10/16/2026: Microsoft Launches Copilot for Legal, Meta warns deep fakes may not be detectable, and more

0:00Victor Li: I had this message pop up right behind my camera. All right, welcome to Legal Tech Week for October 16th, 2026. If you don't know where you are, this is the show where we talk about the top news in legal tech and innovation from the past week. My name is Bob Ambrogi and I have a blog called Law Sites and another podcast called Law Next. And I have a podcast called Legal Tech Week, by the way. which this is. And our panelists today, we've got joined by a sometimes occasional sit-in guest panelist, Doug Austin. Doug, why don't you introduce yourself?

0:45Bob Ambrogi: Thanks, Bob. I'm Doug Austin, editor of eDiscovery Today, a daily blog about eDiscovery and related topics. Been daily blogging for 16 plus years. eDiscovery Today is about six and a half years old and looking forward to the discussion.

1:01Victor Li: All right. And Nikki Black.

1:06Niki Black: My name is Nikki Black. I'm the principal legal insight strategist at 8am, the team behind MyCaseLawPayCaseSpear and DocketWise. And I write legal technology columns for ABA Journal Above the Law and the Daily Record. All right.

1:23Victor Li: Next up, Victor Lee.

1:25Bob Ambrogi: Hi, everyone. My name is Victor Lee. I am assistant managing editor for the ABA Journal. I'm covering business of law and technology. I am not going to be the White House press secretary because I actually have passed the bar.

1:38Victor Li: And last but not least, Steve Embry.

1:41Doug Austin: Hey, everyone. Steve Embry here. I write the blog Tech Law Crossroads, and I write for Above the Law. And I guess I can't be a press secretary either. I think I still have my license, even.

1:56Victor Li: I still have my license. but i mean it is kind of an i thought it was actually an interesting question about when do you call can you call yourself a lawyer when you haven't actually passed the bar i mean i think a lot of people do that frankly despite all the uproar over it but i mean i often refer to i know people like general counsel of corporations who don't have their law who aren't admitted to a bar or at least in the state where their corporation is i don't know if that really matters but anyway that is not a legal tech problem and we don't have joe here to talk about all the stories he wrote about it this week joe is already uh joe says he's already on route to orlando where some of us are going to be going this week for the net documents conference uh and uh continuing our string of way too much way too many conferences and way too much travel for the fall season.

2:48Victor Li: Get back from that. And then we got Clio in another week or so. So a busy time. All right. So I think probably the top story of the week was this news from Microsoft that it has now launched its co-pilot for legal following the lead of Claude, which of Anthropic, which had launched Claude for legal and Google, which had launched Gemini Enterprise for legal and OpenAI, which just launched Astra for law, not all that long ago. And so now we have the four big major frontier models, all having launched extensions into the legal space, which is interesting in and of itself. I remember, I mean, I'm old enough to remember when we all panicked last January.

3:50Victor Li: It was January or February when Anthropik came out with that like one plug-in and suddenly the stocks dropped for Thomson Reuters and RELX and Volters Kluwer all in the same day. and everybody thought that was like the end of the world. And now all of the companies have, all the frontier models have done it. And, you know, of course, what makes Microsoft maybe different than all of those others is that lawyers already live in Microsoft products. They're already in Word. I thought it was, you know, when the ILTA Tech Survey came out last month or so, when they asked lawyers, what AI are you using?

4:36Victor Li: The leader by a wide margin was Copilot. So lawyers are already using this. So it seems significant. I think it raises questions for, what are the intentions of these companies in terms of the legal market? And what does it all mean for vendors and law firms and everybody else with their own AI products? What do you all think?

5:06Bob Ambrogi: I'll start. I feel like MCP is like one of the acronyms of the year because everybody's touting model context protocols to these different platforms. And certainly we saw a couple of weeks ago, you know, Relativity talked about their focus was to be, you know, able to interface with the frontier models. So I think that's interesting. We also saw this week that Cooley is doing something directly with Google on a redaction capability redaction tool for court for case filings. So maybe we're going to see more of that where we'll see some providers and law firms and what have you working directly with the frontier models instead of with the legal tech provider vendors.

5:58Bob Ambrogi: But I don't know. We'll see. But certainly everybody, I think, is covering their bases by making sure they can connect via MCP to these platforms.

6:08Doug Austin: I think it's kind of a startling development in a way. I mean, most of us can remember when a lot of these big companies like Microsoft paid scant attention to the legal marketplace. I mean, yeah, Microsoft did a little dabbling here and there, but didn't really seem to care. And when OpenAI and Anthropic first came out, they didn't seem to be terribly interested. and now they're kind of falling all over themselves to get involved and coming on the heels of the orgy of money at ILTA this year. It sort of is just like, you know, has legal tech suddenly got all that advantageous, all that advantageous and has the ability to make so much money or was it there all the time and nobody really knew it?

7:05Doug Austin: But it's crazy in a lot of ways.

7:09Bob Ambrogi: I guess it shouldn't be a surprise, except for the orgy part, but it shouldn't be a surprise that Microsoft, I mean, because, yeah, like, you know, everyone else did it, so they have to do it too. But, you know, and someone said low friction in the comments, and we talked about this in the show. I mean, you know, so many lawyers already use Office. So many lawyers use Microsoft products. You know, this is a natural move for them. And if that allows them to kind of hold on to some of those users that might otherwise be like, you know what, I'll use Google instead or I'll use, you know, I'll use like, you know, one of the other products.

7:40Bob Ambrogi: And this gives them a built-in advantage because they already have so much of the market captured. So, no, it was a good move. I actually am probably surprised that it took as long as it did. I think it's an indication that what you're going to start to see in law firms is what you're also seeing across large companies.

7:59Niki Black: and it's what Joe always talks about, the cost, like the pipe, you know, you're going to have to start paying the piper here. And because the costs are increasing and they're going to continue to increase, I think you're going to see firms, especially the larger they get, but even in the smaller firms, tell users, tell their employees that for certain types of tasks, you look at the tools that are already built into, and we're not going to get charged for, you know, like Google, like the practice management software or the document management software you're already using. Look at those tools and the capabilities in those and use that for all of these tasks, like correspondence with clients or whatever the case may be.

8:41Niki Black: And then when you have these use cases that are more complex, then we're going to try to find the most reasonably priced piece of software we can that will allow people to accomplish their goals within that software. But I think they're going to start looking at the tools that they've already always had in place, Microsoft, and whatever their document management, billing, practice management, CRM tools are, and the capabilities built into those, Google Suites, so Gemini. and it's only going to be for the more complex tasks, you know, that they're really going to start saying for these complex tasks, we're going to have these specific tools and only certain people can use them because they're the ones that really need to use it for these complex tasks.

9:24Niki Black: But I think cost is going to increasingly matter. And I think that the companies that are already entrenched in these, whether it's not an additional purchase or it's just up a tier rather than a whole brand new purchase, you're going to see more and more firms looking at it that way.

9:40Doug Austin: Yeah, I think you're right, Nicky. And, you know, Microsoft is so entrenched and it's beyond just being entrenched. I mean, there's sort of a trusted commodity by a lot of law firms. I'm not saying whether they should be a trusted commodity. That's a different issue. But it'd be really easy for management to say, just use the Microsoft tools. We're already paying for them. And then there are good tools. We've used them for 35 years. So why would we make a change?

10:09Victor Li: I think one of the things, Doug, you mentioned MCP and the rise of the MCP. I think it's been interesting. At first, people were kind of saying, oh, and by the way, we have an MCP connector. Now they are touting this in a huge way. I mean, I've had two different demos this week where companies were touting their new partnerships between two companies. And you get in on the demo call, and it's like, you've just created an MCP. That's all you've done here. I mean, it's not all you've done. it's a cool thing it's a wonderful thing but they're now like portraying this as as a major integration a major partnership um and everybody's doing it it's like suddenly mcp is the thing you know i think we talked last week on the show or whatever but i was so struck i mentioned this before about phil saunders comment when we were at rel fest about about the fact that relativity is now focusing its sort of AI partnership more with Anthropic and OpenAI and not with the Harveys and Legoras of the world because they think that's where everything is heading, that law firms are all going to have one or the other, either Claude or ChadGPT on their desktop.

11:26Victor Li: And that's how lawyers are going to be using the sort of the power of these frontier models against through mcps against these different data sources and and and whatever else uh they're accessing um you know i'm not so worried i'm not too worried about harvey and lagora actually i mean i used to be worried about them i kind of wondered how they were going to last but i think they'll be around for a while but i do wonder what this means for more of the sort of you know some of the products at the at the the smaller products the more specialized products that can now be replaced by maybe some skills in clod or something.

12:07Doug Austin: It's funny when you were talking, Bob, about, you know, it was major partnership announcement, which is really just applying an MCP. You know, it wasn't that long ago when there's a major AI announcement and we'd all go and listen to it and walk out of the room going, that's new? That's like, really? Right.

12:27Victor Li: Well, for the longest time, it was everybody building something on top of AI and calling it their own AI. And now they're just kind of like throwing in the towel in a sense of saying, OK, that AI is really good. I'm never going to build anything like that, but I've got this cool other set of data or features or whatever. And so why don't we connect Cloud or OpenAI or whatever into what I'm building here? I don't even care if you come to my platform anymore, just as long as you're paying me a subscription fee and using what I have. It's really, it's like a whole different way of thinking about legal tech.

13:07Bob Ambrogi: It's the ultimate hedging of bets. I mean, it really is. It's kind of like, okay, we're building these things, but we're also building the connections too. So one way or another, we're trying to make sure we're a part of the process, a part of the solution. Yeah.

13:22Doug Austin: It is hedging your bets, but you're also placing a pretty big bet on the quads and the chat GPTs and Microsofts of the world, as I mentioned before. Sometimes it worries me a little bit that all these vendors are connecting by and through these large models, and that gives those models a pretty big control over access and price and terms and everything else. So we shall see, I suppose. Tech moguls controlling legal tech.

14:01Victor Li: I don't know. Yeah, actually, one other point in that is just because I was thinking a lot about this is the, I mean, we, you know, when this all first started happening, I mean, so this is November's four years, you know, since ChatGPT 3.5 came out. and you know we've talked a lot about this over the years and talked about how it might change law practice and you know it's clearly I feel like it's one of those things it's sort of so slowly and quietly changing things that we're sort of not even seeing at all we're not seeing the forest for the trees in a sense right now but I mean when you really think about what's happening right and just what's happened over the last six months over the last five to six months with with these companies the big frontier models coming into the space with mcps taking off it's completely changing how companies think about building legal tech how lawyers use legal tech how we interface with legal tech it's like it's it's eliminating the not eliminating but it's uh reducing the importance of platform uis and and all of that uh you know and if you're if you're dramatically changing how lawyers use and access legal tech, then that means you're changing how lawyers practice law in pretty significant ways.

15:24Victor Li: So, I mean, this is really starting to snowball, I think, in terms of the impact on a legal profession.

15:30Niki Black: And I think the bigger picture, you know, when you pull back and look at the bigger picture, I feel, I don't know how you guys feel, but I feel pretty confident that the impact of AI so far has been mostly negative. Like I, both on the profession and on the culture overall, the AI slop is everywhere. Like it's inescapable. And even, and we're all starting to talk like it. You just said, Bob, and I don't know if it was you picking up on the AI slop or not, but we're all incorporating into our language. You just said, it's quietly doing something. That's such an AI term to say things are quietly happening.

16:09Niki Black: And we're even starting to speak that way. You know, it's showing up everywhere. I think it's ruining so many things. It's ruining people's ability to think. It's ruining analysis. It's ruining slideshows. It's ruining advertising. It's ruining even just the flyers you see around in places. Like, everything's getting ruined by it. And I don't think that the impact on legal practice at this point has been beneficial. The courts are overwhelmed by AI slop. stop the and the all anyone's doing is writing ethics opinions and cases and there is so much time being devoted to trying to manage and address AI in the legal space so I four years out I don't think that the impact has been positive and I don't see like huge productivity gains that everyone's talking about I just see a lot of money being spent um and a lot of um risk being taken and I don't see the benefits happening yet so I mean we'll see but I just feel like we're on a freight train headed towards a, I don't know, or like around a ship that can't stop or something like we're headed in a really weird direction and we're going there fast.

17:18Yeah.

17:18Victor Li: I'm going to quietly ignore that you just insulted me, but I won't.

17:23Doug Austin: I think I'd written something about that too, Mickey, you know, particularly for litigators and trial lawyers. It, it's like a, like you have to now speak, you know, You have to speak an AI slop if you want to persuade somebody of something. And it's like, which just makes it worse and worse and worse. I agree with you on a lot of those points. I mean, it's overtaking us.

17:50Victor Li: Yeah, it's hard to say which way we're going. Because what's the AI detector, the pangram? Is that what it's called? Yeah. Something like that. I was reading somewhere where they're saying it's getting harder and harder for them to detect AI language. in part because the models keep evolving, but in part because the language keeps evolving based on the language that everybody is using and that influences the language the models are using. And so it's this chicken or egg kind of situation of slop. I think I used things like quietly before. Sometimes I get accused of AI slop for things I've used for years.

18:28Victor Li: As Stephanie and I have said many times, I'm a long time hyphener person. And I love the M-dash and long live the M-dash.

18:39Niki Black: And just to be clear, I'm not accusing you of, I think it's just a seep, it's seeping into the way that we speak because we see it so much, you know? And that is a term that has a great place when you use it. And with the way you used it made sense, but I also think we're all sort of guilty of you just see it so much that you start to adapt it into your language. Because it is a great word when you're trying to talk about something happens subtly or behind the scenes it's a lot easier to say than behind the scenes you know but i don't know i i i'm not sure i disagree with your basic i mean that i i'm sure i agree with your basic premise i think that ai is in fact uh having a

19:19Victor Li: lot of positive impact on on law practice uh you know like again just like one example just take like the litigation example where you've got somebody who's working on a litigation matter who's in a some kind of a platform now where they've got all of their case data and now they're able to kind of very easily take all of that case data and associate it with perhaps legal research data and then put all of that together into preparing pleadings uh and it it it not it It does make you more efficient in some ways. Yes, you've still got to review it all and you've got to go through it all, but it lets you pull together all these pieces that used to be in different platforms, pull them together and make a lot more sense of them, evaluate, you know, like just a simple thing like timelines.

20:09Victor Li: I mean, Steve used to do a lot of litigation. I mean, I'm sure you can appreciate how hard it used to be to do a timeline and how easy it is. It takes about five seconds right now to get a great timeline done. And there's so many ways that it really is making law practice easier. You know, I don't know.

20:27Niki Black: Well, the counterpoint. So I litigated myself and part of the way that you master the facts, part of the way that I currently am able to speak fluently about the things about which I speak is because I read and analyze cases and opinions and other people's writing. And then I write about it and then I write about it again. and then I talk to someone about it, and then I'm able to speak about it in front of a bunch of people without looking at anything, without having notes, because I've mastered it by going through that whole process. And when you're litigating a case, you have to know those facts like second nature.

21:03Niki Black: And there's sometimes hundreds, thousands of them, right? And the way that you master those facts is you create the timeline. You go through and you find all those pieces and you put them together and you create this timeline. And then you know, all right, I know that the blue card drove by first because I put it, you know, you remember the timeline, whereas when it creates the timeline for you, that's helpful. But then how do you go through, I mean, I am not litigating anymore. So I don't know what the answer to this is. But how do you master those facts in a way that has that process? And you know, they talk about you write it down, and then you read it, and then you talk, say it out loud.

21:38Niki Black: And those are the ways that you learn, then you write about it, like you write a brief. So when the software is doing all this for you, how are you able to actually litigate the case and be artful about it and be able to pull these facts off the top of your head. I don't know what the answer is because I'm not doing it. So, but I worry about that too. I don't know. What do you think, Steve? And I know victory is mitigated.

21:57Doug Austin: Yeah, it's, I think it's, it's not a problem for the, for the more experienced lawyers or I think it's a real problem is for the younger lawyers that haven't acquired those skills. I mean, you know, we had to do it all ourselves. And you're right. That gave us some knowledge and understanding and grasp of the nuances that you wouldn't have if you just used AI. The problem, I think, is not some use of AI. It's the temptation to let it do everything in a case. draft all the briefs read all the documents that's where you got that understanding it's a fair point I worry a lot about the younger lawyers that are coming out that are so tempted or may be forced to use these tools

23:00Bob Ambrogi: we've been talking about this ad nauseum AI is not going to help you do something if you didn't already know how to do it it might help you fake it or it might help you, you know, get to a certain level, but it's not going to make you like a better lawyer. It's not going to make you a better writer. It's not going to make you a better, you know, whatever. I mean, you know, I mean, and you know, like, so Nikki's talking about all the power work that she's done beforehand. It's like, well, yeah, that's because you're using it, you're using it in a way to help just to just cut your time. You're not using it to learn what this stuff is.

23:33Bob Ambrogi: You're not using it as like a substitute for actually having done the work. You've done the work. It's just now this is just a way of just shortcutting through all that stuff so you save time. And it's like, and so for the people that can use it that way, like can, you know, build on their knowledge and their experience and just use AI as a shortcut, then it's great. You know, it saves you so much time. It opens up possibilities that never, you know, never, that never existed beforehand or were prohibitive for you. And so, you know, it'll help you be able to do things that you wouldn't have been able to do beforehand and do them better and do them more efficiently.

24:05Bob Ambrogi: But if you didn't know it or if you weren't good at it or you didn't have the experience, it's not going to be a cover for that. Yeah. Yeah. I mean, I think that there's always got to be a combination of using AI and using your skills. And one of the things, maybe it's because I have trust issues. The case that I covered this morning has four different rulings to it. And so I thought, well, I'm going to throw them into chat GPT and say, give me a timeline of the events, including the rulings. And so that was a great start for me to understand the case and what happened. But then I went back to the rulings to say, OK, so they're talking about this here.

24:44Bob Ambrogi: I want to see what the judge said about this and so forth. So to me, I didn't just take it straight as it was. I went and really tried to understand it. But I used that to help me guide to kind of pick what parts of the rulings I really wanted to focus on.

24:58Niki Black: So I guess just one example that I feel like AI, I also worry that it's not going to make these connections that we as humans make sometimes that can be pivotal in litigation. And my example of this is years and years ago when I was a public defender, I had a simple case. It was a prostitution case. And without getting into the specific facts, you know, my client gave me this. He said, this isn't actually what the police report says isn't what happened. What happened is X, Y, and Z, not X and Y. And so, and X, Y, and Z was a kind of strange set of circumstances that stuck in my head. And the night before trial, I was Xeroxing.

25:39Niki Black: And I saw people would put funny accusatories up on the bulletin board near the Xerox machine. and their lo and behold was the same cop on an arrest report saying x y and z happened not x y and it was the exact same kind of case and so i was able to use that other police report um to cross-examine the cop and i got him to agree that he would never do x y and z and i'm like never no i would never do x y and z and i pulled out this report from another case and he had to admit that he'd done it um but like i feel like now that's an example like physical documentation and stuff, but, and it's two completely different cases, but I feel like when you're using AI, that those types of connections aren't going to be made, and that seasoned litigators are going to skip over steps that sometimes would lead them down, I mean, that's sometimes when you get those smoking guns in cases, it's not always just a piece of evidence, it's making the connections between a bunch of different pieces of evidence, and sometimes the connections aren't always clear, and so I just really worry that AI is going to just make everyone sloppy, even the seasoned

26:43Victor Li: litigators and i don't know i mean those are legitimate worries but keep in mind i mean you were your your experience was in a smaller firm at a lot of larger firms where the litigators where the people who are going into the courtroom are not the people who created the timeline who went through all of the evidence who did all of that they had associates working for them who did all of that so is it really much different for an experienced litigator if an associate creates a timeline or if AI creates a timeline, I don't think it is. I mean, they're still going to have to sit down and absorb that and review it.

27:17Victor Li: I mean, they're going to have to know the stuff before they go into the courtroom in order to handle the case, however it gets prepared. I do totally agree that the de-skilling at the lower end of the totem pole, in a sense, is where the real threat is, that how younger lawyers learn these things. But I mean, this is something else we've been talking about for years because nobody goes to trial anymore. I mean, you know, even younger associates, I mean, how many cases ever go to trial? I mean, you know, some of us when we were younger, we got to do a lot of trials and, you know, just get kind of thrown into the pool to see if you can swim sort of thing.

27:58Victor Li: And that just doesn't happen now, even without AI. So there's just a real issue around de-skilling and younger lawyers and training. Anyway, we should probably move on to another topic before time runs out. And Doug, as a guest panelist today, you get honors to tell about your story this week.

28:25Bob Ambrogi: Okay. Thanks, Bob. So I covered a case this morning. Hat tip to Kelly Twigger, who actually covered it earlier this week. Volta, basically a former employee of the plaintiff, he was terminated, joined a competitor. Plaintiff sought a temporary restraining order and preliminary injunction to stop him from soliciting customers and employees. Basically, one of the debates here was whether he signed the plaintiff's non-solicitation and confidentiality agreement, which despite the fact that they had a signed copy, He claimed he didn't sign it. So they they basically after a after a status conference, the court said, OK, you're going to have to authorize Yahoo to, you know, for subpoena.

29:18Bob Ambrogi: You're going to need to provide access to your Yahoo account where emails that might be, you know, sending to the completed agreement might be and so forth. And so he dragged his feet to provide access to his Yahoo account for a couple of weeks. And they also provided access to its devices. And the forensic examination of its devices determined that the guy had queried ChatGPT on Yahoo's ability to retrieve deleted emails. He basically said, can Yahoo retrieve an email deleted from his account a year ago, even if it was under a subpoena? ChatGPT said no. He asked how long an email permanently deleted would still be retrievable.

30:03Bob Ambrogi: And it said at most seven days. And then he asked and he finally asked if Yahoo maintains a log of permanently deleted emails. And it said no log is is is kept. So that ultimately, with those emails not available, and subsequently they tried to, they expanded Discovery to go after his mobile device and his laptop, which he claimed he was using the same one since he started working at the company. Turns out he had changed mobile devices, and the laptop that he provided to the forensic examiner wasn't purchased or sold until 2024, and this was issues in 2022. So neither of the devices were at the time.

30:54Bob Ambrogi: So ultimately, they not only got costs and fees, but once they realized that the ESI wasn't recoverable, they slapped on adverse inference sanction and also preliminary injunction where he can't go after any, he can't do anything related to his former employer. So it's just kind of another example now of, you know, AI as evidence. And in this case, obviously, direct evidence of his spoliation kind of activities related to the evidence that's, you know, crucial to the case. So I thought it was an interesting one. And it was essentially over four rulings over about a year where all this transpired and what have you.

31:40Bob Ambrogi: So kind of an interesting case. It seems like just, you know, I'm seeing like one or two AI cases a week now in the case law that I'm reading and so forth. So they're just everywhere.

31:53Doug Austin: I have to say, Doug, when I read your report of it, and I read at the beginning that the guy was claiming he didn't sign an agreement that has his signature on it. I sort of said, I know how this is going to turn out.

32:07Bob Ambrogi: Yeah. Yeah. I mean, it's, you know, he claimed he didn't have electronic signature capabilities. So that was part of his, you know, declaration and why, you know, he was kind of saying, okay, you know, that can't have happened. So that's when they were going after his devices, which turned out not to be the devices he was using at the time and all that. So, yeah, there, you know, but the thing was interesting, you know, forensic examination to uncover the chat GPT activity. And we had another case earlier this year where a guy was trying to get out of a deal. So he was acquiring, inquiring of chat GPT.

32:43Bob Ambrogi: How will I how will I force them out and how can I take over their company and so forth? And that ultimately came back to cost him in that litigation, too. So these are the types of things we're going to be seeing in case law more and more, I think.

33:01Victor Li: I was most shocked by the fact that anybody is still using Yahoo Mail. But that's another thing.

33:06Bob Ambrogi: Right. Well, there's that. So I still have my Yahoo account. I mean, it's because, you know, I had it since after. I never had emails when I went to college and then they cut us off after we graduated. So I needed something. So Yahoo was the first thing that I went with. And so I still have mine. I don't use it that often, but you know, I probably still have it.

33:25Victor Li: I may still have a Yahoo email address, but I have no idea.

33:29Doug Austin: Interestingly enough, Victor, I still have my Yahoo account and I do use it. You know, when you, you know, you're supposed to give your email address to take advantage of this offer. And you know that as soon as you do that, you're going to get 10 ,000 solicitations. And so that's where my Yahoo, and I use my Yahoo mail and I never bother opening any of them. Yeah.

33:51Bob Ambrogi: Well, yeah. I mean, that's probably, I mean, probably most of us have old dormant email accounts because Naveen said, I'm impressed with folks still using AOL.com. I still have an AOL account. I'm hardly ever in it. You know, I've switched to using Gmail, you know, a long time ago. But that AOL account's still out there. There's still emails. Occasionally something comes across and I'm like, oh, okay, maybe I should check this out. But, you know, email accounts, especially public email accounts, never seem to die.

34:21Victor Li: So they just still have your three and a half inch floppy disk that would

34:26Bob Ambrogi: activate. Right. Yeah. With the, yeah. And my, what was it? 28.8 or 56 point, whatever. Modem. 14.4.

34:36Victor Li: Yeah. 14.4. Yeah. 14. Yeah.

34:39Bob Ambrogi: Exactly.

34:42Victor Li: So much simpler. Life was so much simpler then.

34:45Bob Ambrogi: Kind of was. But yeah, that's right. Yeah. MFA. Anyway, I thought the case was interesting.

34:52Victor Li: I, you know, it's I mean, it is evidence of, you know, be careful what you search for and be careful how you use chat GPT. But it was also it is also interesting that the judge, you know, basically use that to make an adverse inference about about what was really going on here.

35:13Bob Ambrogi: Right. Yeah. I mean, I mean, that seems like pretty clear evidence of him because, you know, it was a seven day period and he had already inquired two weeks before the status conference. So it was long gone before the status conference ever happened. So, yeah, the timing of it was hard to ignore. Yeah.

35:35Victor Li: Well, speaking of judges and AI, Nikki, you have a story on judges and AI.

35:42Niki Black: I do. This caught my eye just because I like the ethics cases and this one was a judicial ethics case. And it basically focused on the idea of what's a judge to do when they encounter AI and ethically, what can they do? And essentially, the standard that was set forth is there's a two-pronged test, which is to decide that the... Oh, I thought I had that high one second. The applicable standard is that if they have information indicating a substantial likelihood that an attorney's committed a substantial violation of the rules of professional conduct, they have to take appropriate action. You know, hallucination, you know, is a violation of your ethical obligations.

36:35Niki Black: But then it's a discretionary decision, what is an appropriate action. And they have to consider whether the lawyer has a history of unprofessional or other conduct in violation of the rules and any other relevant conduct or factor. And then once they've determined, you know, that they have, once they're trying to determine what's appropriate, there's a lot of different courses of action, including reporting the attorney to the disciplinary committee, counseling and or warning the lawyer, reporting the lawyer to his or her employer, sanctioning the lawyer, and, you know, presumably making the lawyer right on a, you know, chalkboard.

37:08Niki Black: I will not do this again. I don't know. But, you know, they've got a lot of different options. But then they did say that if you do report it to the disciplinary committee, it doesn't always have to be done, but if you do this, there are ramifications you need to be aware of. Specifically, the judge in all cases involving that attorney, both while the matter is pending and for two years after, must be disqualified from the case. So they've got to disqualify themselves for many cases involving that attorney for the penancy of the disciplinary action and up to two years afterwards. So I just thought it was interesting because it gave the judges a lot of discretion.

37:51Niki Black: And there's a lot of different things that judges can actually do. And it also made it seem like it was going to complicate the judge's life a little bit if they did send it off to a disciplinary committee. So, I mean, I think that was a requirement, but I also think that that is one of those systemic things that you're seeing it happen here, but it just happens across the board. Sometimes when you make something a pain and they ask, people aren't going to do the right thing. And, I mean, is it a pain? It just means you have to go through that whole disqualification process every time this lawyer ends up in your court.

38:21Niki Black: So, I thought that was an interesting ramification that might cause judges to do, especially in smaller communities, And especially like if perhaps this attorney handles a lot of assigned cases that, you know, ease the judge's burden, you know, in terms of addressing all these cases in front of him or her. So it might be the situation where they'd be less inclined to report him to the disciplinary committee. So I thought it was an interesting case. I like to track those ethics opinions. And, you know, it was interesting to see it applied towards judges this time instead of attorneys. So I don't think it's earth shattering, but I love to track those things.

39:01So.

39:02Victor Li: But it also speaks to the increasing frequency of this issue and the fact that they need to the judges are trying to, you know, are dealing with how do we respond to this and what do we do when we encounter it and we see it. It's it's it's something that's happening more and more. You mentioned in your article that the was a name, Damien Charlatans tracking of all the hallucination cases, which is well up, well up over 2000 cases at this point. So it's a fact of life these days.

39:34Doug Austin: I thought the disqualification angle was really the most interesting part of that. I mean, it didn't strike me that that was that obvious of a thing that a judge would have to do. I mean, if a judge reports a potential violation of ethics to the Bar Association, I guess the Bar Association does the investigation. It's not an adversarial proceeding per se, but it's not the judge versus the lawyer. It's more like I'm reporting a crime or a possible crime, and you guys go decide and deal with it. But I guess if the judge was going to be a witness or something, but even then, not sure. I mean, it's either hallucinated or it's not, right?

40:22Doug Austin: It's not terribly, not a lot of gray area there. And if it is hallucinated, then their sanctions are appropriate or they're not. It really, I don't, so it struck me. I just couldn't quite see how you would get to that, necessarily get to that decision.

40:46All right.

40:49Victor Li: Something we used to talk a lot about on this show was whether law firms were going to make their employees go back into the office. And it's been a while since we've talked about that. And yet. It's in the news again this week. Victor, you got to start.

41:13Bob Ambrogi: yeah so um so i guess well well before i just wanted to plug a podcast that i did this week with uh damien real clio where we talk about whether ai could end humanity and actually you know it is possible so there's that um but anyway yeah so i chose this if damien says that was true now i'm scared he thought he thought he thought we might be able to we might be able to squeak by um but um uh so yeah um the reason why this caught my attention this quinnemmanuel thing so i interviewed um uh i don't remember exactly that one of the one of the one of the people in charge i don't remember his managing partner or what um right when right when they announced their policy that they would let they would let all of their all their lawyers work from wherever they wanted um and you know they were very gung-ho about it this was you know This was right after some of the pandemic restrictions started to lift.

42:11Bob Ambrogi: But he was very gung-ho, like, oh, this is how it's going to be, and we feel very good about it, and we think our attorneys are more productive when they can work from wherever they want to, as opposed to forcing them to come to the office, blah, blah, blah. So they were very strong, and they felt really good about this policy. So to then see them kind of hedge back on it a little bit was interesting. And I guess kind of a sign of the times. Because after all, again, it's like these lawyers, these big law firms, especially a big law firm like Quentin Emanuel, that they have these big offices. They're probably paying a lot of money on the leases and whatnot.

42:50Bob Ambrogi: And just law firm culture in general is just this idea that you need to come in and show your face. and, you know, they'll have like those late night water cooler sessions or whatever they call it where, you know, inspiration and genius, you know, meet and intersect. And so basically like they're moving off in the sense where it's like they're not telling people that they had to come to the office every single day or every three days or every two days or whatnot. It's now it's like associates now need two years of experience before they can work outside the office. And I think people who ladder all in, there's some restrictions as to whether or not, as to how soon they can get that work from anywhere status.

43:29Bob Ambrogi: So they're moving off a little bit, but they're not forcing people to have to come back in. It does beg the question whether or not the next step is to just be like, you know what, we're just going to make everyone come in at least, you know, two, three days a week and whatnot. So I thought it was just interesting just how, yeah, like one of the big proponents of this whole, you know, remote is the way to go. You know, we don't all have to be in the same place. And now they're coming back off that a little bit.

43:52Doug Austin: So, yeah, the reason I was chuckling about it is as I've read it. So you have to have two years of experience if you're an associate being in the office and then you have to had to spend one year as an associate in the office. And so the whole notion of being in the office for younger lawyers is is your water cooler bumping into the partner at the water cooler, which in my experience never really happened. And but so so but now you've got requirements for associates to be in the office. But I didn't I didn't see that word partner in place, that requirement. So the associates are all going to come into the office and the partners are not going to be there.

44:35Doug Austin: So I guess they can enjoy talking to each other. And I saw the thing about the lateral and I wondered to myself, if you brought if they bring in a lateral with a$10 million book of business, Does that lateral have to come to the office every day for two years? Really?

44:55Doug Austin: That's true. That's probably some flexibility there. What's this line from the Johnny Depp movies? It's really not a requirement. It's more like a guideline.

45:09Bob Ambrogi: The rules don't tend to apply to people who bring in that kind of money. Right.

45:18Victor Li: So it's a policy for some and a guideline for others, and maybe not a very effective policy, but interesting.

45:28Bob Ambrogi: But then again, I mean, like we were talking earlier about sort of like, you know, young lawyers who, you know, don't get the training or don't get the experience or whatnot. I don't know if this would necessarily address that because, you know, they could just go to the office, just play on it, you know, just play on chat GPT all the time and not learn anything. But yeah, I mean, I guess maybe you could always say, well, this forces them at least to have to interact with other partners and whatnot and maybe pick up some things. I don't know. Yeah.

45:56Niki Black: I was trying to find, there was a study about, I think this one talks about a study about ego. I can't remember where I saw it, but this one talks about there's a correlation between ego and narcissism and return to office and our bosses that have big egos and are narcissistic. But, you know, I've read some other interesting things that talk about how at home the bosses don't get the adulation that they get. You know, they don't have these people kissing their asses everywhere they go. They just got a wife who's like, get the shit done. And so when they go into the office, it gives them this little fake world where they're at the top of it and they can make all their little minions run around.

46:34Niki Black: But it's not a lot of fun if you don't have the minions there. And this, I'm sure, applies to women as well, but not just men. But because of the structure of our law firms and offices, there's a lot more men at the top of these than the women. But I think that it's just interesting that there's a lot of that involved. It's not a lot of fun to be the boss when you're working from home and everyone else is working from home. So I think there's a lot of that playing into this. And the psychology is what interests me more than anything else, because clearly the when you look at the revenue generated and the expense of offices and everything else, the math isn't mathing.

47:16Niki Black: You know, it's not about that. It's about something else. And it's about the psychology of people that rise to the top of these firms and businesses and everything else that require these things. So I think that's the most important part. Yeah.

47:29Doug Austin: And the argument is always you've got to be in the office to be adequately trained. And there could be some truth to that. The problem is that most of these firms make younger lawyers come to the office just to come to the office. I mean, there's no formal training program. Just I hope you bump into the, you know, the super duper partner at the water cooler, or you get to work with a partner that, you know, happens to be a really good mentor. So you get really adequately trained and then somebody else happens to get a worse choice. it's not i mean it's the whole it's great to say come to the office to get trained better training but if you go to the office and there's no formal training then you're not really accomplished very much so i've not been a fan of all that come back to the office stuff for a long time and this

48:19Victor Li: doesn't help well it was all just about facetime there's something to be said for zoom i mean if if we were all like we're all legal tech writers if we all say worked at the same publication in the same office building, we probably talk to each other less than we do now. At least once a week, we get on this call and we talk to each other. If we were in an office, we'd be running around doing stuff, running out at the end of the day. I don't know. I'm not even sure we'd even see each other all that much. So I'm not really sure the office structure does anything.

48:48Bob Ambrogi: We won't have any idea how tall we are though. That's one of the challenges.

48:51Victor Li: That is a problem. I have been surprised by a few things. Steve is only three feet tall and I never knew that.

48:59Bob Ambrogi: Actually, no kidding. The first time I met Caroline, for whatever reason, I don't know why. I thought she was shorter than she was. Maybe because very often on these calls she's crouching a little bit. But when I first saw it, I was like, whoa, you're tall. I totally did not see that coming. I don't know why.

49:15Victor Li: Yeah. You didn't think the same of me, Victor?

49:17Niki Black: I'm a little insulted.

49:19Bob Ambrogi: I've met you before, though.

49:21Niki Black: Oh, all right. I haven't met Caroline in person. But you think I'm tall. Is that what you're saying?

49:26Bob Ambrogi: Yes. I think you are eight feet tall.

49:31Victor Li: Yeah, that happened to me actually. I'm on a community mediation. I'm on a board of directors of this community mediation group. And for like all through the pandemic, we never met. I hadn't never met most of the people I was on this board with. And then after the pandemic, we had this like our first in-person meeting. And it was this like tall guy standing next to me. And I'm like, I've never seen him before. And it turns out he's like this guy I talk to all the time, but he just looked so different in person. I thought he was short. Anyway, okay. All right. Well, we can now move on to another, what's becoming another, unfortunately, regular topic, but that of deep fakes in the courtroom.

50:13Victor Li: Steve.

50:14Doug Austin: Yeah. Actually, for those in the audience, this show is entirely a deep fake. I wasn't supposed to tell that, but, you know, in good conscience, I thought I would do it. No, this is an article that - You're deprogrammed, Steve. I will not be here next week. This is an article that originally appeared in Reuters, which is behind a paywall. So I got the copy out of Yahoo now that we sort of all slammed Yahoo. But it was a pretty scary topic. Apparently, a couple of representatives from OpenAI and Meta went before this U.S. Judicial Conference Advisory Committee on the Federal Rules of Evidence. And they weren't just like heon kind of people.

51:03Doug Austin: I mean, one of them was the policy council and the other one was a director of product management. And what they had to say was really pretty scary. I mean, they were saying that the ability to create deep fakes is already so far advanced that it's really difficult to tell what's real and what isn't and it's continuing to improve that the cost in time to create these deep fakes is substantially reduced that there's no science proven scientific method to to detect them and the systems that they do have to detect them really aren't all that reliable and and they went on to add and may never be all that reliable or completely reliable.

51:51Doug Austin: Now, the funny part about this is that committee has been in existence for, I don't know, three or four years. And the best I can tell, they haven't really done all that much. I mean, haven't produced all that much. And they tend to rely upon the survey. And the survey was reported in the article that was done earlier this year. and out of 930 some judges, only 15 of them reported encountering deep fakes. And I gave a presentation with Judge Rodriguez, who most of us know, at Tech Show, I think maybe a year before last, or maybe it was this past year, where we asked the people in the audience how many people had come across deep fakes and maybe one or two hands went up.

52:42Doug Austin: But it sort of reminded reminds me of you guys probably remember that the old thing that used to say about about data breaches in law firms. I mean, there there are those that have encountered deep fakes in their courtroom and there there there are those that don't know that they've encountered deep fakes in their courtroom. And I think there's a lot more of the latter than there are of the former. But if what these guys are saying is accurate, this is going to become a really significant problem. Not to mention, I mean, if there's no scientific method or no completely reliable method, then I can just see constant battles of the experts and judges placed in this horrible dilemma of trying to figure out what's real and what's not real.

53:31Doug Austin: And I mean, it's a mess. And I hope that this committee, you know, gets a little more serious about dealing with it other than this thing. Oh, it doesn't happen very much. So let's not worry about it, which is kind of sort of seems to be the attitude. Not that I follow it that much, but I haven't seen much, much substantive discussion of it. So I thought it was a it was a pretty scary article in a lot of a lot of ways.

53:59Niki Black: I think it's particularly concerning in terms of the impact on the criminal justice system, because there was just a case where an expert, not an expert, it was somebody was analyzing evidence in a lab. Oh, gosh, I can't remember what the evidence was. I think it was DNA evidence. Someone got convicted of it was a scientist who was just faking all the evidence and all these people got convicted and they had to reverse all the convictions. And so she was so before deep fakes, you had some people that were in the name of justice falsifying evidence because they thought these people did it. So they falsified the evidence.

54:37Victor Li: And now we had a big drug testing scandal like Massachusetts where this one person at the lab was like falsifying all the results.

54:44Niki Black: I think that's what it was. Yeah. And she just it was a woman and she just got convicted. did. Yeah, that's what it was. But, you know, so this was happening before deep fakes. And now, just the idea that the police could be manufacturing evidence against people that they know were guilty, you know, they know this guy did it, but they can't, you know, evidence got withheld or whatever. But then another situation that's slightly different, but there was, it was a rogue AI, like an agent, AI agent was assigned to do one task, and it went on the internet and did all these things. And one of the things that decided to do that had nothing to do with the task it was assigned was to put in a fake tip.

55:21Niki Black: It called a hotline for a murder case and put in a fake tip, you know, which apparently got screened as spam. So it wasn't actually considered. But there are just so many ways that AI is both impacting and potentially impacting the criminal justice system in the upcoming year so that it is alarming. And, you know, the flock cameras are another situation. That's AI facial recognition. They're going to start building facial recognition into them. Right now, it's just the license plates. But even those, they're falsely identifying people being in places where they weren't. So I think this is just going to exponentially increase, just like AI is exponentially advancing.

56:04Niki Black: So I really worry about the criminal justice system more than anything else in terms of how it's going to impact things. Because that's people's lives. It's not just money. depending on how you look at it some people think money is a lot more important than people's lives

56:15Doug Austin: but well and you know it's it's it's it's a it's a double double problem because you got a problem with saying okay we're going to ferret out the deep fakes but then you have the problem where everybody says i don't i i believe this guy's guilty and so you know the fact that you've got something that shows he wasn't at the scene of the crowd must be a must be a deep fake even though it is i mean it it just it it turns every piece of evidence on its head and puts lets the decision maker sort of assign themselves whether they're going to believe evidence because they think it's

56:53Bob Ambrogi: it's it's falsified or fake that's just a mess it's yeah the liar's dividend um yeah i just covered a case this uh this morning on on my kitchen sink uh where basically a guy claimed that you know, he had a recording of assaulting a woman and he was claiming it was a deepfake. And they used the circumstantial evidence of how could he reproduce the room exactly the way it looked? How could you, you know, all these different things to, you know, to to determine that to basically say it wasn't a deepfake, it was legitimate. But you're going to have those cases, too. And you're going to have a lot more probably deepfake, you know, lawyers, dividend, deepfake defense kind of cases where they're going to claim something's a deepfake for a while, then you actually have the deepfakes.

57:40Bob Ambrogi: But eventually, you're going to have some deepfakes that are going to be really hard to identify, and they're getting harder and harder.

57:46Doug Austin: It's sort of like saying you didn't sign a document that you signed or something like that. Right. Exactly. Yeah, exactly. Right.

57:53Bob Ambrogi: It must have been a deepfake. Yeah, I agree. That signature. And ChatGP2 told me so. So there you go.

58:02Doug Austin: I mean, you already see it just in, you know, in public life with, you know, political advertisements and videos. I mean, you see them and you accept them as real based upon what political persuasion that you are. Right. You know, so now, you know, you're going to have a juror saying, you know, their propensity would be to find this person guilty. So they're going to look for things that could, you know, support that and label things that don't support it as deep fakes. You know, our litigation system is already a frigging mess, too expensive and too time consuming. This could just overwhelm, be a tsunami wave that overwhelms it completely.

58:47Victor Li: all right well we are out of time and uh so that's gonna have to do it for today

58:53Doug Austin: good thing we are out of time because if we got into that p shit we'd be in trouble

59:03Victor Li: yeah i guess we don't we don't have to ask today it's over uh we could damien real uh proclaiming it, making it official. We're all screwed. That's it. As a matter of fact, we may not even make it to CleoCon where he'll be speaking.

59:19Bob Ambrogi: He did with optimism that we could, you know, that we could save humanity, but, you know. Yeah, we could, but we're not going to.

59:28Doug Austin: Why would we do that? No money in that. All right.

59:35Victor Li: Well, we will be back next week, I hope, and talk about then, everything that happens between now and then. And Doug, a big thanks to you for sitting in with us today. Always a pleasure to have you.

59:49Doug Austin: Thanks, Doug.

59:50Victor Li: And to everybody in the audience, thanks for being with us today. Until next week, see you then.

59:54Doug Austin: So long, guys. Bye, everybody.

59:56Niki Black: Bye now.