ICP Logo

S5 E20: Year of the GPTs


ICP Logo

S5 E20: Year of the GPTs




Harnessing the power of AI in the legal field has never been more vital, and ChatGPT-4 Turbo is leading the charge. Law firms are finding themselves at a pivotal juncture where adapting to technological advancements isn’t just beneficial; it’s essential for staying competitive.

In this week’s episode, Liel and Grace take a deep dive into how ChatGPT-4 Turbo is revolutionizing law firm operations. They discuss its remarkable ability to access and utilize local state regulations and legislation, a feature that proves invaluable for content creation and strategy formulation in legal practices.

The conversation also touches upon the multifaceted role of AI as a supplementary tool in the legal industry. From enhancing content creation to aiding in document reviews, image creation, and audio voiceovers, AI’s utility in modernizing law firm operations is underscored. This episode is a must-listen for law firms looking to leverage AI technology to boost their efficiency and maintain a competitive edge in an evolving legal landscape.

Resources mentioned in our episode:

If you enjoy the show, subscribe and leave us a review! Don’t forget to send us your questions and comments at ask@incamerapodcast.com.


Liel: [00:00:00] ChatGPT-4 Turbo is revolutionizing the way law firms handle information. Openai reports that this new version is not just faster, but it can seamlessly integrate the latest online data, like up-to-the-minute court decisions or legislative changes. This means law firms can make smarter decisions faster, staying ahead in a competitive field. I’m Liel Levy, co-founder of Nanato Media and author of Beyond Se Habla Español How Lawyers Win the Hispanic Market, and this is in Camera podcast, where we watch law firms turn into powerhouses, thanks to AI. Welcome to in-camera podcast private legal marketing conversations. Grace. Welcome back. How are you today?

Grace: [00:01:12] Good. How are you, Liel?

Liel: [00:01:13] Doing great. Grace. We are December 1st the day that we’re recording this episode. Here we are, another year going by. And especially for me, this quarter has been just nuts. Insane. Grace, I know it’s been for you. It was really nice to catch up with you in person about three weeks ago, that we’ve met very briefly in Arizona for the Business of Law Summit or conference by the National Trial Lawyers Association. So that was great. And ever since then, things have been good. Thanksgiving great has been good. How about you? How’d you been?

Grace: [00:01:44] I’ve been good. I mean, Thanksgiving, like you said, was amazing. And, you know, I go on my crazy baking streak, so it was definitely wonderful. Yeah. And it was great seeing you NTL Business of Law. That was a really cool conference. Yeah. So it was great to see you there.

Liel: [00:01:57] I liked it, yes. And I also really enjoyed hearing you talk about, well, moderate a panel about Irtc. And I think it’s a topic that at this point a lot of people have already heard and feel that have certain level of awareness. And I think what has, you know, as someone from the audience, just hearing and listening and being part of the conversations that people were having right after it, it was kind of like, interesting to see that. How many people thought that the boat has already sailed in terms of like being able to do something like it’s too late and realize that it hasn’t, right, that there is still room to do something. And you know, what is the sense of urgency of that? So I think that was much appreciated. Um, session that I think for some it’s going to mean thousands of dollars back in their pockets. Right?

Grace: [00:02:54] It certainly does. I mean, tens of thousands in most cases, because on average it is. And this is a conservative estimate because we’ve always been conservative with our estimates here, $12,000 per full time employee, right, as a credit to the firm. So it’s, you know, it’s a significant business credit that you’re due, right? I mean, I actually had somebody come up to me afterwards, and one of the first things they said was, so this is what it feels like to be a client, which is an interesting. Exactly. It’s true.

Liel: [00:03:24] That was probably one of the wildest parts of it. Of them all is kind of like, see, lawyers react to the fact that those who are primarily leading the efforts to help law firms recover this money from the IRS are working on contingency. And they are kind of it’s interesting to see how it doesn’t it doesn’t land on them right off the of the of the top, that they’re not necessarily having to hire someone and pay them upfront to do something. It’s all based on results. And, you know, it’s one of those things that it really is an offer that it’s too good to, to decline. But the problem is that it’s also too good to be true from the standpoint that or sorry, is it a scam. So I think that’s what, you know, even now, a year and a half or two years into this, people are still only starting to kind of like, oh, so I don’t really need to do anything, and the IRS is not going to come and retaliate against me because of doing this. And like all of those things where steal coming up now. And I must say that the room, you know, you’re not talking about your average unaware law firm owner. You’re talking about pretty bright and and active and up to date people. So it’s a good reminder. Look, my lesson, what I take out from this as a business owner, as someone who is on the marketing side of things, as someone who is in the messaging side of things as well, is how much repetition there must be, how much you have to go over and over and over your message and never assume that people have heard it enough and that they understand, because chances are that a lot of people still don’t get it, they still don’t get it, and then also try different ways of explaining things. And not necessarily that this is the case, but also keep the vocabulary as simple as possible, because that’s it is kind of the case.

Grace: [00:05:32] Do you think so this you’re 100, oh, 100% because these lawyers, they’re not tax attorneys. And that was one of the first things that they said is like, I wasn’t aware that you could do this on contingency. My accountant said my CPA said, yes, your accountant and your CPA, they’re going to charge you per the hour. That’s their model. They have to. Right? That’s how they make their money. As a lawyer or a law firm, we are allowed to charge on contingency if we choose, or flat fee when it comes to certain case types. And that includes tax law. So there was somebody on the panel that was a partner of a firm he charges, not contingent. That’s right. He charges a flat fee. Yeah. So you know they’re just not aware of tax law. And I think when somebody is extremely educated like these lawyers are and they’re not aware of the law on that side of the law, it is kind of disconcerting because you don’t know if it’s true. You don’t know what the truth is. You don’t know the law behind it, so you just don’t even touch it because you’re like, my CPA told me this, my accountant told me this.

Grace: [00:06:42] They’re the professionals. They’re the tax professionals. So they should know better, right? But here we have tax attorneys. That’s your version right? A PI attorney on the tax side. So you definitely want to go with somebody that you can talk to that you can trust. And so that’s why it was a really great presentation and panel. Because they had all sides. They were able to see the different aspects of it or see in layman’s terms, even for an attorney they need the layman’s terms. It’s requirement for anybody to understand anything. Like you said, you being a marketing, me being in marketing, we have to constantly explain to people why they should get involved in something, why the buy right? Why the bite? So I think that it definitely helped, but it was a really good reminder to me that people were telling me, oh, this is what it feels like to be a client. Oh, you can do contingency in tax call. I didn’t know you could do that. Yeah, it’s like this. These are lawyers asking me these questions that are simple questions that I would have even thought they knew the answers to, but they don’t. Yeah.

Liel: [00:07:45] Yeah. No, it was an all in great conference. You know, I think one of the nice things about the business of law is that it’s relatively small. And as I believe we’ve introduced this conference in the past, is kind of like that. Greatest Hits conference, where you get to hear a little bit of what’s been going on on all of these independent, well known, well established masterminds and law firm growth groups. And here you get to kind of like get a taste of all of them and then decide a little bit, know, like, I want to go this path, I want to go that path. Right? I really liked or feel gravitating towards the way that those things, the way that, you know, fireproof supports you and a bunch of others that were there. So I think it’s a really good place to just get started and talking about the Trial Lawyers Association. We have the TLS, sorry, trial lawyers summit. Right. Trial Lawyer Summit is coming now up next month in. Isn’t it wild to say next month? It’s still almost like two months, but still very soon in Miami. And that’s big. That’s big because that has a business of law track. But it also has a lot of trial insights.

Liel: [00:09:10] And it’s, you know, the biggest event from the Trial Lawyers Association. So it’s it’s definitely something worth keeping your eyes on if you are thinking of potentially going to some conferences next year. And Grace, I did not know that AJJ in February is coming to town. They’re they’re here. They’re in Austin. They’re going to be right here in my hometown. So, you know, I’ll guess I’ll have to swing by and, and and reconnect with good old AJJ, which I love. I think they’re an amazing cause. I just don’t necessarily think that they are the best venue for our agency to promote ourselves, or at least it hasn’t really been very impactful. But hey, you know, there are going to be here. So that’s that’s exciting. I’m really excited about it. Ever since we’ve opened our agency, I think this is the first time that we’re going to have one of these bigger, nationwide, um, conferences happening in our in our town. I mean, obviously, there’s always the, the, the local Texas trial lawyers and those things happening every year. But, you know, it’s nice to have one of those kind of big conferences, big conventions. So that’s nice. Hopefully you’ll be here too. Oh that would be so much fun.

Grace: [00:10:26] I wish I could I love going to Texas as a whole, but I haven’t. Austin is the one of the few I haven’t been to. Oh come on.

Liel: [00:10:32] You have to. You have to. You have.

Grace: [00:10:34] It’s so beautiful.

Liel: [00:10:35] You have to.

Grace: [00:10:36] I know I, I will do my best. I can’t promise anything. You know how hard it is for me to get out. Yeah. Be let out.

Liel: [00:10:42] Tell Michael Blum to sit this one out and come. Come to come to Texas. All right. Grace, look, it’s our last episode of the year. Come on. Let’s be honest. It’s been hard for us to keep up these episodes coming, but that doesn’t mean we don’t like doing them. We love it. And we really have a lot to say and to share and since the last time we spoke Grace, I think probably one of the most meaningful and important times in what’s going to go down as the significant moments in history, I believe, just happened. Right. And that is the enhancement to ChatGPT to its fourth turbo version, which really enables a lot of power in a very, very, very easy way and kind of like introduces. A new way of using ChatGPT that is way, way, way simpler. And the first thing I’m going to say about this before we jump into talking and getting all excited about how is it that we’re using it? Um, it’s such a good example of the domination that that that GPT now holds over AI, right? Because like. Google. You’ve probably heard that Google recently also got an update in that, you know, they’re pretty much on par in a lot of the things that ChatGPT can do, but that they they don’t get a fraction of the attention that. Anything happening on the Chat GPT platform gets. And that’s because they are pretty much established market dominance. Like why would you why would you? And the other thing, the other thing Grace here is kind of like. How having a very strong, very, very, very, very strong foundation in terms of. The capacity of your platform. Is gonna allow you. To really put out of business anyone who’s trying to build a. Business that is innerwear and another utilizing your model. Right. So I’m like okay so that was my thing. I want to hear what you have to say about that.

Grace: [00:13:22] Yeah. I mean, you know, you know, I’ve been dealing with it since it was like an infancy of open eye. I mean, even 20 years ago. Right. So, you know, you and I talk about chat and chat bots and everything. Chat and how this particular new language models and true AI. Right. Because this is true, I this is not chat bots. This is not no, this is us actually talking about artificial intelligence, synthetic language, custom language models, all that tech jargon that, you know, people start to glaze over about this. They they took over the market. They had the early adopters. And for people like me who wants to have access to things like open source information, open AI has been open source from day one. They have allowed developers, people who want to play with their stuff. And they were early to this whole concept of AI. And so they’ve just continued to take over the market. I actually just got a Google alert, not an alert, but a notification for their two three week conference. It’s going to be online, hands on, remote on AI. You and I both know they’ve got barred. Yeah, barred. Can’t touch the heels of what ChatGPT is doing. So to just to your question and your comment, I think that them taking over the market and being such an early model and really just all the developers and people like myself who are on the Reddit forums, we communicate with each other every day on a constant basis about things that are developing to the second. Google was too far behind and they’ve always been ahead. So it’s going to continue on this path and it’s going to be very difficult for anybody, just like you said, to use any other model to create their chats to create their AI bots, right? Because they’ve got the establishment, they’ve got the security, they’ve got all of these things in place. And for Google even to catch up, it’s going to be very difficult. Yeah.

Liel: [00:15:35] Yeah. I think just people, they don’t care, right. They don’t care. And as far as I know, bar doesn’t doesn’t have paid version. Right. And right now ChatGPT the paid version is on hold. You cannot get it. Even if you wanted it to get a paid subscription. You can’t. They’re just said we’re not going to allow new members to come up as we are basically running over capacity. And this is extremely powerful in terms of a strong message because, yes, you’re right. I mean, it’s actually very interesting because in light of all of these, you know, ousting and then reinstating of Sam Altman from OpenAI and a lot of things have come to light, right. And one of that is kind of like, well, wait a second, what was the origin of OpenAI? Why did it existed? Why did it got founded in the first place? What was their initial, you know, mission, vision, however you want to frame that? And I was really they kind of almost established themselves as a nonprofit looking to balance the power of AI. I mean, I’m using my own words to explain this, but my understanding is that they came up as a way of trying to level the access and use of AI so that it’s not all sitting in the power of Google, which at that point apparently had about two thirds of dominance of everything that has related to that was related in computing power and such for AI. And they were kind of like, we cannot it’s too dangerous to have a company monopolizing basically AI, right? So we need to be the, the, you know, the protectors of integrity and that. And so, so a lot of, a lot of the of the people that are, you know, the executives and such in, in AI and definitely in the board, they were loyal to that mission.

Liel: [00:17:35] But we all understand now that that mission is kind of like gone down the drain like, hello, $90 billion valuation by Microsoft. And, you know, very, very lucrative opportunity here. Now seems like they’re in the business for monetizing and generating shareholder value for the companies that are backing them up. And so it looks like that that is no longer a top of the top of the list on the agenda, disregarding where are we going with all of this, is that Google is no longer leading the race in AI. They are no longer the main player, and I think that is something that they themselves, you know, they acknowledge like in every. Every single earnings call in every single announcement that Google’s been making. They’re taking that role of like, you know, we’ve not given up. And you will see. And like we still have a lot to show. And yeah I mean sure you do. But the thing is like will people still care? Will people still be interested. So anyhow, Grace, that’s the, the, the the state of the corporate world in AI. Let’s talk about the fun stuff which is using it. Right. And the first thing I’d love for you to explain, Grace, for those who you know, have not really had an opportunity to have a hands on experience with using ChatGPT for turbo. What is it that primarily has changed in ChatGPT for turbo? And then if you can explain us also, what’s this whole thing about being able to create your own GPT?

Grace: [00:19:20] So the main thing that changed with ChatGPT is users can basically create their own version of it, right? So currently when you use ChatGPT, you basically have to tell it what you’re looking for and prompt it. These are called prompts. Prompts are basically directions like if I would tell you okay, turn right here, turn left there. You have to go 60mph because that’s the speed limit. It’s very similar to that kind of directions that you would tell ChatGPT this is what I’m looking for. This is what I want you to do. And this is what I want you to spit out. Now with the GPT four turbo, you can feed up to what they’re claiming up to 300 book pages of text as context. Yeah, that’s around the insane. It’s insane.

Liel: [00:20:12] It is insane. So yeah, that is one of the of of the biggest increase in capabilities that it has. One of the much awaited updates has been the ability of actually creating images inside of ChatGPT, and not necessarily having to rely on another third party or a plugin or something else, but rather just kind of like a functionality that is natively embedded into ChatGPT. And that is done through Dall-E three, and that is now something that you can do. You can go into ChatGPT and say, hey, ChatGPT, we’re going to be doing a community event where we’re going to be doing a turkey giveaway. We want to call it Thanking our Partners 2023 and sponsored or hosted by the Smith Law Firm. And we’d like to have a turkey and a gobble. Right. And can you create it and 30s in? You’ll have it. I’ll come up with something that it may or may not look exactly like you wanted it to look, but then you can kind of like, fine tune it a little bit with, with some direction, and you’re actually going to have a graphic for your event. It’s a pretty, pretty cool, pretty, pretty cool. And there it’s a good example of how you can use this, you know, new powerful capability that it has. And I like to go back to what you’re saying about the ability of configure your own ChatGPT kind of like creating your your own GPT is creating your own chat bots that already have pre configured data points to lean on so that they can give you better or more fine tuned answers or results to what you need.

Liel: [00:22:04] And that is because now you know, before, one of the most painful things about having to deal with ChatGPT is that you had to make very, very thorough prompts. But here is a problem. If your prompts were too long, the system wouldn’t read them or it would confuse them. It will completely throw them off. It wouldn’t understand what is it that you want, and it would start contradicting some of the things that you may be asking, right? So you may say only use three times this word, and then another 25 guidelines and then that word that you said only used three times. Then probably it’s going to be used three times in every sentence because it will get confused with something else that is further down the, the, the instruct the prompt. And it really threw them. It threw it off. And so that obviously made it very hard to use it for some more sophisticated purposes. Right. Or for things that have an even greater detail in, in, in some aspects. Now, with the ability of creating your own GPT, you can actually pre-populate all of this information to the extent, as you said, like you can give 100 pages worth of context to it. And on top of that, you can upload to the actual GPT configuration page up to, I believe, 10 or 12 different documents. This could be PDFs.

Liel: [00:23:31] This could be images that can further help the bot to understand how you’d like to give things. This could be, for instance, right? For instance, if you’d like to create for your law firm. But that can create. Any type of communication using the tone and voice of the firm. You can upload their your editorial guidelines, right. And you can upload their your brand kit, and you can upload their examples of previous campaigns that you have run or of letters that you’ve sent to your clients. And then you can say in your kind of like master prompt, the one that it’s on the main dashboard of configuration. Use all of the information added here in this here in your database to create documents that are serving the purpose of communicating to clients or potential clients about notifications, activities and such. And they have to it has to be done in the same style and voice and all of these great things. Right. And then the beauty about this is because you’ve already fed to the GPT all of the information that it will need to perfectly understand how this paragraph should start. What’s the signature should be? What are the traditional calls to actions that you’re using? All of that is already available to the bot because you’ve fed it to them, to it through these documents that you’ve uploaded, through the examples that you’ve uploaded. And at the same time, you could further fine tune how to use those documents in that master prompt that you put there, and that what that allows you to do is then you have a bot, and here is the beauty.

Liel: [00:25:22] You can then share that bot via link with anyone, right? You can share with anyone, any one of your team members. Now they don’t have to worry about having a very long or building a very robust prompt so that they can actually get the same end result. You’ve already created a baseline model that’s going to yield the same result to every single one of your team members. And now the only thing that you need to do is click this link. It will open up the bot that it’s been pre-trained on your brand voice. And then with a very simple prompt which you can also create, it’s a like call it like a micro template that kind of like is going to specify who is this letter going to and what is the main topic. And so you right there in a line this is for Michael Smith. And the topic is we have your disbursement check ready. Come and pick it up. Right. And then third line is there anything else that I should know about this. No there isn’t. And then just hit create and there you go. You’re going to get something that pretty much looks, feels and reads like what you’ve been creating manually or through very outdated or impersonal templates. And it’s just going to work beautifully. Right? So it’s really really A) that easy and B) that powerful.

Grace: [00:26:51] So you know we’ve we’ve been paying for the resource allocation before it got closed out. So we have ChatGPT and we’ve had turbo for a little bit the beta version of turbo. So we’ve been able to create our own chat bots for quite a few months now with our own prompts, back end and information where the change for us has been significant. It’s in the increase in the amount of knowledge that you’re able to feed it, and the reduction in the hallucinations, like you were saying, right? That that for us is the biggest difference. Now, the other significant difference, and for us, what has made a change with this turbo is the fact that the information cut off was previously really old. That’s right. Like September I think you.

Liel: [00:27:36] Said it was. Yeah, it was 2020 1st September 2020. It was ridiculous.

Grace: [00:27:40] It’s now. Yeah, it was a long time ago. It was old. So now it’s right.

Liel: [00:27:46] And and it can actually search live on the internet. But I’ll be honest with you. Yes. Yeah. I’ll be honest with you. It hasn’t worked really that great for me. I like when it tries to pull off information live from the internet. It just doesn’t. It’s just not not very reliable. But but but yes, what you’re saying about being up to date until April of 2023 allows for it to, you know, by its native database, give you so much more valuable and up to date information on almost anything. And, and and just think about it from, I mean, from a legal standpoint, the fact that it can actually pull up the local state regulations, legislation, whatever that is, it’s extremely powerful, extremely, extremely powerful. Because now you can reference to that. You can reference to that. Like for instance, if you’re going to be using it for content creation, you can actually say that in the way that you are explaining your problem resolution for whatever thing it is that you’re writing about, you need to keep in mind the latest local legislation, right, of the state of this, and it’ll probably going to do a much better job than it was doing before. Because, you know, we know that there’s changes, right. Like this is a thing that evolves. So it’s great. Yeah. No, that’s a really, really good point, Grace. Really, really good point.

Grace: [00:29:25] That’s where it’s made the difference for us because, you know, we’ve been working on this project for just I’d say maybe for about a year, really playing with it for six the last six months, I’d say, you know, as things have expanded, gotten easier to use, and just getting involved on the, the customizing part of it. Right. Because right now it’s it’s kind of it’s open to a lot of people. It’s open to a lot of things. And as a law firm, we have to worry more about security and compliance than anybody else. Right. So, you know, we’ve been messing with the security side of it and the custom customization mostly. And so with this difference or this change and this update for us, it’s been a lot easier to get information from the web as a whole. But we we have been feeding our own database of information from the beginning because we have our own custom model, right in a secure Azure server with a, you know, non reachable. Yeah. All of that risk management stuff that, you know, I’m always talking about because it drives me insane. And you know I have to be compliant. So you know I don’t usually like to talk about it. And this is a component that I it’s not that I don’t like to talk about it. It’s just I know a lot of people don’t really think about it, but since we speak to lawyers, a lot of times I have been asked this question over and over again about AI.

Grace: [00:30:47] And that is, you know, what kind of security do they have in place and what are the regulations? Right? Because we have Can-Spam act when it comes to emails. We’ve got to pee on text messaging. So what are the regulations they’re supposed to look like for a call from an AI bot? Well, I mean, you know, I personally, having followed this for so long and followed compliance for as long as I have, I can read into the regulations and understand that to be compliant, you have to let people know and give them the option that this is a bot, that this is not a human being, and you can’t pretend that you’re something that you’re not. As a law firm, we’ve taken the tactic that we giving people the option to communicate with you in the way that they want to. And this is, you know, 30 plus years that Ed’s been in business. That’s how he communicates. If you want a text, you get a text, you want an email, you get an email, you want a call, we’ll give you a call. We’ll do all three if that’s how you want to communicate with us. Now with AI, that adds a fourth communication option for people that don’t necessarily want to talk on the phone, they might not want to. Right? So our concept is if you give them the opportunity and the option to communicate with you how they want to, including an AI bot that could call them, give them a status update on their case, or call them and take them through an intake as an example, which, you know, that’s kind of what we’re focusing on is on the intake side, at least to start and let them know that this is a bot calling them.

Grace: [00:32:24] But if they ever want to speak to a live person at any point in time, press one. You know what I mean or say I want to speak to a live person. That is where it’s for us, has been a game changer and a significant user experience because it’s all about the client, right? If you don’t have clients, you don’t have business. And so the more you can take care of your client and be there, be there in the method that they want to communicate with you, you’re serving them and you’re serving yourself. So for us, the compliance situation and the concept of AI as a whole, we’ve embedded it as part of our communication methods instead of creating it as an outlier. Because AI is a tool, it’s not a it shouldn’t be a complete replacement really to everything that you do, but it’s a tool and it should supplement everything you’re doing. And that’s what we use it for. We use it in content creation. We use it in document reviews. We use it, I mean, in image creation, in audio voiceovers. We use it in a lot of different aspects because we have it embedded as part of our processes.

Liel: [00:33:31] Yeah. In the firm. No. I think it’s a very interesting angle that you’re taking up in there because obviously, you know, you know, maybe for us, we are a generation that we’ve grown used to, you know, personal way of communicating. And maybe for a lot of us it would be like, I wouldn’t want to talk to a bot ever in my life. But you need to think that, you know, in in ten years. Eight year olds now are going to be 18 and they’re going to be reaching out to lawyers, and they probably are not going to want to speak to a live agent because they want to deal with a human being in the pod and the training, which we’re going of isolation and such, which is very, very sad. But at the end of the day, it’s something that people may want to reserve. There’s. Interactions for people that really care for right and not for transactional things. I think that’s been the trend that we’ve been following for the past few decades. And again, I’m not I’m not saying that this is good. I’m not I’m not an advocate for this. But the bottom line is that consumer behavior is going to determine the way that that brands, that businesses are going to have to make themselves available to have those conversations with their customer base. Now, that’s that’s that’s great. Grace, I really like the angle that you’re taking. Austin’s there. Now let’s let’s bring this down to, you know, a more kind of like your average business owner user version of this. Let’s let’s wrap up this podcast instead of doing takeaways. Why don’t we give each a couple of ideas of how they could best use an revamped version of ChatGPT so that the year of 2024 can just be easier in whichever way they want to make it easier. So do you have something already ready, Grace? All right, so go ahead and start.

Grace: [00:35:29] I sure do. So you know, it’s funny because I think about this and we have changed our marketing department a little bit recently. And that includes because of what’s available now. Right. In terms of tools now we use, you know, a variety of platforms. But Hootsuite has kind of been one of the easier ones to publish multiple places. Right. And they’re very good. So we use Hootsuite. Hootsuite has AI embedded. So I would suggest for those that are trying to get into it, aren’t familiar with it, aren’t used to it, to look for a piece of software that has AI embedded and I’d say a good 80 to 90 even percent, I’d go as high as most of these. Social media and content creation platforms that are software have it now embedded. If they didn’t before, they do today, and Hootsuite is one of them. So, you know, use a platform that has AI and start using it because you will not know how good a tool it can be for you and your use, particularly with this new update, right, with the customization. And just like Liel said, and I want you to listen to this part, you can now upload your brand kits, you can upload your tone, you can upload old blog posts. Right? So upload all the stuff that you’ve already created to date, even if it’s 1 or 2 things, because it’s going to tell the new ChatGPT it’s going to help you create other things, new individualized content that you’re still going to look at, obviously, and you’re still going to review, but now it’s going to be done so much faster, more productively, more efficiently for you, your team, and everybody else.

Grace: [00:37:11] But if you’re not familiar with it and you’re uncomfortable with using new pieces of software, get one that has it embedded, like Hootsuite, because it just makes it that much easier for you to start using it. You need to. I mean, it’s a tool for a reason. It’s I don’t think it’s ever going to take over, honestly. Liel, between you and I like completely maybe at some point, you know, maybe in 20 years from now it might take over completely in terms of original content creation, because it has access to all knowledge around the world. Maybe. But for now, it’s going to be a tool that you’re going to use that you still have to review. And as a law firm or a lawyer or writing content for legal, you absolutely need to review it for regulations and compliance and verbiage that’s allowed versus not allowed.

Liel: [00:37:59] Yeah. And a good one’s really, really good one’s Grace. You know I’m going to I’m going to say yes I mean absolutely a lot of tools. What are the ones you’re using or not using in your in your tech stack are starting to. Are AI powered tools that in most cases come at an additional cost, right? If you want like kind of like you have your membership and sometimes they say, well, we now have the ability of being able to have an AI assistant that will do this, that, that and that for you. And sometimes it’s very much worth it. Okay. It is. It’s 100% worth it. Now, if that is what’s stopping you from being able to use the AI? Because, as Grace said, not everyone has the comfort levels of going into a ChatGPT that even though it’s nowadays very, very user friendly, it still it still requires for you to have certain ability of think, organize structure prompts, gather documentation. I mean, you know, it takes some effort, right? So granted, it’s not all fully automated for you from scratch. You need to build it. Okay. But it’s easier to build than it was before to code. Now what I’m going to say is yes, 100% use AI to just be able to ease the process in which you create something that is part of your operations and that, you know, for me, it’s always easier to think about marketing things. So I’m just going to go ahead and give an example of a marketing thing.

Liel: [00:39:28] If you’re sending weekly newsletters, okay, weekly newsletters, and they’re traditionally always had a structure to it, right? You have a section where you talk about legal stuff, you have a section where you talk about community stuff. You have a section where you have your recipe of the week. You have a section where you have your favorite neighbor of the week, whatever that is. Right? This structure exists, and you probably already have at least a few or a dozen or three dozen or 50 of these newsletters that have been created in the past. So what you want to do with that? Well, you want to start creating a PDF document that actually has a lot of these examples put together. You want to start looking at how is this newsletter actually structured and tried creating a prompt that describes exactly each one of the sections. Okay. So we always start with a headline. The headline usually has this many characters, and that’s the first thing. And it needs to address in a slightly playful way, the topic of the week. Right? Then you want to have a subtitle, usually these many characters in there. And that kind of like gives a little bit more of information on that. Then we usually write one paragraph that is an intro, and then jump right into 2 or 3 takeaways as to what are the main insights, and then the actionable items that we recommend people to do, which are usually three and four.

Liel: [00:40:46] We never do them more than two lines and whatnot. Right. And now you’re really giving everything that ChatGPT needs to truly understand. Okay, so when I’m creating your next newsletter, I know exactly what you want to show. I already have a very, very good idea of what is it that you’d like us to like, in what tone and in what style you want us to write? Because you’ve given us a lot of examples, and we can see the common themes and the common aspects that are connected between all of these newsletters. And you’ve also maybe given us an editorial guideline if you have that. Not a lot of firms have it, but if you have that, you can have it there. And you know, what’s so wonderful about these is that then this thing that took you probably half a day to create because let’s face it, it’s not going to be put together in a matter of minutes. Oftentimes you can maybe do a GPT in less time. Right. But something that is, you know, as it has to be as effective as this, you want to dedicate some time to it. And here is the beauty. Then you need to make to work on your micro prompt. So you have your master plan, you have your GPT trained, everything is in there. Next thing you need to do is okay. Now I need to create the mini prompt that every single week I’m going to have to fill out so that I give enough information to the bot so that it can generate the full blown newsletter.

Liel: [00:42:08] And so that mini GPT is going to be what’s the topic? We’re going to talk about new traffic laws here. Okay. Do you have a resource to from where I can pull information specifically on this topic? Yes. Here there’s an article that was in the news. Look at this link. Then who is the neighbor of the of the week? This one. What’s the recipe of the week? Here’s the link of the one that we want to share. And you hit enter and you’re done. And you’re done. The the eight hours that it used to take, the one person in your team to pull together this newsletter. It’s now done. It’s now done. And here is the best part. You never had a newsletter because you never had someone that could spend eight hours doing this. Well, now you don’t have to because you can do this and it could actually be of great quality, great, great quality without having to have someone invest a lot of hours in there. You just need to get this ready, create it, build a template and such. Now, what’s the painful part about GPT steel is that it’s. Feel very much based on copy pasting, right? It’s created it. I need to copy paste it and such. Now, what we’re not talking about here, Grace, and we’re probably going to leave this for next conversation or sometime in the near future.

Liel: [00:43:23] Once we once we also better acquainted with this is is the amazing integration that now GPT has through many other platforms through Zapier. Right. You can actually trigger from prompts into actions happening somewhere else through Zapier, right? So you can put up a prompt, you can actually create that, an action that as part of the result of the prompt, some of the information that the prompt itself generated is going is getting inserted somewhere else. You don’t have to manually go and move things around. And this is very, very, very effective okay. Very, very, very effective okay. Just, you know, to give you an you’re probably have other examples of how this could potentially be used. Grace I’m just out of the top of my mind like one that I can, you know, connect. Let’s say that you’re having you’re using Call rail to track your calls, your leads. You’re using conversational intelligence to track and transcribe all of these calls. And so you can see that person that has called recently mentioned during the call that they were in a car accident. The call gets tagged with tagged car accident and injury. Those two words were mentioned in the conversation by the person who actually called, not by the person who’s asking the questions, because you can actually monitor and trigger actions based on that. And then that information gets automatically through Zapier flagged to ChatGPT, who then goes and crafts a personalized text message for the person to say, you know, we’re sorry that you were in a car accident.

Liel: [00:45:07] And thank you so much for calling us and we are going to be here for you every step of the way, right? That’s what you can do. Nobody has to lift a finger. Nobody had to go and qualify the call. Yes or no? Is it a car accident or not? Everything happened in real time as it got spoken. And I think that’s kind of like very, very, very, very cool. I don’t know, I mean, we’ll see. And I’m, I’m still thinking here in traditional ways. I mean, this could be far greater. This could be your website getting automatically optimized based on the data that it’s actually gathering in real time. I mean, this is really, really, really amazing stuff. So it’s exciting. Grace. And, you know, my final thought for this podcast in which we are wrapping our our 2023, is that, I think, you know, 2024, we usually do a predictions like what’s going to happen and such. And you know, it’s always the same thing. Grace. It’s social media and you need to create content and have great client experience like that never goes away. That’s always there. But what I do think now, like it’s there and it’s going to completely create two groups, those who those who are going to prevail and there are going to be able to just outgrow anyone that is in their competitor set and those who are going to be left behind.

Liel: [00:46:41] And and that’s really going to be those who are capable of leveraging artificial intelligence to. Just put themselves on steroids and just be way faster, more efficient and more powerful in what they’re doing. And and I think that is that is that’s what every single law firm, business owner, individual trying to build up their career should be thinking and doing now is how can we do and be more effective and do better using artificial intelligence. And and the best advice that I can give to anyone that doesn’t know where to start is create a membership. Bombard on ChatGPT whichever one you want, and start playing around. Really, literally start playing around. Playing around as in like. Try to break it. Try to ask things that are, in your opinion, ridiculous. And then eventually ideas are going to start coming up and eventually you’re going to start seeing, wait a second, if it can do this, could it actually do this? But for my business. And that’s going to be just, you know, potentially revolutionary. Don’t be afraid. This is not a social media platform. You’re not connecting. You’re not expecting for anyone’s acceptance. You’re not looking for likes. You’re not being judged by others as to whether you’re doing it good or bad. This is just you and a bot trying to get the most out of their capabilities. Grace, what are your final thoughts?

Grace: [00:48:16] I want to I want to add to what you said because it’s such a good point. Right? And it’s something that I always tell people, don’t be afraid of software. Don’t be afraid of tools. Log in, sign up and play with it. I cannot emphasize the word play enough, because what Liel is saying here, ladies and gentlemen, you really when you play with something and you have fun with it, you can understand it so much quicker and so much better because it is not. I mean, I know I’m restating what you said, Liel, but I don’t think people are. I want them to understand that this is not a social media platform. Developers like things very introverted. Most people that are coders don’t want to interact with other people. This is the opportunity that you have to play with a system that is not going to talk back to you, that isn’t going to tell you that you’re doing something wrong. It’s just going to tell you and give you information that you’re asking it to give you. And you can ask it information that you already know. So you can see if it’s giving you information that you might want to know. Right. So that’s why Liel is saying play with it. Because really, that’s what you need to do. Log in, sign up, play with it. That’s it. That’s all we’re telling you to do. And don’t be afraid of new pieces of software.

Grace: [00:49:32] If I can’t say that enough, I will continue to say that because new pieces of software and new things out there are here to help you. There’s always trial versions of things. In this case, it doesn’t cost you to do it. So pick the one that you want barred or open. And you know, obviously after what we’re saying, it seems like ChatGPT is the one to go for because it’s the most viable at the moment. That is technically the easiest for people to use, but it doesn’t matter. Use barred, you know, play with barred first because you can just ask it in right directly in the barred platform without having to log in for a subscription for ChatGPT. Go ahead and do that and you’ll still get the same familiarity, same kind of concepts of prompts and how chat and AI works. So just get into it. Start doing it because there’s so many ideas out there. We’re using it for so many things right now. I couldn’t tell you it’s embedded in like 90% of our platforms and it’s native to Zoho CRM already. We have voice of the customer, clouds, Liel that I get every once a week, and it tells me in red tagged words that the customers are saying on calls that could be a cause for issues, right? Like if they’re saying negative terms, positive terms, sentiment analysis. Yeah, it’s doing it for us.

Liel: [00:50:56] And that part of the technology has been it’s been around for a while now. The beautiful the beauty of this now is that you can actually do things with that data without necessarily having to spend a lot of time analyzing it. Right? I mean, a lot of a lot of, you know, a lot of the things that we’ve one of the challenges that we’ve in, in, in the time that this podcast been going on that we’ve actually talked about is like, yeah, yeah, yeah, I mean, it’s great. Like we have so much data, we all have so much data. It’s so easy to gather data, but it’s so hard to actually do something with it, because analyzing data is not something that is necessarily easy, and especially not when you have thousands of data points. So this is the beauty that ChatGPT has. I mean, we did not really mention this, but you can go in and throw, you know, drop a spreadsheet with a lot of information that you wouldn’t know how to get Excel to filter out. If you’re not very proficient at Excel, and you wouldn’t be able to to to get it to tell you, but you drop it out in chat GPT and you say, hey, can you please let me know which ones are the highest rated points in according to this report, which ones are the ones scoring the lowest, and what is the delta between them? And I mean, you can do a lot of that without necessarily having to know how to type a formula in Excel.

Liel: [00:52:14] And that is that is the beauty about this. You just need to speak out what you want. That’s literally what you have to do. You have to speak out what you want. You can dictate in words to ChatGPT what you want and ChatGPT will do it for you. And if you spend a little bit of time training ChatGPT to understand you better, you’re just going to get even. Better quality result answers. So yeah, it’s very very very cool. Grace. So yeah 2024, the year of the GPT is very exciting for me. Wrapping up the year with such a great outlook. Thank you so much for a wonderful 2023 full of great insights, knowledge, fun conversations. I’d like to believe and I’ll see you in 2024. Grace.

Grace: [00:52:59] Thank you Liel. It’s been fun this year.

Liel: [00:53:01] All right. Take care. Bye. If you like our show, make sure you subscribe. Tell your co-workers. Leave us a review and send us your questions at: ask@incamerapodcast.com. We’ll see you next week.

Leave a Reply

Your email address will not be published. Required fields are marked *

ICP Comments

Leave a Reply

Your email address will not be published. Required fields are marked *