends of any legal arguments or provide insights about precedents via being able to readily peruse thousands upon thousands of prior cases. Generally, there is minimal controversy associated with arming the judiciary with so-called “smart” computer-based systems that serve at the will of human judges and can ostensibly bolster adjudication toward greater efficiencies.
The real war of words arises when you start thinking about fully autonomous AI that imbues legal reasoning and can seemingly render judicial decisions without the need for human hands or minds. That type of AI would not be supplemental and instead would outright replace human judges (this is not to suggest that there would only be AI-based judges, as there might be a mixture of sometimes the use of a human judge and in other instances the use of an AI one).
For details on this and other AI and law topics, see my book entitled “AI and Legal Reasoning Essentials” at this link here: https://www.amazon.com/Legal-Reasoning-Essentials-Artificial-Intelligence/dp/1734601655
The shorthand version of the spirited controversy is whether we should allow AI on its own to serve in any such judicial capacity.
Some argue that there will never be a viable means for this AI autonomy to be utilized, even if it can be crafted (none of this heightened AI yet exists). Others take a decidedly opposite tack and emphasize that we could use AI across all levels of the courts and in any and all judging roles. Assuming that AI can be established to provide the equivalent of what human judges do, the hope by proponents of this kind of technology would be to use it broadly and deeply throughout our courts.
Another viewpoint is that maybe these autonomous AI judges should be confined to a particular focus or segment of our courts. At the International Bar Association (IBA) annual conference in November 2020, Dean Tania Sourdin of Australia’s Newcastle University Law School remarked that the lower courts might be the appropriate setting: “‘It seems to me that judges at lower court levels are likely to be phased out” (see The Law Society Gazette, November 9, 2020 article entitled “IBA 2020: Robots Don Black Cap For Lower Court Judges).
This lower court infusion of AI-based autonomous judges might not necessarily apply across-the-board entailing all areas of the law. Professor Sourdin indicated that family law and various aspects of criminal law are likely carveouts. Legal avenues that seem particularly amenable would be “the simple civil cases, possibly personal injury cases, and certainly very simple contractual matters,” according to Sourdin.
Okay, for sake of argument, let’s take at face value the plausibility of using AI-based autonomous judges in the lower courts. Meanwhile, agree too that such AI judges won’t be used at the appeals courts and thus intentionally not applied in the higher courts.
What might be the line of argumentation that supports this angle or slice of mechanization and what might be the counterarguments?
Going Toe-to-Toe On The Lower Courts
We’ll start with the foundation in favor of the AI autonomous judges as serving only at the lower courts level and thus distinctly not being used at the higher courts. Notice that there is no need to delve into the argument of not using such tech at all as judges since we are momentarily stipulating that we’ll forego that qualm. Presumably, without undue bickering, let’s assume that this kind of AI is going to be used somewhere in the courts, albeit in a narrower rather than an all-encompassing manner.
First, a popular supportive argument is that the use of AI at the lower courts would tremendously help in coping with the voluminous amount of court activity that occurs in the lower courts. There are ongoing calls that more judges are desperately needed, along with the decrying of justice potentially being delayed as a result of an insufficient number of qualified judges at the ready. AI-based autonomous judges could easily scale-up or scale-down, being on the go and always available for whatever need for judging might be at the lower courts (i.e., the AI provides the supply to always meet the demand).
Secondly, the AI would provide a semblance of consistency that is unachievable with the use of human judges. Human judges act not simply by the nature of the law, they also embody personal preferences and styles as permitted within the range of the law and the myriad of court procedures. There is also the point made that human judges are in fact human, meaning they are subject to human emotions and human foibles. Many assume that AI-based autonomous judges won’t be (this is a contested topic since there are indeed real concerns about AI-related inherent biases).
Thirdly, by having human judges at the higher courts, we preserve our overall sense of the law and can assume and rely upon those human judges to be a backstop or failsafe. The assumption is that even if the AI goes astray, doing so on occasion at the lower courts, the human judges at the higher courts will catch these maladies and decisively rejuvenate justice accordingly.
Now, let’s wade into the counterarguments.
For most citizens, their first contact with the judicial system arises at the lower courts. At that point of entry, assuming there are these widespread AI judges, the public at large is being handed over to a machine, as it were. Some eschew the lack of humanity that would be conveyed and the cold and calculated message that it would send.
Another concern is that the number of appeals might grow astronomically. The lower courts would simply become a spitfire feeder and overwhelm the higher courts.
And, for some the most frightening aspect is that the lower courts are nothing more than the start of a slippery slope. If this AI does a sufficient job at the lower courts, you can anticipate that the higher court days of human judges are sorely numbered.
For those of you that are pining away about these AI-based autonomous judges, the good news is that you’ve got time before this capability of AI emerges.
Meanwhile, do not be surprised to see the supportive style semi-autonomous AI that soon becomes the best pal of human judges, and be ready to make your case to a bench that has got a double-punch of both human and AI sitting there in judgment.
For the latest trends about AI & Law, visit our website www.ai-law.legal
Additional writings by Dr. Lance Eliot:
And to follow Dr. Lance Eliot (@LanceEliot) on Twitter use: https://twitter.com/LanceEliot
Copyright © 2021 Dr. Lance Eliot. All Rights Reserved.