Judges Admit AI Mistakes & NBA Gambling Scandal

0:00So there's a lot of noise about AI, but time's too tight for more promises. So let's talk about results. At IBM, we work with our employees to integrate technology right into the systems they need. Now, a global workforce of 300 ,000 can use AI to fill their HR questions, resolving 94 % of common questions. Not noise. Proof of how we can help companies get smarter by putting AI where it actually pays off, deep in the work that moves the business. Let's create smarter business. When you own your own business, you own every decision. Now own the card that rewards you for it. Chase Sapphire Reserve for Business is a pay-in-full card that elevates your travel experience and offers premium benefits that will take your business to the next level.

0:45Sapphire Reserve for Business offers 8x points on all purchases through Chase Travel, 3x points on social media and search engine advertising, airport lounge access, and more. Chase Sapphire Reserve for Business. It's the card that gives back all you put in. Learn more at chase.com forward slash reserve business. Chase for Business. Make more of what's yours. Accounts subject to credit approval. Restrictions and limitations apply. Cards are issued by JPMorgan Chase Bank N.A., member FDIC. When you're running a business, the best days are the ones where priorities stay on track. For midsize and large companies, risk can affect multiple parts of the organization at once, from property and liability to cyber and regulatory challenges.

1:26At that level, managing risk becomes an ongoing discipline. At the Hartford, the focus is on helping businesses manage risk before it turns into something more disruptive. And when losses do happen, that work is paired with insurance coverage shaped by years of underwriting, risk engineering, and claims experience. Learn more at thehartford.com slash risk mitigation. Policies provided by Hartford Fire Insurance Company and its property and casualty affiliates, Hartford, Connecticut. This is Bloomberg Law with June Grosso from Bloomberg Radio. A lot of work goes into legal research and writing, so using AI seems like a way to replicate some of the very time-consuming tasks such as researching court cases and citing them in legal briefs.

2:16And the legal profession has seen a rise in the use of AI in recent years, with lawyers facing increasing scrutiny and sometimes fines from judges for the alleged misuse of AI, as fake case citations and other errors show up in legal filings. You may ask, how can lawyers fail to vet the output of AI-generated legal documents. And now another question, how can judges fail to vet AI-generated orders before they go out from their chambers? In a response to an inquiry from Senate Judiciary Chair Chuck Grassley, two federal judges, one in New Jersey and the other in Mississippi, have admitted that members of their staff used artificial intelligence to help prepare orders that Grassley said were error-ridden.

3:09Joining me is former federal judge Paul Grimm. He's the director of the Bolch Judicial Institute at Duke Law School. Are clerks and interns, are they using AI and doing research and drafting opinions? Is that sort of the norm? Yeah, I think right now the genie's out of the bottle on that. Two or three years ago, the law schools were really struggling with whether or not students should be able to use artificial intelligence in their classwork, particularly during exams. And I think that probably the needle has moved to the point where the law schools recognize that this demographic, these young people, they're using this in all other aspects of their lives and to try to expect them not to use it in the practice of law, especially when law firms have invested heavily in the use of generative AI in their own business and practices is a futile thing.

4:07And so I think what the law schools are trying to do now are teach the students the proper use of it and are trying to teach them the do's and don'ts of how it should be used. And I think that as for the intern, I think that was in the District of New Jersey, that use of AI was inconsistent with the judge's own internal policy in chambers and also inconsistent with what the law school's policy was for use of this. So I think that while there's no way of knowing whether it's most students coming out of law school now, I think probably it is. The overwhelming majority are using this and probably are at schools where they're being taught the proper way to use it.

4:48And it's clear that, you know, it's been over a year that the American Bar Association, its ethics committee released formal ethics opinion 512, 512, that deals with the ethical issues associated with the use of artificial intelligence in the practice of law. So that's over a year old. And the fact that the ethics people are already starting to tell people how you use it ethically is pretty clear indication that it's being used broadly in law schools. And I think that the students realize that they're going to be expected to have this skill because the law firms are expecting it, and that's how they're going to survive and thrive in the law school environment.

5:31So I think that probably the answer is, yes, they're using it, and that will only increase. In the New Jersey case, the use of AI resulted in a temporary restraining order that referred to parties' allegations and quotes unconnected to the case. So even if you proofread the restraining order, wouldn't you see that there were all these problems? Yeah, in both of those cases, both the Southern District of Mississippi and also the District of New Jersey, I think that in retrospect, it looks like if there had been rigorous proof reading by the law clerks and or by the court, the judge themselves, that this should have been caught.

6:11And indeed, the Administrative Office of the U.S. Court has a task force which is dealing with the use of AI by judges and chambers that should be a national policy across the United States in all the district courts. And that was issued on July 31st of 2025. And in his response to Senator Grassley's letter asking for comments from the Federal Administrative Office of the U.S. course about any policies dealing with the use of generative AI in chambers, Judge Conrad, who is the director of the AO, made reference to that. So there's even interim guidance. And I think that one of the things that is pretty clear from that guidance is that the judges are responsible within their chambers to make sure that AI is not used for any core judicial functions, such as decision making or adjudication and strongly recommends independent verification before any generative AI is used for any purpose.

7:10There are a number of purposes for which it probably safely can be used, and there's guidance on that. The National Center for State Courts, which is a terrific organization that focuses on the state courts, and of course there are far more state court judges than there are federal judges, issued a 17-page document on August 7th of 2024 titled Artificial Intelligence, Guidance for the Use of AI and Generative AI in the Courts. And again, it goes through extensive definitions of what AI is, what kind of software applications there are, the ones that are problematic, the ones that are more useful, how chambers should operate, and similarly this year, The Sedona Conference, which is a think tank that has been around for many years and is very influential in dealing with cutting-edge technology issues as they apply in the law, issued a monograph in their journal called Navigating AI in the Judiciary that was co-authored by three state court judges and two federal judges and a professor from Waterloo University that gives, in eight pages, some really good guidance to judges as to how they should and shouldn't use it.

8:19So I think the key takeaway here is, you know, the genie's out of the bottle. This stuff is being used. There's growing guidance in terms of how it can properly be used and how it should probably not be used. And just like courts initially were sanctioning and cautioning lawyers who didn't use it correctly when they were citing, Gen AI created content without checking it. I think the judges themselves are learning that lesson that they're responsible for everything that comes out of their chambers. and they have to manage interns and law clerks and make sure that their procedures are complied with.

8:54Coming up next, I'll continue this conversation with Judge Grimm. So how much do judges actually vet the work of their clerks? You're listening to Bloomberg. Support for the show comes from Public.com. If you're actively involved in your portfolio, you probably catch yourself repeating the same actions. Buying the dip, manually sweeping idle cash, putting on a hedge. On public, you can now create AI agents that handle all these tasks on your behalf. Just describe what you want to do in plain English. Like, if the VIX hits 25, buy a put option on the S &P 500. Or, if my cash balance goes above$20 ,000, move the excess into my direct index.

9:33You approve of the workflow, and your agent handles the rest. Monitoring the market, watching for your conditions, and executing your strategies exactly as defined. An investing platform driven by your intent, not just your clicks. You can also get full read and write access to your account via the public API. Go to public.com slash market and fund your account in five minutes or less. That's public.com slash market. Paid for by Public Investing. Brokered services by Open to the Public Investing, Inc., member FINRA and SIPC. Advisory services by Public Advisors, LLC, SEC Registered Advisor. Complete disclosures available at public.com slash disclosures.

10:13So there's a lot of noise about AI, but time's too tight for more promises. So let's talk about results. At IBM, we work with our employees to integrate technology right into the systems they need. Now, a global workforce of 300 ,000 can use AI to fill their HR questions, resolving 94 % of common questions. Not noise. Proof of how we can help companies get smarter by putting AI where it actually pays off. deep in the work that moves the business. Let's create smarter business, IBM. When you own your own business, you own every decision. Now own the card that rewards you for it. The Chase Sapphire Reserve for Business card brings the best Sapphire Reserve benefits to business owners who expect hardworking rewards.

10:56Designed to meet the needs of business owners at scale, this pay-in-full card elevates your travel experience and offers premium benefits and value toward business services that will take your business to the next level. Fuel your business and maximize rewards with 8x points on all purchases through Chase Travel, 3x points on social media and search engine advertising, annual partnership credits, and more. Make every journey more rewarding with a$300 annual travel credit and access to a network of airport lounges, whether you're looking for pre-flight productivity or time to rest and recharge.

11:30Chase Sapphire Reserve for Business. It's the card that gives back all you put in. Learn more at chase.com forward slash reserve business. Chase for business. Make more of what's yours. Accounts subject to credit approval. Restrictions and limitations apply. Cards are issued by JPMorgan Chase Bank N.A. Member FDIC. Two federal judges have blamed faulty rulings on the use of artificial intelligence tools by staff members. That raises questions about how much they scrutinized documents issued under their names. Federal Judge Julian Neals in New Jersey and Henry Wingate in Mississippi admitted to the A.I.

12:10foibles in letters to the Administrative Office of the U.S. courts. The answers were sent in response to questions by Senate Judiciary Committee Chair Chuck Grassley. I've been talking to former Federal Judge Paul Grimm, director of the Bolch Judicial Institute at Duke Law School. So you have interns, you have clerks. Does it depend on the judge how much a judge puts into, let's say, an opinion? Do most judges at least read them over? Yeah, it's a great question. And the answer is it varies greatly with the chambers. I think as a bottom line, I don't think it's a federal judge on the court now or a state court judge who's hearing cases that wouldn't say that they read everything that goes out of their chambers with their name on it.

12:57I mean, unless it's a scheduling order that's automatic, you know, that says, you know, counsel will by such and such a date provide me with the following information and we'll have a scheduling call. But certainly no order coming out of the judge's chamber that deals with a substantive ruling on a case or an evidentiary motion, you know, something that's going to decide the merits. I don't think any judge would say that it's appropriate for a judge to issue an order that deals with a resolution of a substantive issue in a case without having read it and verifying to the satisfaction of that judge that it's legitimate.

13:31Now, some judges have different ways of doing it. What I would do is once I had the initial conversation with them and they sent me a draft, I would ask them to attach to the draft a PDF copy of the cases that they cited as dispositive. Now, as you're aware, lots of times there are cases that are just sort of like a string site. So you've got five cases and there's One, they had a recent court of appeals or Supreme Court case that had the ruling and you were going to rely on it. I would have them attach a PDF copy of that and highlight for me in yellow highlighting the language that they were relying on for what they said.

14:07And so I can independently check it that way. By giving it to me that way, it sped it up for me. I didn't have to go on Westlaw or Nexus and, you know, input it all and read it. A lot of times, you know, there's a lot of part of the case, the initial part talking about the jurisdiction or the standard review that's really not important to what the ruling is, but you want to look at the discussion on one particular point of law. And by having the actual case and having the highlighted section, I then had at my fingertips what I needed to go through, look at the opinion, look at the authority, decide whether I thought, you know, it was adequate for what was necessary.

14:42Look at the party's briefing again and verify that before it went out. But to your question, June, you know, the caseload of an average federal judge probably runs between a low of 200 to 300 cases to a high of a thousand cases. And it depends upon what kind of an order it is and how busy the court is. And, you know, no matter how hard the judges work or what kind of safeguards they have in their chambers, it's possible for things to slip through the cracks. And when it deals with new technology that people are perhaps not as familiar with, there's going to be a learning curve at the start. But I think the cat's out of the bag now that the learning curve time is over and you need to start using it.

15:25You've got guidance that tells you how to do it. I am quite sure, I'm quite sure that Judge Robin Rosenberg, who is the director of the Federal Judicial Center, which is the legal education arm for the federal judiciary, She's a fantastic judge, and I am quite sure that the baby judge schools, the judge training programs that they have every year for new judges and for judges, magistrate judges, bankruptcy judges, district judges, and circuit judges, I am sure that there will be a segment on the proper use of AI there because Judge Rosenberg will want to make sure that the judges have the most recent guidance and are aware of the pros and the pitfalls.

16:08Let's say you have both sides using AI. Do you think that this will lead to less innovative legal theories or legal thinking because they're both using, you know, whatever is on AI? That is a worry. There's a worry among law professors that if students are taught to use AI, that their own sort of creative approach to a subject will actually be the critical reasoning skills. And I suppose if the lawyers are doing that as well, there's always a risk that it applies to them as well. But what they submit, even if they verify it and they get rid of the hallucinations and they submit the stuff which is real, that they got there through a curated analysis that was done for them by AI.

16:53There are some law and technology people, computer scientists, who have raised the specter of a problem that can occur when you've got these large language model programs like generative AI is that, you know, one of the things that you worry about with AI is bias. And it could be intentional bias or it can be inadvertent bias. But when you use a generative AI tool and you put sort of a narrative inquiry, you know, what are the key defenses to an end-of-trust case that alleges the following things in the Second Circuit of the United States courts? and you ask a question like that and it comes out and gives you a list and you follow up with a series of follow-on questions, which is sort of a natural progression, that the algorithm senses what you're interested in, senses what you're looking to do, and starts to provide you with answers, even if they're based upon legitimate legal sources consistent with what it intuits your desired outcome is.

17:59And that's a real problem because, you know, I'm a dinosaur. When I started doing legal research, you know, we had digest. There were no computers. You would go to like contracts and you'd look through this digest that was like an old fashioned encyclopedia. And you'd be looking, you know, and you'd say, is this close to the problem I have? And then you would go to a real book and you'd open it up and you'd read the case. And then when you got a case, You had a whole series of processes to make sure the case had not been overruled or modified called shepherdization. And what you did was in reading and kind of going, triangulating through all these different things, you might say, well, you know, I thought it was contract, but it's really warranty, which is sort of a sub-issue of contract.

18:44but it has its own set of rules. And you kind of would learn through this trial and error that this intuitive approach to what the answer was. And sometimes it would be from a different field that you had initially thought it might be from. You miss all that when you don't have that kind of give and take thinking about it. I can remember sitting in the law library with like 50 books open, you know, all surrounding me, you know, looking at something to try and figure out what my particular case really needed to have to be the solution for the position that I was taking. And I think there's worry that you lose that.

19:23Now, some people would tell you though, June, that if the parties, one thing you can do, one accepted use of AI, if it's done properly, is you can feed into the artificial intelligence generative AI search program, the briefs of the party, and ask the particular tool if any of the cases cited by the parties are hallucinations, are not proper. And you can also say, did they miss any controlling cases in this jurisdiction that deal with this subject matter? And so it may be possible that AI will allow the judge to sort of check to see whether or not the lawyers joint use of it has been siloed to where they missed some authority that the judge really thinks needs to be reflected upon.

20:12But I think that right now, people who talk about the appropriate use of AI worry that that sort of initial creative and analytical process, the thought process that law schools have championed for decades as the critical thinking skills that lawyers have to have, that they may become a sort of atrophy and a little bit less well-used if people rely upon AI too much to the exclusion of the old skills that are learned. Now, we heard some of those same kind of problems about the use of computers back in the day. I remember lawyers saying that, well, you could never use computers in litigation. You could use it in wills and estates because you're using basically forms, the same forms over again, but you'll never use it in litigation.

21:02I think that people would split their sides open laughing at that notion today because computers are used in every aspect of law. So, you know, we're at the beginning of this. And keep in mind, you know, AI has been around since the 1950s, right, in one form or another. But generative AI, because of technological advances, no one was talking about generative AI. I know this because I used to write opinions as a judge and scholarship as an adjunct professor and a professor at Duke about evidentiary issues associated with technical evidence. And no one was talking about generative AI until about 2022 when the whole chat GPT thing came out.

21:44So this is still relatively recent. And what's confounding is how explosive the use has become in such a short period of time. Explosive is the right word. Thanks so much as always, Judge Grimm. That's Judge Paul Grimm, Director of the Bolch Judicial Institute at Duke Law School.

22:23S &P 500. Or if my cash balance goes above$20 ,000, move the excess into my direct index. You approve of the workflow and your agent handles the risk. Monitoring the market, watching for your conditions, and executing your strategies exactly as defined. An investing platform driven by your intent, not just your clicks. You can also get full read and write access to your account via the public API. Go to public.com slash market and fund your account in five minutes or less. That's public.com slash market. Paid for by Public Investing. Brokered services by Open to the Public Investing, Inc., Member FINRA, and SIPC.

23:03Advisory services by Public Advisors, LLC, SEC Registered Advisor. Complete disclosures available at public.com slash disclosures. So there's a lot of noise about AI, but time's too tight for more promises. So let's talk about results. At IBM, we work with our employees to integrate technology right into the systems they need. Now, a global workforce of 300 ,000 can use AI to fill their HR questions, resolving 94 % of common questions. Not noise. Proof of how we can help companies get smarter by putting AI where it actually pays off, deep in the work that moves the business. Let's create smarter business.

23:40IBM. When you own your own business, you own every decision. Now own the card that rewards you for it. The Chase Sapphire Reserve for Business card brings the best Sapphire Reserve benefits to business owners who expect hardworking rewards. Designed to meet the needs of business owners at scale, this pay-in-full card elevates your travel experience and offers premium benefits and value toward business services that will take your business to the next level. Fuel your business and maximize rewards with 8x points on all purchases through Chase Travel, 3x points on social media and search engine advertising, annual partnership credits, and more.

24:16Make every journey more rewarding with a$300 annual travel credit and access to a network of airport lounges, whether you're looking for pre-flight productivity or time to rest and recharge. Chase Sapphire Reserve for Business. It's the card that gives back all you put in. Learn more at chase.com forward slash reserve business. Chase for Business. Make more of what's yours. Accounts subject to credit approval. Restrictions and limitations apply. Cards are issued by JPMorgan Chase Bank N.A., member FDIC. NBA Commissioner Adam Silver spoke out on Friday about the bombshell gambling scandal that's rocked the league.

24:55I was deeply disturbed. There's nothing more important to the league and its fans than the integrity of the competition. And so I had a pit in my stomach. It was very upsetting. Shock waves ripple through the sports world after Portland Trailblazers head coach Chauncey Billups and Miami Heat player Terry Rozier, along with 30 others, were indicted as part of two separate criminal schemes related to illegal sports betting. Joseph Nocella, the top federal prosecutor for the Eastern District of New York, described the two schemes. The first involved sports betting. Between December 2022 and March 2024, these defendants perpetrated a scheme to defraud by betting on inside non-public information about NBA athletes and teams.

25:50The non-public information included when specific players would be sitting out future games or when they would pull themselves out early for purported injuries or illnesses. The second involved rigged poker games. These defendants, which include former professional athletes, used high-tech cheating technology to steal millions of dollars from victims in underground poker games that were secretly fixed. The games in the New York area were backed by the Bonanno, Gambino and Genovese crime families. Billings and Rozier face money laundering and wire fraud conspiracy charges. Joining me is Derek Hogan, a partner at Tully Rinke.

26:35Tell us about the charges here broadly. You know, the first one involved providing, let's say, inside or confidential information to people placing wagers on games. So, for example, the NBA player involved, Terry Rozier, there are certain bets that an individual can place, you know, whether a team wins or loses. the score goes over a certain amount of points, or there's something called individual prop bets where bettors can make bets on whether a certain individual is going to score more or less than a certain number of points, get a certain amount of rebounds, certain amount of assists, whatever statistical categories they put out there.

27:11And in this instance, it's been alleged that Mr. Rogier purportedly told individuals that he was going to leave a game early, you know, before he reached a certain amount of points, rebounds, assists, and that if the individuals bet the under on that, then they would obviously win their bets. So it's alleged that he passed along that information and then that other people took that information, distributed that information to others and distributed information to others to make bets on it. And I think it's important to note in those types of situations that each league, whether it's the NBA, NFL, Major League Baseball, National Hockey League, at some point they issue like an injury report.

27:53And that injury report will let people know who's playing, who's not playing, who may be playing, probable, questionable things of that nature. And that information is released. So ultimately the bookmakers can put odds out there that are, you know, equivalent to who's playing, who's not. And so what was happening, Mr. Rozier's situation, you know, that information was allegedly being disclosed to others prior to any type of injury report or not even making it on your injury report. It's basically inside information. Think about like in a stock tip, right? Like you might know about a company going public before it's supposedly out there.

28:30And then you use that inside information to benefit to yourself. The other involved an illegal poker game in a sense that it was a cheated card game. It was through the use of technology from what I read, whether it's like technology within the table that the players were using, technology within the cards that they were using, essentially that it was a rigged poker game. And where the high-profile defendant, Mr. Chauncey Billups, who was the head coach of the Portland Trailblazers, it's being alleged that he was kind of used to lure individuals to that game, to lure like an unsuspecting victim to that game.