Just Delete It?

In August 2025, I noted the hubbub surrounding two federal judges in Trust but Verify (August 2025). These instances involve federal judges "in Mississippi and New Jersey." Judge Henry Wingate (78) of Mississippi was appointed in 1985 by Ronald Reagan. Judge Julian Xavier Neals (60) was appointed by Joseph Biden in 2021 and has been on the bench for four years. What do the two share? Well, neither one is part of the tech-savy generation "alpha" or "z." Judge Neals is Generation X and Judge Wingate is a Baby Boomer, but an early Boomer. They each made the news for artificial intelligence and its implications.

Repetition of one critical point is necessary in every post—you cannot trust artificial intelligence. It is a boon, a powerful tool, and it should be used. As time passes, it will become ubiquitous and will be more commonly used and relied upon. I was shocked on a recent AI presentation when the sponsor deployed an impromptu poll, "Are you using AI?" and the results did not support universal commitment to the technology. That will change. That said, more than half of the poll respondents admitted to currently using it.

More recently, reports noted that the U.S. Senate Judiciary Committee inquired about the circumstances surrounding Judges Neals and Wingate. Both federal judges explained that the "error-ridden" "decisions ... did not go through their chambers' typical review processes before they were issued." Simply stated, there was trust there, but insufficient verify.

Judge Neals explained that "a law school intern used OpenAI's ChatGPT for research without authorization or disclosure." Judge Wingate similarly placed responsibility on "a law clerk ... (who) used Perplexity 'as a foundational drafting assistant.'" There is no explanation in the news report regarding why an intern or a law clerk was releasing decisions in any event, or how that would occur without the review and agreement of the judge.

Some will believe that all case decisions are made by judges. That should be a fair and reasonable assumption. Judges are elected and appointed to make difficult decisions. The news article seems to suggest that in at least these two courts, decisions are made and published by interns and clerks. Delegation is one thing, but responsibility is another. No decision is issued by the OJCC except by a judge. Perhaps that is true more broadly (most tribunals), or perhaps our judges are somehow an exception?

Some will take my thoughts as criticism. Certainly, others have been more direct in their thoughts regarding the Judges, their decisions, and their deflection onto staff. See Admit the Obvious, Above the Law. Nonetheless, my thoughts are not a criticism of judges or courts. That said, lessons can be learned by all whenever any of us makes an error. 

Federal judges are governed by the Code of Conduct for United States Judges. This is arranged in canons, as are the various state codes that have been periodically discussed here. Canon 2(A) reads:

(A) Respect for Law. A judge should respect and comply with the law and should act at all times in a manner that promotes public confidence in the integrity and impartiality of the judiciary.

Canon 3(a)(5)

(5) A judge should dispose promptly of the business of the court.

There is no provision that suggests that law clerks and interns should dispose of the business of the court. 

There is the potential that decisions being published by clerks and interns could impact "public confidence" in courts and their decisions. As has been frequently emphasized, these Codes preclude not only impropriety but also the "appearance of impropriety." The commentary to Canon 2 includes: "A judge must avoid all impropriety and appearance of impropriety."

Judge Wingate concluded that the failure in his chambers "was a lapse in human oversight." That is, a failure in the "verify" element of the Trust but Verify suggestion. Judge "Neals said his chambers has since created a written AI policy and enhanced its review process." Potentially, this means interns in that chamber are no longer issuing decisions.

Judge Wingate, caught in the wake of this mistake, reportedly deleted "the order (that) was riddled with mistakes" from the case docket, published a new order, and backdated that new order to the date of the original hallucinated one, according to News from the States. Does deleting the error make it go away?

Each judge asserts that they have instituted new processes to assure the errors will not be repeated. According to the Magnolia Tribune, Judge Wingate's solution is simple. Now, he assures, 

all draft opinions, orders, and memorandum decisions undergo a mandatory, independent review by a second law clerk before submission to me.

Does this mean that the previously published decisions were, in fact, submitted to Judge Wingate before publication? If so, responsibility for the hallucinations rests with the judge. 

Would any tribunal, anywhere, excuse a lawyer who signs a document full of hallucinations on the premise that "my paralegal" or "my associate" wrote this and I just signed it? I am dubious regarding the efficacy of such an excuse.  

What does "verify" mean? There has been seeming confusion about that in the world of workers' compensation. I have striven to deliver suggestions in A Tool Kit (October 2025) and Simple Steps (October 2025). There are various methods that will help you to ensure the absence of hallucinations in your filings.

What does "verify" not mean? Well, that you had one paralegal, clerk, or intern check the work of another paralegal, clerk, or intern is not likely compatible with your personal professional responsibility. 

Similarly, knowing that AI hallucinates, would you ask a second AI (that hallucinates) to verify the output of your primary AI? Doubtful anyone would view that as a rational process. 

No, in the end, the signatory (judge, lawyer, engineer, doctor) is responsible. They can perhaps delegate some of the work, but it is their job to do the verification and verifying. In these two instances, as in all occasions where some staff, associate, or partner does work, it all comes down to the person who signed the final document. The buck must stop there. 

The Nancy Reagan method is most effective ("just say 'no'"), meaning just don't use AI. I highly recommend this one regarding legal research, writing, and pleading. Use AI to check grammar, spelling, or similar. Do not trust it to do your work for you.  

Second, there is the Ronald Reagan method, "Trust but Verify," which means the person who is hired to do the work (judge, lawyer, engineer, etc.) may delegate but must remain personally responsible. 

Finally, there is the laissez-faire method in which this new and proven unreliable technology is engaged and trusted without question, as are clerks, interns, paralegals, and more. I cannot recommend that last one, but the decision is up to you. 

The person signing the document makes the decision. Appropriately so. It is their name that will be on the line. See Do You Care About Reputation (June 2023). The lawyer, doctor, engineer, judge, or other person who signs the document is responsible. This will remain whether you claim your dog ate your homework or your paralegal, intern, or clerk made mistakes in judgment or process.

The paralegal, clerk, intern, associate, partner, etc., will not be the one whose reputation is impacted, impugned, or ruined. Review the Reuters article. What are the names of the clerk and intern in these chambers? They are not identified. Various searches for their identities have come up short. In this instance, it appears that the judges instead will be the ones impacted. That said, the names of clerks and interns may eventually surface. 

It will persistently be the lawyer, doctor, engineer, or judge who is embarrassed or liable for errors. The implications may be an "error-ridden decision" published by an intern, a hallucination-filled pleading filed by a paralegal, or a set of structural plans completed and mailed by an engineering intern. The result will likely be similar in each instance: embarrassment, ridicule, or worse. 

Will the world remember the name (or even ever know the name) of the paralegal, clerk, or intern? Unlikely. Various queries have failed to identify either the clerk or intern by name here. Will those called upon to review the results take pity on, or commiserate with, the judge, engineer, or lawyer who seeks to cast blame on their staff, team, or contributors? As unlikely. In the end, the world will look to the professional to be right, diligent, and thoughtful in the process for which they are paid.

It is entirely possible that the world will hold those people to a standard that it would not hold itself. Will judges who fail to supervise be empathetic towards lawyers appearing before them who have similarly failed to supervise their staff?

Will it take a bridge collapsing through unchecked math? Will it take a lawsuit dismissal? Will it take a lawyer disbarred, a judge impeached, or an engineer stripped of a license? What must happen before professionals take the challenges and shortcomings of AI-produced material seriously and check the work on which they are placing their name?

It is encouraging that each of these judges has implemented plans to prevent these errors from recurring. Every lawyer, engineer, accountant, doctor, etc. should take heed of the potential AI embarrassment to themself, their office, and their professions. A sound AI policy is critical. And no, having some other clerk or AI check someone else's (or some other AI's) work is not a policy. 

You can delegate tasks, but not responsibilities. When you get caught, deleting it will not make it go away. Deleting it will just make for more questions. 

Another Arms Race

AI will change much of our world. There will be opportunities for measurement and evaluation not yet dreamed of. One example has already begun and provides an overview and perspective worth our consideration. There will be more examples as industry and individuals employ tools to their individual benefit. 

The news recently provided insight into a device that uses multiple cameras to inspect vehicles. It has an array of cameras and can identify, evaluate, and memorialize conditions. The obvious deployment for this is likely vehicle fleets, such as car rental. But the implications are likely much broader from the perspective of safety, maintenance, and repair.

The news story focuses on the rental car application and provides a photo of a car being driven through an arched scanner of many cameras. The vehicle is thus viewed from all angles and analyzed by an artificial intelligence (AI) tool, and assessed each time it is returned. It is not clear whether the car is scanned as it leaves the facility.

In this instance, the AI tool found a small ding on the car. Perhaps unnoticed or even unnoticeable. The customer was charged for the damage. Was it there when the car was rented? Did it occur during the rental when the vehicle was parked? Did it occur in the rental facility between rentals? Whose responsibility is it?

Suppose the car is scanned only when driven into the rental agency (return). In that case, any damage that occurs to the vehicle while it is in the rental facility might go unnoticed or unnoted until the next time the car is returned. A customer might be more comfortable with the process and the technology if a scan occurred upon departure and return. Or, perhaps the customer must look out for themself? Caveat emptor?

In this instance, the customer was later notified by Hertz of a charge for repair related to a minor cosmetic flaw (ding) that the AI noted. She is contesting the charge and denying responsibility. I have rented a few cars over the years, and it is always possible to incur such charges. Many times, the renter is time-obligated to leave the car and head to a flight, trusting that the vehicle will be checked in and the contract closed. 

I almost always do a walk-around video of a car before I leave the lot with it. I never drive off in a moving truck or rental trailer without doing so, because moving can so easily include scrapes, nicks, and dings. Despite my careful engagement, I was charged a pet cleaning fee once, when I never had an animal in the car. I now do a walk-around and interior video when I depart and return.

This illustrates, again, my contention that technology generally and AI specifically may be in an Arms Race (May 2024). The ascent of the AI car scanner shows the potential for more rapid vehicle return and more thorough examination then for potential damage. This may empower the rental agency to both deliver service and avoid inadvertently absorbing the cost of vehicle repair. And to do so without incurring the labor cost associated with a detailed inspection.

Nonetheless, there may be disputes with customers, as illustrated by the news story above. Any customer might take exception to the AI conclusions and fight the rental agency's demand for damage reimbursement. That may be a challenging undertaking in the face of evidence produced by the AI scanner, essentially a full PET scan of the departing vehicle.

Enter the consumer response, reported by the New York Post. There are now phone applications that similarly allow the renter to "create their own ... tamper-proof, AI-powered before-and-after damage scan in seconds.” And, in the realm of evidence authentication, the app "not only identifies scratches and dents but also timestamps, geotags, and securely stores the images to prevent alteration." That is a step above my habit of making videos. 

In this, some will see the resolution of a conflict between the car owner and the car renter. Others might see collaboration in that each side merely seeks to accurately allocate responsibility for damage or loss to the person appropriately at fault or responsible.

But, in the broader perspectives of AI and work, the main point that struck me was the loss of jobs. In all of my encounters renting cars, I have dealt with a human who checked me out of the rental facility and another who checked me in. That process is both labor-intensive and time-consuming. A human on either rental or return might merely assure you have taken/returned the correct vehicle, or they might do a detailed inspection. I have not had many rental or return experiences that were detailed. 

In such a human interaction, there is room for human frailty and error, perhaps worse. A security guard has been arrested in Miami for allegedly turning the other cheek or facilitating the theft of more than a dozen rental cars, according to WPLG. The rental agency targeted was also apparently Hertz. Humans can fail at an assigned task, and so there is some inclination to technology. Yet, technology can also fail. 

My first experience with eliminating some of the human interaction in car rental was an app. About three years ago, a rental company encouraged me to download their app; with it, I can often arrive, identify my vehicle, gain access, and exit the garage all without speaking to anyone. That is admittedly convenient. It has worked swimmingly at some locations and at others, not so much. 

Thus, there is no ideal solution to all challenges, but persistence that may require dynamic and periodic assessment and management. Staff may require supervision to do the job (inspect the vehicle studiously) or to prevent malfeasance (stealing cars or helping those who do). Management may have to balance the challenges of technology and the demands of keeping the service or product flowing. 

Nonetheless, these AI scanners could be a tool for both the rental and return processes. There is the potential for more rapid completion of both, without the attendant expense of labor. The technology may persistently improve with both the business and consumer tools innovating, upgrading, and competing. This illustrates once again the potential for AI to impact employment.

So, it may be that the scanner technology is not yet universal, or that the reliance on it is less at some vendors than others. Some may make consumption decisions based on their perceptions of this or that vendor's deployment and reliance. If a customer does not like these scanners, they may gravitate to rental agencies that do not use them. 

Some vendors may make similar decisions regarding who they will or will not accept as customers. If a customer is engaged in damage disputes, an agency might decide not to rent to them in the future. 

Customer and vendor, each with an AI tool, each striving to preserve or project. Each tool strives to accomplish a primary goal but is also persistently revised and improved to answer the demands and shortcomings of the other tools. 

The implications are broad and deep. As AI enters the workplace in the role of assessing and evaluating work quality, productivity, and efficiency, might workers similarly respond with their own AI that affords such evaluative assessment at the point of production? In the realm of labor and management relations, it is likely that each will engage and employ such tools in a check-and-balance not dissimilar from the car rental scanner scenario.

Carrying that similarity on, some employers may be less inclined to recruit or retain those who would resist such tools. Some humans may similarly avoid working for companies that deploy such tools. And in any event, it is practical to anticipate that deployment, response, and equilibrium will be dynamic and unpredictable for years as this AI arms race continues to evolve. 

AAOS 2025

Sometime around 2015, I became acquainted with the American Association of Orthopedic Surgeons (AAOS). The introduction was an invitation to participate in their Workers’ Compensation and Musculoskeletal Injuries Course. Since then, I have been fortunate to travel to some great locations including Atlanta, Chicago, Las Vegas, and San Antonio to deliver some thoughts, usually in the vein of credibility. In fairness, some destinations have been more engaging than others. 

This year, I will participate again in the 27th Annual. Thus, despite a number of years invested, this program has been ongoing far longer than I have been involved. 27 years is a long time, taking us back to the last century.

The program has a simple (perhaps deceptively so) purpose: the education of "Orthopaedic surgeons and other physicians who:"

"Treat workplace injuries or evaluate medical liability cases

Perform IMEs or wish to cultivate their treatment approach for occupational injuries."

The diagnosis and treatment of injury and disease is a challenging part of medicine. I have learned much about that by listening to various physician speakers over the years. The physicians who drive and teach this program are incredibly talented, focused, dedicated, and committed to both the art of medicine and the science of diagnosis and treatment.

The last time I travelled to participate was to San Antonio in 2022. The setting was spectacular, although that city's fastidious tendencies proved distracting. They clean the streets each morning with gasoline-powered leaf blowers. That noise at 04:00 can be disturbing to many who find the dark hours each day an opportune time for sleep and rest. Not me, mind you, but some people. 

Then the programming returned to "virtual" in 2023, where it will remain in 2025, for the third year. Much can be learned in virtual settings, but I lament the lack of personal contact and interaction. In various instances, I have seen outstanding speakers withdraw from participation in virtual opportunities. They do not see the value, do not appreciate the virtuality, and struggle with the tech. I get it.

To produce a virtual program, the Workers’ Compensation and Musculoskeletal Injuries Course necessitates pre-planning and preparation. The speakers have to produce and submit an outline months in advance. The PowerPoint and other materials had to be completed in July, and I recorded my video in August. The program is not until November.

Despite that long lead time and the feelings many have about virtuality, this program draws an incredible faculty. Because of that faculty, this program draws an outstanding, dedicated, and engaged audience. I have never been disappointed. 

Recently, I attended the "rehearsal" for the program. The faces were familiar, and one commented that rehearsal was not really necessary because everyone present had presented in this format and on their topics for so many years. That is incredibly notable; these folks have been dedicated to this program for many years, some for decades, and a fair number of them have been involved for all 27 years. That alone is incredible.

Another point that is incredible is the dedication of attendees. I have delivered a few lectures over the years and travelled to some wonderful destinations. It is very common to find seminar attendees in the halls, on the town, or otherwise distracted. But the striking observation of my AAOS trips has always been the persistent attendance. The physicians who attend this course are in the room, in the conversation, and in the groove from start to finish. They make no excuses; they commit to the course. That is inspiring.

As I am honored to participate in this program, I think we could all reflect on the inspirations. We can develop and dedicate ourselves professionally. The faculty listed below are testament to that. We can all focus and attend, grab the knowledge and perspective, and be in the moment. When we take the time from family and day-to-day work to learn, we need to focus, remain in the moment, and really learn.

It is amazing to work with and be surrounded by dedicated, studious, and engaged experts. I am looking forward to yet another opportunity in November this year. I am grateful to be invited and proud to participate in the laudable purpose of raising consciousness, adhering to science, and delivering value in the challenging world of occupational injury, treatment, and impairment determination.

The Workers’ Compensation course faculty includes:

J. Mark Melhorn, MD, FAAOS

The Hand Center at Mid-America Orthopedics, University of Kansas School of Medicine

Wichita – Department of Orthopaedics

Wichita, KS

Robert J. Barth, PhD

Barth Neuroscience, PC

Chattanooga, TN

R. David Bauer, MD, FAAOS

Orthopedic Independent Medical Examinations

Garland, TX

Charles N. Brooks, MD

Orthopedic Evaluations Northwest

Bellevue, WA

Eugene Carragee, MD, FAAOS

Standford University School of Medicine

Redwood City, CA

Marjorie Eskay-Auerbach, MD, FAAOS

SpineCare and Forensic Medicine, PLLC

Tucson, AZ

Barry Gelinas, DC, MD

International Academy of Independent Medical Evaluators

Dundee Township, IL

Kurt T. Hegmann, MD, MPH

University of Utah, RMCOEH

Salt Lake City, UT

Emily E. Heid, MD, FAAOS

IME Montana, PLLC

Missoula, MT

Mark H. Hyman, MD, FACP, FIAIME

University of California, Los Angeles

Los Angeles, CA

Doug Martin, MD

Center ofr Neurosciences, Orthopaedics, & Spine

Dakota Dunes, SD

David C. Ring, MD, PhD, FAAOS

Dell Medical School – The University of Texas at Austin

Austin, TX

Mark Ross, MD

Riverside Health

Gloucester, VA

Richard, E. Strain, Jr, MD, FAAOS

Orthopaedic Associates of South Broward

Hollywood, FL

Matthew Winterton, MD

Midwest Hand to Shoulder Surgery

Chicago, IL

Marilyn L. Yodlowski, MD, PhD, FAAOS

Lake Oswego, OR

The Expert Witness Course that precedes it each year includes:

George B. Holmes, MD

Expert Witness Course

Rush University Medical Center

Chicago, IL

J. Mark Melhorn, MD, FAAOS

The Hand Center at Mid-America Orthopedics, University of Kansas School of Medicine

Wichita – Department of Orthopaedics

Wichita, KS

Margaret P. Battersby Black, Esq.

Levin & Perconti

Chicago, IL

Steven A. Kodros, MD, FAAOS

Endeavor Health Orthopaedic Institute

Chicago, IL

Michael L. Vittori, Esq.

Quintairos, Prieto, Wood & Boyer P.A.

Chicago, IL

Opportunity Cost

In 1978, Brother Bluto delivered an epiphany in Animal House (Universal Pictures):

"Seven years of college down the drain."

The line was sarcastic and irreverent. In those days, no one needed seven years to complete an undergraduate degree (but some did move that slowly). That made the joke funny, in a backhanded and enjoyable manner.

Nonetheless, many spend that long and more on their education. We see many individuals pursue degrees beyond the four-year bachelor's programs. A relevant analogy is lawyers, who spend three years in law school supplementing their college studies. Thus, seven years on the nose. Medical doctors typically spend eight. A PhD might take three to five years following undergraduate school, seven to nine overall. 

An interesting aside, there are those who pursue multiple such degrees. People like Dr. Marjorie Eskay-Auerbach have always amazed me; she is a medical doctor and lawyer. I run into law professors who have both Juris Doctor and PhD degrees (at least 10 years of post-high school). Simply astounding from the perspective of invested time, intellect, dedication, and more.

These various degrees are investments. They are not cheap. Some contend a law degree will cost you $90,000 to $165,000. Medical school costs more like $238,420. The PhD is likely a similar investment. Most of that, for many students, is invested debt (they borrow money at significant rates and spend it on living and studying).

That money is a "sunk cost" and remains a resource drain in any event.

If a student has $90,000 in hand, that can safely yield $3,600 a year in interest. The $238,420 would similarly yield $9,536 a year. Thus, investing in education needs to yield a similar return. But there is additional cost to consider. The "opportunity cost." In other words, the doctor should make at least $9,536 per year more with that degree than in any vocation they could work in without it. 

That is reasonably easy in the medical world. Demand for physicians is significant. And yet, early career positions have poor financial returns. A medical resident might make as little as $52,331. That is a figure for Alabama, where the average starting salary for a college graduate is $56,748. The average college graduate makes $4,417 more than the average medical doctor resident who invested $238,420 for the privilege.

As we ponder why there is a limited population of physicians, we might consider that. A young person invests $238,420 to qualify for a job, making less than those who choose to enter the workforce with a college degree? And, this insult is greater if they borrowed that $238,420, because they are facing the expense of interest on top of the insult of the salary.

Those who are investing in college, professional school, or a PhD are simultaneously limiting their participation in the workforce. Though they may earn some during school, they are pouring time into studies and deferring some level of economic benefit because of the earnings lost during that time. 

That work over a college and law school career (7 years) or medical school (8 years) or PhD (9 years) will likely not be at a similar wage earned by those working full-time and on a career path. A high school graduate in Alabama might make $30,545. But a college student working part-time, 20 hours per week, will make less, perhaps $15,000. The college student is investing in a future, and deferring over $15,000 per year in earnings ($60,000 over four years). 

Those four years of part-time, low-wage work are a significant loss. The 7, 8, or 9 years are merely greater examples. The entry-level work positions for new doctors, lawyers, and others may generate the same or less income than college graduates. Those people are investing when they pursue education. 

There is an alternative. The high school graduates who choose a trade instead have a short-term potential to reach a living wage, $50,000 to $69,000 is cited as a reasonable expectation. This earning is more immediate and does not require the investment of years or dollars for tuition. Some will note that these high school graduates are starting at nearly the same level as the 8-year educated medical resident with $238,420 invested, perhaps in debt.

At the end of medical school, at 8 years, the new physician is about half a million dollars behind the person who elected a trade.

That economic reality is a force in decision-making. The numbers are stark and to some perhaps even shocking. And, they ignore that the medical student, the new medical resident, is facing the debt-service expense on the $238,420. The interest must be paid from the annual earnings mentioned above. With 8 years of trades experience, that worker is likely earning at least $60,000. And the comparison is valid for college graduates also.

After the minimum three-year residency for medical students, or three years of work as an associate attorney, the debt-burdened professional is still likely one-half million dollars behind the plumber, electrician, welder, or other tradesperson. Doctors reportedly do better after residency, earning an average of $250,000. 

Most young lawyers struggle to gain that income level three years out of law school. This illustrates that economic recovery is possible, but not necessarily equitable. The supply and demand (lots of lawyers, not so many doctors) impact that with inherency. 

By 15 years post-high school, the medical doctor has likely equalled the income total of the college graduate and the tradesperson. In a long career arc, the doctor is likely to do well financially. The work of a physician or attorney will involve stress, hours, and other challenges but will avoid some of the physicality (lifting, carrying, climbing). But this all merely illustrates that there are choices.

This series of thoughts was instigated by an Artificial Intelligence (AI) expert who, or is it "that" opined in August 2025 that "Students could end up 'throwing away' years of their lives, as technology is moving so quickly." He recommended against pursuing doctoral degrees as a path to "standing out" in the workplace and job market.

According to his analysis and other AI experts cited there, "undergraduate degrees have lost their payoffs thanks to AI." The next shoe to drop will be a similar devaluation of the benefits of "advanced schooling" for many. Despite the urge to blame AI, this same expert claims that "AI itself is going to be gone by the time you finish a PhD." No, AI will not be gone, but he means that the challenges of building, deploying, and adapting AI will be solved.

These experts are suggesting that there is a decreasing value represented by the investment in higher education. They are advocating for careful consideration of the cost-benefit analysis of education. Those who face a future career arc of 40 years, as they graduate from high school, should be cognizant, conscious, and deliberate about training, education, and debt.

The reader should know that Utopia is not Coming (October 2025). No one will ever take care of you like you would yourself. You owe it to yourself to be your own best advocate, advisor, and actuator. 

Are we on the brink of the end of days as the prognosticators suggest? There will not come a day when boomers accept the word of a robot regarding illness, diagnosis, or treatment. You read it here first. But that day may come for the tech-ready iPhone generations that followed. That said, it is very likely that our daily lives and repairs by electricians, plumbers, air conditioning techs, and similar trades will remain human tasks regardless of consumer generation or age.

AI is indeed changing the world. It will undoubtedly alter our perceptions, exertions, and economics. Now is not the time to lament or fear, but a conscious, careful, and objective review of realistic cost-benefit is warranted and recommended. What are the probabilities of the choices you will make about training, education, and borrowing? 

Make informed choices about that education investment. Carefully consider the opportunity costs and investment costs. You might be better off investing that $238,420 in the stock market than in an M.D., JD, or even that B.A.?

Utopia is not Coming

I have touched on Universal Income in the past. Many prognosticate that enforced socialism will be the inevitable American, if not world, eventuality. See Let Them Eat Brioche? (September 2018); Universal Income - A Reality Coming? (November 2016). The unifying theory of Universal Income is that somehow there will be wealth, and that the government will decide how and when to seize it and then redistribute it in some manner it deems fit.

The subject arose again recently with Fortune reporting on a "legendary investor" issuing "a stark warning regarding the future impact of artificial intelligence (AI) and humanoid robots." The outcome predicted is one of "dramatic increase in wealth inequality."

That outcome of inequality might be interpreted as requiring shifting human motivations and endeavors. Or, there is the alternative of centralized socialism.

The investor contends that a great many jobs will evaporate, and he "questioned the need for lawyers, accountants, and medical professionals." These occupations, he says, will be replaced by "highly intelligent robots with PhD-level knowledge." The thinking in society will be done by a “humanoid robot that is smarter than all of us and has a PhD in everything.”

Therefore, there will be some mechanism for both "redistribution of money" and how we might "put people to work." Today, putting people to work is largely a market-force proposition. Goods and services are demanded, and businesses strive for a method of being selected to fulfill that need for a particular individual.

Those who produce the best results, achieve the most perceived value price, or mount the most persuasive advertisement campaign win the contest, deliver some service or good, and are recompensed. As they say, "Et voilà!"

The consumer decides that this or that car company, law firm, hospital, or other provider prospers. We consume, and they succeed. These decisions are driven by individual needs, desires, proclivities, and even vanities. I drive a Hudson Hornet because "Now You're Face to Face with Tomorrow." That is why I picked it over the Studebaker across the street. Advertising draws, brand loyalty draws, looks, feels, and more sell.

We make other market choices that may or may not be "free" choices. Some of us decided in high school to pursue college; others went on to medical school, earned a pilot's license, learned to weld, or otherwise qualified for some vocation or profession. As a result of those distinguishing criteria, some of those were allowed to do tasks forbidden to others (I have had more people decline to let me do their surgery, and I cannot figure that out—I have seen it on the Discovery Channel enough).

In all this, there is an economic driver called scarcity. Decreased supply increases price. If it were legal for any person to do appendectomy surgery, the cost would go down dramatically. If anyone off the street could pilot a Boeing 777, then the salary for pilots would plummet. Scarcity rules the economic realm. If diamonds covered the beach like sand, then no one would pay thousands for a diamond ring. If the Gulfstream G650ER were not $150 million each, we would all own one (I would have two).

Scarcity governs economic decisions. The price tag that keeps me from the Gulfstream similarly keeps me from living on Key Largo, lunching on Beluga caviar, and so much more. Scarcity drives markets. Jessie J. insists "its' not about the money, money, money," but I fear it truly is, always has been, and always will be. (Price Tag, Universal Music, 2010). 

Taylor makes more because she is talented, produces desirable music, and puts on a compelling show. Any might do any one of these, but she is the complete example--scarce, and price increases. The same works for Beyonce, Shaquille Oneil, Tom Brady, and more. Will they bring their talent, expertise, and panache? Why will the Musks and Jobs continue to innovate, explicate, and develop? What will drive them to the next revolution? 

The pundits who posit that we will soon live in a society without economics never explain how, when, and by whom decisions will come regarding who gets to live on the water in Paradise and who has to live on the river in Cleveland. With Universal Basic Income, how will economic decisions be made? If we all have basic income, will stores stock products that are not basic? If they do, to what extent will there be a consumer base willing (or able) to reach for the Tillamook instead of the store brand?

If everyone in the proletariat gets the same UBI, what is the motivation for self-improvement? Who will study, stretch, achieve, and perform? Where are we left as a society if no one is reaching for the stars? There are those who believe that the destination is worthwhile, but a great many more see the value in the journey, the effort, and the opportunity. Can that human nature be rewritten?

In the midst of drafting this exploration, which barely scratches the surface of UBI and macroeconomics, Elon Musk iterates yet again that

"society won't just need a universal basic income — it will need something bigger: universal high income."

Mr. Musk says that we will all have way beyond "basic." We will all have the

"'Best' Medical Care, Food, Housing and 'Everything Else' And A Universal 'High' Income."

The siren song is attractive. I picture myself resting my feet in the ocean on my private beach while I reflect on whether to take my Bugatti or my Ferrari to the airport to grab my G-4 for the hop to Cyprus for lunch. And yet, I wonder if there might be others who would compete with me for the space and possessions I would pursue? In other words, might my neighbor steal my Bugatti while I am at lunch? Not if she has her own, but might I then take hers because it is newer than mine?

At the end of the day, I am certain that Mr. Musk and a great many others are far more intelligent than I am. I am confident that there is much about our world that I do not comprehend, and a good bit of it I would not understand on your fifth or sixth attempt to teach me.

That said, I struggle to see a world of limited resources in which there will not be competition. I see human competition persisting on a personal, collective, and even national level. I see it on a rational and even emotional basis. People will compete for resources. Wars will be fought over resource access and needs, from survival to luxury and everything in between.

I predict that whoever/however resources are meted and distributed will not be fair, equitable, or impartial. Some undeserving will thrive, and some exemplary will stagnate and suffer. In that, the future will likely be significantly similar to today. There is, without doubt, a great deal of inequity and inefficiency in the distribution of resources in our present. On what basis would one expect that the future will magically produce a perfect, or even more efficient and effective, distribution than today?

Socialism has failed before. This has happened historically because of simple truths. Economist Mark Perry noted, years ago, that "Socialism does not work because it is not consistent with fundamental principles of human behavior." That will remain. Central planners produce outcomes, people will perceive inequity, and human nature will rail against the mindset of utopianistic central control. This will be true whether we have "Baxic Income" (c), 2025, or Mr. Musks's enhanced "Bext Income" (c) 2025.

As we strive today, grasping and adapting to the world that AI beckons us toward, know that there are some inalienable truths. Among these are human nature, competition, and independence/freedom. There may be economic shifts, false starts, and untoward outcomes coming in our future. 

But, there is no utopian Brave New World coming to theaters near you by 1984. There will be no "Inner Party, the administrative Outer Party, or Proles" any more than humans will be categorized into "Alphas, Betas, Gammas, Deltas, and Epsilons." Humans will remain inspired, competitive, and self-interested. 

Yes, the future is coming. Yes, AI will change the world as we know it. No, it is not going to be some socialist utopia. There may be some bumps and miscues. There always have been. There will be winners, and there will be the rest of us. But there will be opportunity, progress, and more. 

Appearances

Years ago, I sat with the family at a turn-lane stoplight in Jacksonville. I perceived smoke or steam behind me and concluded aloud, "I think that guy just hit us." The others decried my conclusion with statements like "I didn't feel anything." I exited the car and found a Cadillac impaled on the trailer hitch of my SUV. It was a mess.

The Cadillac driver was likely at least an octogenarian. He apologized profusely and pulled a wad of Benjamins from his pocket. He was eager to avoid an accident report, saying that his son had warned him of lost driving privileges upon "one more accident." We were unhurt; the poor Cadillac had not even scratched my hitch's paint. I empathized with the kindly man. I declined his money and wished him a good day.

How often does that play out across the country? I suspect it is frequent. Perhaps more so in the promised land of retirees. 

Recently, it reportedly played out similarly, and coincidentally in Jacksonville. Maybe this is a Jacksonville thing? Doubtful.

A 48-year-old backing his Mercedes from a space reportedly struck a Ford F-150 on the morning of October 7, 2025. But then there is the Paul Harvey (i.e., "the rest of the story"; see Make Your Point (July 2025)). This parking mishap occurred at 03:00. It allegedly occurred in the parking lot of a business called "Wacko's Gentlemen's Club." And there was some suspicion that the Mercedes driver may have been drinking.

The Daily Mail reports that the driver, a fellow named "Guy," exited his car and briefly spoke with the truck driver before attempting to leave the scene." He was detained by a parking valet and others, and the police were summoned. Perhaps such events are common, but this instance made the news, appearing on multiple platforms.

This may be one of the few times one might utter, "Some Guy hit me," with full accuracy. But I digress. 

The Brunswick News reports that he did not leave the scene because he "was unable to do so as he appeared highly intoxicated." That report alleges the Mercedes driver "refused to exchange information and instead offered him (the truck driver) $500 to settle the damages." The power of Benjamins is perhaps expected to entice?

The Mercedes driver was allegedly "Superior Court Judge Robert Guy Jr. " of Georgia. He has served on the bench for about nine years (2016). At the time of the trip to Jacksonville, he currently served as "president of Georgia's Council of Superior Court Judges."

And, unlike the kindly gentleman whose Cadillac was once impaled on my hitch, Judge Guy is neither an octogenarian nor was initially perhaps perceived as particularly apologetic about the alleged collision. Rhiana is playing in my head for some reason (Take a Bow, 2008, Def Jam).

The judge reportedly self-reported the incident to the "Judicial Qualifications Commission" immediately. The New York Post soon reported on October 12, 2025 (collision plus 5 days), that Judge Guy has "resigned as president of Georgia’s Council of Superior Court Judges." 

Those were both appropriate reactions to the allegations. There is merit in preventing, to the extent possible, having personal allegations implicate such a council, and it is always best to self-report potential discipline issues. 

On October 16, 2025 (Collision plus 9 days), the Atlanta Constitution reported that Judge Guy had resigned from the bench. The resignation mentions the "honor and privilege" of serving on the bench and includes praise for judges, lawyers, and court staff. In all, a gracious departure. Some might see fault in the delay in resigning; others might say that it is reasonably rapid. Perspectives. 

There are moments that draw us all in. They may be celebratory or less. I have found that people visit bars, imbibe, and socialize from a variety of motivations. Nonetheless, Mom's old saw that drinking should never precede driving resonates. Driving a vehicle is a decision.

 

Sobriety has its place in this world. I have written about that. See Sober as a Judge (December 2019), A New Miranda Warning (April 2020), and A Judge Under Surveillance (June 2025). Sobriety would have likely benefited the protagonists in each of those instances. Drinking does not mix with driving. 

My favorite judge post involves the Conferences and Consequences (November 2019) story of strip clubs, heavy drinking, White Castle burgers, and sign language communication. One general rule I recommend to young and old is that nothing good happens after midnight (unless you count the publication times of these gripping blog posts). 

The judges in Conferences and Consequences ended up involved in violence, embarrassment, and more. I reflect this morning and wonder how those judges view their experience in May 2019. Do you suppose the ones that were shot view it differently than others? I have never been shot, but I suspect it is memorable. I hope I never find out. 

Yes, there are a few out there who have made grievous errors in judgment. More recently, we have seen it in Georgia. See She Must Go (May 2024) and Adjectives and Appearances (June 2024). We have seen it elsewhere also, including Florida. See A Judge Under Surveillance (June 2025). 

There are easy lessons in life, and then some not so easy. If you find yourself driving 75 miles to a strip club, it might be worth reconsidering. If you are in any club at 03:00, same. If you strike another's auto and are inclined to pull out the Benjamins, same. There are signs of impending difficulty. Driving to drink will mandate driving back after having done so. 

There will be those who will criticize this post. I will hear "preachy" and "unempathetic." I am neither. But I do believe that we cannot live long enough to make all the big mistakes ourselves (Eleanor Roosevelt). We owe it to ourselves to think on the misfortune and decisions of judges cited in these various posts. We might consider our course a bit harder before heading out for burgers or otherwise in the middle of the night?

Kudos to Judge Guy for truncating the need for investigation by the qualifications commission. Kudos for removing the potential for negative appearances of the bench by resigning the job. Kudos for taking responsibility and moving on with the personal challenges that are inevitable in any prosecution. His days will likely be difficult in the near term as he deals with these allegations. 

For the rest of us, judges and otherwise, consider the implications. Do so before you imbibe, instead of trusting yourself to make valid and safe judgments after you have had a few.  Think, then act, and avoid such allegations and consequences if possible. Hint: it is usually possible. 

A Tool Kit

On a recent conference call, an attendee spontaneously shared praise for the efforts of the Texas bar regarding Artificial Intelligence (AI). It is fair to say that AI has been criticized and feted. AI is either the best innovation since the wheel, or it is the harbinger of "the end of days." Perhaps, it is actually something short of either of these two extremes.

I have noticed that the older folks are struggling more with AI than the younger ones. That is subject to exceptions. One of the biggest proponents and fans I know is a septuagenarian, and one of the critics I ran into is a Gen Z. There are exceptions to most rules. Nonetheless, we of the "OFC" (old folks club) are struggling with the acceptance and adaptation of this new tool.

So, it came as no surprise that the conference attendee who provided this information was a card-carrying member of the OFC. And, as a fellow member, I was drawn to that recommendation. I am sharing it with you. No, the Texas effort is not the Rosetta Stone or the Missing Link. If you seek a be-all, end-all panacea about AI, I wish you good luck and safe travels. Your quest is noble but misguided.

The punch line, essentially, is that there ain't one.

That said, the Texas efforts are worth your consideration. Titled the AI Tool Kit, there is a wealth of analysis, advice, and patience. There are helpful takeaways.

This reinforces critical points on ethical use of AI. It fails to credit me, but endorses my advice on Trust but Verify, stating:

"You must independently verify Gen AI-generated results—never rely blindly."

There are admonitions to "double-check" and to focus on the critical lawyer obligation of "candor to the" tribunal. Layered on top of that are admonitions on billing that are worthy of consideration.

The AI Tool Kit addresses a spectrum of issues with "Managing AI Risk." That is positive. Let's all get over the panacea fallacy. There is risk with everything you do. Over 200,000 injuries are reported each year occurring in people's bathrooms. Risk surrounds us. Don't deny it, manage it.

So, following that analogy, perhaps there is less concern with tooth brushing than with shaving (sharp object) or with stepping into or out of the shower (slippery surfaces). The AI Tool Kit strives to identify and discuss the "least risk" to the "high risk." A sound analysis. Don't forget you can still get hurt with the low risk. Toothbrush injury may be rare, but explore it and make intelligent prioritization decisions.

What of privacy? Security? Well, folks, those are issues for anyone who possesses data. If the information is on paper, it can still be misplaced, mistransmitted, or stolen. That the data is digital may change the "how" of precaution/protection, but not the "if." Well, using AI is an extension of that "how" analysis.

A critical portion of the AI Tool Kit advises on the client. In the end, the purpose of every lawyer is service. We make no widgets. The lawyer is a service provider, whether in analysis, drafting, remediating, or litigating. How can the lawyer deliver service, and what is implicated by AI in that process?

The bottom line is that the world is changing. AI is an example, but not the only one. It is changing method, pace, and more. We would be remiss not to note it. We should be curious and engaged about it (unless you are committed to retirement in the next 24 months, it will likely impact you). We should be as respectful of it as we would be of any efficiency tool (think how much more efficient a chainsaw is over a bow saw; it will cut you more rapidly and deeply).

Is there magic in the AI Tool Kit? No. There is no panacea I have yet found. But, there is wisdom there, and at least enough knowledge to make you begin to understand how and when to ask more questions. I recommend the Tool Kit as critical reading for all who are engaging AI. It is a worthy investment of 30 minutes.

Phone a Friend

There are frequently opportunities in life to learn from others. Eleanor Roosevelt once said, "Learn from the mistakes of others. You can't live long enough to make them all yourself." There is merit in that. Life is, indeed, fleeting.

I am often reminded of that admonition when judges face discipline. A recent instance involves a County Court Judge appointed in 2023, who will reportedly resign at the end of October, according to WCJB. A quick review of the Circuit website confirms extensive experience and notable educational and professional credentials.

In October 2024, the Florida Judicial Qualifications Commission filed a Notice of Formal Charges concluding probable cause existed that the judge failed to comply with Canons 2A, 3B(2), 3B(4), and 3B(8). Specifics of each were quoted in the Notice.

There were conclusions as to "due process, ... patience, dignity, and courtesy." The Commission also noted a failure to "act at all times in a manner that promotes public confidence in the integrity and impartiality of the judiciary."

It is fair to say that this instance illustrates some critical points. Nonetheless, the one most worthy of study is perhaps the necessity of community. The judge was called upon to preside in truancy proceedings, in proceedings with difficult parties, and in emotional situations. The Notice describes these.

The Judge filed a response and answer to the Notice. The response provides another perspective and explanation of the interactions and instances that were viewed by the Commission on courtroom recordings. The recordings, nonetheless, are compelling. The Judge did contest the validity of "the time stamp" on one recording, which might be inaccurate, and noted in another instance that the entire tenor of proceedings in the courtroom was "not necessarily fully reflected on the videotape." Further, that "the video is not the best evidence of fully showing the behavior of" a party in the courtroom.

This is respectful of both, with various iterations to the effect of:

"Admit content of verbatim communications at the hearing but deny argumentative wording/conclusions set forth in the paragraph."

Thus, on various points, there is not disagreement as to what was said, but the assertion that different perspectives might coexist regarding the words, tone, and more. That is an important consideration for all involved in litigation. Tone and volume may be interpreted differently by others.

There is an allegation in the response that the Judge suffered physically on one day. There is illness described, and it is axiomatic that illness, stress, and other influences can change how we act and react. In retrospect, the judge noted that because of illness, "she should have continued the hearing." She noted that she proceeded because "there was no other judge at the courthouse to cover it."

There is mention of tone and temperament. One instance involved a party who elected to converse with another attendee rather than devote his attention to the judge. Later, that party and another party became involved in a conversation that was not directed to the bench, and the judge was challenged to interrupt. There was no gavel at hand, and, without yelling, the judge found it "necessary as a practical matter to speak over them" to regain control of the hearing room.

That is always a concern. People wait and plan for hearings. Those involved in litigation often have high expectations for their day in court. Having a delay or the surprise of a substitute judge can be disconcerting and disappointing.

In various instances, the judge's response notes that "as a new judge, she had not had the benefit of Judicial College." There was "no formal training ... provided ... prior to taking the bench." There are multiple descriptions of the Judge preparing for new and unfamiliar responsibilities or duties by "read(ing) case law," "reviewing the rules ... and ... statutes," and consulting with colleagues.

On the "prior to the bench," I will never forget sitting down with the chief judge when I was first appointed. Her advice resonates to this day: "Welcome, and good luck." Not so helpful. In speaking with many older judges, I found her aid and succor were similar to each of us in this regard. 

The judge in this complaint noted a pertinent point that should weigh on the mind of every new judge:

"for ... years ... before taking the bench (the judge) was an advocate, not a judge and ... there was a transition period ... adjusting to that new role."

There is great merit in education. Training is critical for a great many occupations and vocations. The job of adjudicating is no different. Nonetheless, the demands of the role are immediate, while the opportunities, such as the Judicial College, may require time.

Ultimately, the judge acknowledged that the judge "made mistakes at times." These included misunderstandings of the law and misstatements.

The lessons here are worth discussing. Every new judge will be inexperienced. That said, maybe there is some experience, education, or foundation that better prepares one of us than another. But at some point, we are all new. We may come from litigation or not, may have seen great judges in action or not, may have had an opportunity to study, observe, and emulate, or not.

Even if you had the chance to observe the best, speak to the best, and emulate the best, know that the view from that side of the bench is not the same. That is a ludicrous contention, but it is true.

The practice of law differs from its adjudication. Running a proceeding is challenging. Parties and witnesses can be challenging; those who lack an attorney particularly require a slower pace and greater patience. There are a variety of potential twists and turns in the law. And, there is the persistent pressure to get it right. A judge bears a huge responsibility to deliver due process, to remain attentive, to preserve patience, and to deserve respect.

Eons ago, in St. Elmo's Fire (1985 Columbia Pictures), a complex character utters a defining line, "Me? Oh, you know, it ain't easy being me." I suggest that this is true for anyone who would undertake being a judge. Having a gavel may be an aid (get attention and quiet a room). Wearing your court outfit may be an aid. Reading, writing, and studying may be an aid. But, nonetheless, "it ain't easy," ever.

What avenue did the judge in this complaint have that was not effectively engaged? As I read the Judge's Answer, I could think only of Who Wants to Be a Millionaire (1999 Valleycrest). That show put people on the spot (as parties do with judges). The person on the spot had to analyze, differentiate, and decide. As the questions became more complex, the contestants were offered help.

The contestant could ask the producers to reduce the potential answers from four to two, called a "50:50." There is no producer in court, but if a judge can eliminate some potentials from a spectrum, the decision may be easier (this is perhaps of use in issues such as whether it is or is not hearsay, or whether an exception to hearsay does or does not apply). Parties can bring very complex issues, and finding a path to make sequential, smaller decisions in a process like hearsay may be more manageable.

The contestant on the show might also ask to "poll the audience." That one will not help in the context of the judge. The questions are all a matter of judicial decision. But the judge can always take a break. That affords time out of the public eye to look at (poll) the rules, the statutes, and the prior case decisions. This is not "polling" in the classic sense, but it both allows reflection and may minimize errors. It affords a look beyond personal knowledge, as a poll might.

Finally, the contestant on the show might "phone a friend." This was done live on the show, but in our context, it could be done in the same recess already suggested above. There is no better tool than "phone a friend." They are unlikely to give you the answer. But a call to a mentor can help with explaining your own perceptions, gathering other perceptions, and moving forward purposefully.

On the show, each contestant only got one of each of these opportunities (on the show, they were called "lifelines"). They judge, however, may take as many as they like. Ponderous is not suggested, but slow the roll as you need to in order to be effective and deliver the service the people deserve. 

Over time, confidence and comfort grow, and you can evolve from needing the lifeline to being the lifeline. Know that you will never know everything. This role of adjudicator is persistently novel, challenging, and frankly amazing. 

As an aside, judges should rely only on other judges in the advice process. See Canon 3B(7)(c). That said, these three lifeline suggestions might work as well for the attorney who is struggling, and they might benefit as readily. Their call to a friend should likely be to a lawyer mentor, to the same end of aiding critical thinking and analysis.

What can a judge who lacks formal training, a judicial college, and more do? These three pop-culture suggestions are worthy. I have trained many judges. The most important advice ever has been to address uncertainty by "calling a judge colleague you think will disagree with your first impression." Get the other perspective. It may lead you away from error or may commit you to your original course. But, either way, it will be an intellectual and helpful exercise for your thoughts and mind. 

The judge in this story has announced a resignation. It will therefore never be known how the Florida Court might have ultimately perceived the circumstances or conclusions. Nonetheless, there are lessons here worth consideration and contemplation. 

Simple Steps

I have been critical of lawyers and judges who cite hallucinations in their writing. I used to say "hallucinated cases." Attorney Middlemier* admonished me lately for referring to "hallucinated cases" and "hallucinated authority" from time to time. The point is somewhat obvious: such references are neither "cases" nor "authority," they are merely hallucinations. Let's all get that part straight going forward. Some might argue they are "fraud," but that is a bit judgmental. 

Attorney Middlemier also criticized me for my statements that every lawyer now knows better and that avoidance of hallucinations is easy. Middlemier's contention here is that no one has ever provided simple, easy-to-follow advice on avoiding hallucinations. I stand contrite and throw myself on the mercy of the reader. Apologies (see below for a more thorough apology).

Lawyers, judges, "lend me your ears." I come to save your name and reputation, not to bury you. (adapted from Julius Caesar, Billy Shakespeare, 1599). 

I strive today, instead, to provide a working suggestion for the legal professional on avoiding the citation of hallucinations in legal writing. It is offered with the best of intentions and hopes. The reader will perhaps forgive some reference to ancient pop culture, but there is sound advice on this point from A Fish Called Wanda (1988 Prominent Features), more on that follows.

How to Avoid Hallucination in Legal Writing:

Step One - do not use artificial intelligence (AI) large language models (LLM) such as ChatGPT, ClaudeAI, Gemini, and similar for legal research or writing. These are not search engines that scrub the World Wide Web for data, information, and knowledge. They are assimilators that strive to assemble information in a way that is pleasing. Some compare them to golden retrievers who wish to please you. If they cannot find the stick you seek, they will bring you something else instead. Unlike the gentle and kind golden, they will lie to you about the something else and try to convince you it is indeed a stick. 

LLMs are not built to do research. And, they are offered free of charge. Hint to the masses: if you are not paying for a service, you are not the customer; you are the product. The LLM is trading your information for its own. It is learning from you as you try to learn from it. Caution to lawyers: anything you put into an LLM, particularly a free one, becomes usable by that LLM. Your client's confidences and privacy may be at risk from such use. You could be liable for that in legal settings or Bar disciplinary proceedings. 

If you use an LLM and it produces a polished, convincing argument with multiple citations, and the result is the absolute best you could hope for in terms of simplicity and completeness ... this is a red flag. Be wary of golden retrievers bearing gifts. 

Step Two - (for when step one does not work and you use an LLM instead of doing lawyer/judge work), verify the results. As Wanda (Jaime Lee Curtis) did in the epic film in her name, verify:

"Now let me correct you on a couple of things, OK? Aristotle was not Belgian. The central message of Buddhism is not 'Every man for himself.' And the London Underground is not a political movement. Those are all mistakes, Otto. I looked them up."

The point here is that Otto (Kevin Klein) hallucinated (made things up). Wanda is not arguing, implicating, or officiating. She is simply doing the legwork. She found the fallacies in a straightforward and simple manner. She "looked them up." Ingenious! (She is the protagonist after all). "Look them up," "look them up," "look them up." Step Two is simply "look them up."

How does one effectively look them up? Know that even paid services like Westlaw and Lexis AI can and do hallucinate. See the Stanford Study, May 2024. 

So, for statutes, rules, or regulations, it is easiest to use a search engine (Google, Edge, Safari, etc., also called a "browser") and search for the statute and rule citations you have selected. This should lead you to government websites that publish statutes and rules. Strive in Florida to find results on the sites of the Legislature (https://www.leg.state.fl.us/) or the Senate (https://www.flsenate.gov/). If a statute exists, your search by number (e.g., for 440.25) should yield a result on both of these sites. If you do not find it on these sites, that is a red flag.

Use a trusted subscription service for your case law citations. In Lexis, Westlaw, Vlex, or a similar database, search for the case by name (use the non-AI tool on those services; see Stanford Study above). If that search yields a positive result, verify your citation matches the one in the database. If it does not yield a positive response, then search the citation you have. 

Negative results from either such attempt should be a red flag (you should likely delete that citation as untrustworthy and return to the drawing board). If you unequivocally love the case you are searching to authenticate, perhaps keep striving. But it is highly recommended that you abandon the case or authority if you cannot find it on these subscription services. 

Having failed to verify this way, you could turn next to the internet. If you turn to the internet, use the same search engines (browsers) to search by case name and citation. It is possible you may yet find an actual copy of an opinion. However, there will remain doubt due to its absence from the commercial databases (Lexis, Vlex, Westlaw). That does not mean the authority you are striving to verify cannot be real, but you should be harboring suspicions and reservations by that point in your process.

For example, if you have cited Gusmorino v. T.G.I. Friday’s, 928 So. 2d 446 (Fla. 1st DCA 2006) or Specialty Emp. Leasing v. Davis, 596 So. 2d 695, 696 (Fla. 1st DCA 1992), paste those names into a browser and evaluate the results. Be careful - some browsers provide an AI LLM response; e.g., Chrome may provide a Gemini analysis that is as much hallucination as the original LLM provided creation.

You may see search results for your case in databases like the University of Florida Repository (https://scholarship.law.ufl.edu), the Florida State University collection (https://ir.law.fsu.edu), University of Miami (https://repository.law.miami.edu), The Myfloridacfo site, and other public sites like Vlex, Justia, Digital Commons, Findlaw, and more. If your case is not generating such confirmatory results, this should be a red flag. I implore you at this point to abandon that LLM-generated authority, no matter how much you may love it. 

Step Three - this one is seemingly the toughest part. Having verified the existence of the authority, now you must actually read it. If you find that Fla. Stat. §440.38(7) and Gusmorino exist (it doesn't), that does not conclude the analysis. As a lawyer, you are seeking relief, making an argument. You are asserting, when you cite authority, that the authority stands for something in particular. You cannot know that unless you read it. 

Not to put too fine a point on things, the lawyer or judge receiving a pleading, or hearing an argument, with such citations should also engage in both Step Two and Step Three. Trust, but verify is the critical point. 

This is a good moment for the lawyer to pause and review Rules Regulating The Florida Bar, Rule 4-3.3. This is about honesty-in-fact. Lawyers are supposed to tell the truth when speaking or writing to a tribunal. When you cite or argue hallucination, that is not the truth; it is fantasy. With all you have invested in becoming a lawyer, is that job, reputation, license, and more worth keeping? The lawyer may certainly decide, but "choose wisely" (Indiana Jones and the Last Crusade, 1989). 

As you read the authority, the analysis is whether it says what you think it says. For a comedic example, refer to Inigo Montoya's reference to "inconceivable." Princess Bride (1987)("you keep using that word. I do not think it means what you think it means"). You have to read the case, statute, or rule. You have to decide what it means. You have to articulate its meaning, holding, relevance, and applicability.  

You cannot merely rely blindly on what your clerk, paralegal, partner, associate, priest, doctor, or spouse says. Yes, those are all trustworthy people. Yes, if they tell you the best pie in town is at Pizza Planet, try one (you won't lose your license for trying a pizza). No, Virginia, lawyers cannot rely blindly on what those people, or any people, say about legal authority. Lawyers cannot blatantly make up authority or facts (that is lying) and cannot rely on what others make up either. 

Lawyers have to do the work, read the authority, make judgments, and arrive at their own conclusions. It is singularly the role and responsibility of the lawyer signing the document or uttering the argument. If a computer could do it, then we would have no need for lawyers, law schools, professionalism, and so much more. But no computer can do it, so you must. 

Hint: In case you missed it, lawyers have to analyze and make judgments. They cannot blindly rely on their most trusted human compatriots. Similarly, you cannot blindly rely on a computer program (AI LLM). When the lawyer signs their name, the tribunal will rely. We count on you to be honest, accurate, and forthcoming. Reread Rule 4-3.3. 

To simplify and reiterate:

1. Do not use LLMs for legal research and drafting (use them to check grammar, structure, spelling, etc.). 
2. If you do use them to research and draft, verify every authority with a subscription, search engine, or both (do this if your associate, paralegal, or partner is used to research and draft). 
3. Read every authority you cite, regardless of whether you found it, your paralegal found it, or some LLM found it. 
4. Remember that it is your name, your reputation, and your license on the line if you are less than complete and honest in your statements. 

Epilogue

An apology was mentioned briefly above. Perhaps a longer apology like that offered by John Cleese (playing a lawyer) in A Fish Called Wanda (1988 Prominent Features)(perhaps one of the most outlandish, ridiculous, and humorous films of all time):

“I offer a complete and utter retraction. The imputation was totally without basis in fact, and was in no way fair comment, and was motivated purely by malice, and I deeply regret any distress that my comments may have caused you, or your family, and I hereby undertake not to repeat any such slander at any time in the future.”

As a matter of full disclosure, I have not used any LLM in the production of any of my blog posts. All of the authorities and references cited herein have been personally reviewed by the author. Do we need or want some rule requiring a disclosure like that on every filing? Likely,  Rule 4-3.3 is sufficient. 

*Horace Middlemier is not a real person. He is a figment of the author's imagination, a foil, an "everyman," used purely as an illustration and to protect the identity of many who provide fodder for these pages. Any similarity to any real person, living or dead, is pure coincidence and not intended.