Founders and Futurists

In this episode of Founders and Futurists, host Josh interviews Ben Haber from Monark Markets. They explore the future of private markets, the challenges in democratizing them, and Monark Markets' unique B2B approach to providing access for individual investors. Ben shares his journey and advice for aspiring entrepreneurs, highlighting the importance of solving meaningful problems and building trust.

00:00 Introduction to Founders and Futurists

00:21 Ben Haber's Journey to Monarch Markets

02:25 Understanding Private Markets

04:39 Challenges and Opportunities in Private Markets

06:54 Monark Markets' Unique Approach

10:15 Market Research and Customer Discovery

11:38 Comparing Monark to Other Platforms

This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit workingitwithai.substack.com

Transforming AI and Backend Architecture: A Conversation with Ravi Ramachandran and Dave Brewster In this episode of WorkAI, host Joshua Schoen is joined by Ravi Ramachandran and Dave Brewster to discuss the evolution of backend architecture with AI. They delve into the concept of agents as atomic units in a microservice framework and the impetus behind their current project. Ravi and Dave share their personal journeys and insights on the collaboration and challenges in deploying AI-driven microservices in the enterprise. They explore the practical applications of large language models (LLMs), the importance of timing, human interaction in AI workflows, and the innovative features of their framework, Eidolon. The conversation also touches on future outlooks for AI and backend architecture development.

00:00 Introduction to WorkAI and Guests

00:51 Founders' Background

01:47 Challenges and Opportunities in AI

04:18 Building Effective AI Agents 09:40 Real-World Applications and Use Cases 16:00 Future of AI and Enterprise Integration

31:25 Eidolon Framework and Deployment

38:38 Closing Remarks

This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit workingitwithai.substack.com

AI and Legal Tech: Insights from Leo Park In this episode of Work AI, host Joshua Schoen interviews Leo Park, a seasoned professional in software development, engineering, and legal tech, with a special focus on Lex Machina.

We discuss the iterative process of extracting and classifying legal data, the advantages of large language models (LLMs) in legal applications, and the importance of using refined queries and rag-based systems. The conversation delves into the current and future challenges of using LLMs in legal research and how small language models could offer efficient solutions.

00:00 Introduction to Claude's Internal Thinking Process

00:40 Meet Leo Park: A Multifaceted Professional

02:25 Journey into Legal Tech with Lex Machina

03:55 Challenges and Solutions in Legal Data Extraction

09:03 The Role of LLMs in Legal Applications

12:25 Critique of Stanford's RAG-Based System

23:59 User Interaction and UI Design in LLMs

26:03 Future of LLMs in Legal Tech

34:23 Experimenting with Fine-Tuning Models

35:18 Conclusion and Final Thoughts

This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit workingitwithai.substack.com

Reducing LLM Hallucinations with Retrieval Augmented Generation: A Conversation with Ofer Mendelevitch

In this episode, Joshua Schoen of Work AI discusses the complexities and innovations in the field of Retrieval Augmented Generation (RAG) with Ofer Mendelevitch from Vectara.

They delve into topics such as the importance of accurate data feeding to Large Language Models (LLMs) to reduce hallucinations, the capabilities of vector databases, and how Vectara provides a comprehensive RAG-as-a-Service solution. Offer also shares his journey from working with early GPT models to co-founding Vectara. They explore various use cases for Vectara's technology in legal, healthcare, and other sectors, and end with a demo showcasing the practical applications of RAG in legal document management. 00:00 Introduction to AI and Hallucinations

01:29 Offer's Journey with GPT and

Syntegra 02:42

Joining Vectara and Its Mission 03:06

Vectara's Growth and Funding 03:31 What is Vectara? 06:54 RAG vs. Fine-Tuning

09:30 Reducing Hallucinations with RAG

12:39 Building a RAG Pipeline with Vectara

14:45 Customer Use Cases and Future of LLMs

18:51 Live Demo of Vectara's Capabilities

22:00 Conclusion and Final Thoughts

This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit workingitwithai.substack.com

In this episode of WorkAI, host Joshua Schoen interviews Chris and Nicole Hedley, the founders of Headstart, an AI-powered services company. Chris and Nicole discuss the significance of structuring natural language prompts, the role of AI consulting, and the future of AI in business. They explore the practical use of AI to generate significant revenue, reduce operating expenses, and enhance customer delight.

The conversation covers the benefits of smaller, agile AI service firms over traditional large consulting companies and includes real-world examples of how Headstart delivers high ROI through swift AI integrations.

00:00 Introduction to WorkAI and Headstart

00:45 The Role of AI in Modern Business

01:37 Challenges and Solutions in AI Adoption

04:54 Nicole's Journey and AI Integration

10:54 Chris's Perspective and Career Shift

12:35 Future of AI in Engineering and Business

18:11 Client Success Stories and ROI

This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit workingitwithai.substack.com

1. 🚀 Dive into the future of AI with Mark Lemley, Stanford Law Professor! Discover his insights on copyright, fair use, and the legal landscape of generative AI. Don't miss it! 🎓🎙️ #AI #Law #Technology

Mark Lemley is the William H. Newcomb Professor of Law at Stanford Law School and Director of the Stanford Program in Law, Science, and Technology. Mark Lemley is a distinguished authority in intellectual property, antitrust, and related areas of law.

Mark Lemley's credentials speak for themselves. He has shaped legal thought in multiple domains as a prolific author of 11 books and 218 articles. Among his accolades, he has nine of the most-cited law review articles of the last 20 years and was recognized as the third most-cited legal scholar globally from 2016 to 2020. His extensive experience includes active participation in over three dozen Supreme Court cases and a diverse client base, ranging from tech giants like Genentech to unique groups like Dykes on Bikes and generative AI companies.

We kick off with a fundamental question: What constitutes copyright infringement?

Mark explains that the basics of copyright revolve around the ease of obtaining protection and the critical act of copying. He then addresses how Generative AI presents novel legal conundrums, particularly regarding the legality of training on vast datasets of copyrighted works. Lemley elucidates that these datasets are essential for teaching AI systems about language and concepts, often involving temporary copying.

This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit workingitwithai.substack.com

This week, we interview Horace Wu, CEO of Synthiea, a Contract Analytics platform that helps large law firms make sense of their data and improve their work. Horace is one of the top applied legal technologists, and we explore what GenAI can do and what it can’t.

Thanks for listening to Venturing in Work AI! Subscribe for free to receive new posts and content at the forefront of AI revolution.

Horace's Transition from Law to Tech [00:01:00]

Horace Wu shares his journey from transactional lawyer to founder of a technology company. He describes his experiences with his initial consumer tech venture and the challenges faced when major companies like Facebook and Google launched similar features.

Discovery of Language Models [00:04:00]

- Wu talks about his discovery of language models in 2017 and his transition to legal tech. He elaborates on how he got involved with a German NLP company and subsequently founded a legal tech firm.

Synthia's Unique Approach [00:06:00]

- Horace outlines what makes Synthia unique in the legal tech space, focusing on boosting a lawyer's knowledge through past documented experiences and creating a reliable database of provisions.

Detailed Workflow and Data Utilization [00:09:00]

- Wu explains Synthia’s complex data handling through a four-tiered pyramid model consisting of expertise, matters, documents, and provisions. He details how data is structured and utilized across these layers.

Enhancing Lawyer's Efficiency and Knowledge [00:12:00]

- The discussion focuses on improving lawyers' efficiency by structuring and querying information. Wu illustrates how Cynthia manages to make data actionable and relevant.

Preparing for Legal Tech Innovations [00:17:00]

- Wu advises on what law firms need to consider technologically and commercially to prepare for implementing AI and data-driven technologies.

Legal Tech and AI Landscape [00:20:00]

- In closing, Horace provides his insights on the current and future state of legal tech and AI, focusing on the application of generative AI and its impacts on the legal industry.

t.

This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit workingitwithai.substack.com

While it is a little outside my traditional purview, I was excited to speak with Maggie Sin, the CEO of MyCompanions.ai, a creator platform that allows creators to communicate with their fans in their voice using AI. Given the recent news with Scarlett Johansson and OpenAI and Wired’s recent piece on the people who drive the creator economy, I was excited to speak with one of the pioneers in the ethical use of AI with creators and how users currently interact with them.

Thanks for checking out Venturing in Work AI’s Podcast! Subscribe for free to receive new posts and support my work.

The transcript is below:

Joshua Schoen: [00:00:00] Hi, I'm Joshua Schoen and I'm here with Maggie Sin who is the CEO and founder of my companion's ai. An AI enabled creator economy company makes tools for creators to automate their businesses and help them maximize engagement in monetization with their fans. Really excited to have you here, Maggie.

to discuss the future of work, what you're doing and what this means for our world and economy and everything in between.

Maggie Sin: Thank you. I'm so happy to be here. Thanks for having me as well. Super excited.

Joshua Schoen: Yeah. I'd love to learn more about kind of your story, your narrative, how you got here and what made you come up with Mycompanions.

Maggie Sin: Yeah, all so I personally have been a social media influencer for about 10 years now. I took breaks here and there. And I first hand saw how hard it is to engage with my [00:01:00] fans. So I get a lot of inbox requests, a lot of messages and comments, and I just don't have the capacity to go through each and one of them.

And I would like sometimes find pretty good messages like a year later or so if when I finally like support through it if I ever do and so I realized there's definitely something we can do about this. And I was talking to my co founders one who is like an AI genius with a PhD from Cambridge and another one who is an investor entrepreneur.

And we decided to start Mycompanions together to target these issues that these creators can face. So we launched our beta for our first product in November last year. And which is a telegram bot that uses AI to mimic the influencer's personality and can communicate with the fans 24/7 via text the influencer's voice and photos and videos.

And that has really helped creators engage with their fans automatically with less effort and also keep the fans happy as they're able to like finally talk to this person who they've followed and they want to be able to learn more about this [00:02:00] person and get precise messages. So we have over 20 influencers and tens of thousands of users.

And our influencers have been making like passive recurring revenue off our platform And engage and they're and they see their fans just more engaged than ever and seeing the early success here where we have a beta product now that is More on self service like bot creation.

And so then more influencers can use our service and also integrations with other creator platforms.

Joshua Schoen: So what were, obviously not all influencers are, actually talking with their fans. But this was but there, we didn't always have generative AI to capture people's voices. What what were they doing beforehand?

Maggie Sin: Yeah. All right. Yeah. So beforehand, a lot of creators would just manually, so there's Different things creators can do. So some creators would just like manually like message people or not and just have to manually look at all the [00:03:00] comments like messages see track their views and like data to see like what will resonate more with their audience or or the fans just keep messaging them and never get a response back.

Some other creators will leverage work with agencies that will take over much of the work. So negotiate with brands, I'm talking to like some of their fans combating back to these fans on their posts and that actually increases engagement and pushes the Instagram algorithm to show these posts more.

They also hire like outsourced labor to help service these like manual requests coming in from these fans. Yeah. So that's why we see that our software and our AI technology really helps these creators because we basically mimic the influencers. We give them like 100x leverage in the conversations they can have.

The creators can zoom in and zoom out of the conversations they want and the fans Are able to learn more about the creator, see exclusive content and the AI provides better than that than any agency or outsourced labor can. So we [00:04:00] increase in quality and also like the costs are lower than hiring outsourced labor.

And so we enable creators to have a business that they can choose to be hands on or hands off and also like start getting that money that they put so much time into. And to go into a bit more into that so basically that means that all these creators on Instagram or even TikTok or just any social media platform in general they post all this content.

They get a bunch of users on because they're like these users and fans who want to see more of the content and Instagram doesn't pay them for like pushing more content out. In fact, like Instagram gets paid for ads. On their platform. And so the creator, just pushes content and doesn't get any of those ad kind of revenue, they get money from sponsorships partnerships and like other like brand part like things but not really from Instagram platform itself.

And the only time I would say it's been like. Instagram will be like here's some money to put something. It's like when they like launch like [00:05:00] stories or like reels, like they were offering like money to creators to post more. And if you just post in a certain amount of cadence, then like you'll get some like cash from them.

But it's generally like a lot of the content that these creators produce, like they don't see the money. That they should be getting. And so my business helps enable that.

Joshua Schoen: Yeah. So it means that, so you're saying what the creators are more able to communicate with their fans on a more intimate basis than they would be otherwise, while they're creating the top of funnel to get them into the conversation,

and so what would you say, what would you say is the Delta? What do the creators get out of what they've, when they start using Mycompanions versus what they were using on other platforms?

Maggie Sin: Yeah, that's a good question. And so I just so I really ask about our influencers, like how much they make on other platforms, but I will [00:06:00] say that all our creators are very happy being on our platform.

And for some of our creators who we started with they they They made this is like the most they've ever made. And for some it's they've made like thousands on other platforms, but they see like this as a, it's given the amount of effort and work they need to put into because it does it for them, like they, they also get like a big chunk of money from here.

So for some, it's this is like their only like venture, like business venture that they get money from it. And for others, it's this is like a supplement to it. But either way there's only upside for the creator. And yeah and the demand we're seeing from these users and their fans I think every creator is pretty shocked to see how much demand people have to talk to them.

So that's like a good thing. And it just depends on every influencer and what they currently use like what kind of platforms that they're using agency. And we work with the creators of all sorts. So yeah, it's variable.

Joshua Schoen: Yeah, so you have creators of all sorts. Oh, people talking about health and [00:07:00] lifestyle.

Probably in positive encouragement. Tell me how that engagement works on that level. And how has that been for those kinds of creators?

Maggie Sin: Yeah, so yeah, engagement depends on a few things. So number of followers, they have number engaged followers, they have. So it's like reach, engage, reach, and also how much times they promote this bot.

So are like their Telegram bot. And so they can promote it on their story, their posts, a real on their LinkedIn bio, or like even other social media platforms. But we technically we mostly work with Instagram models. Or like craters, Instagram craters. So yeah, in terms of engagement, like we see it adhere to all types of different people.

Like it's more about it's more about like how much they, like how much engagement they have. So we also we get a lot of visibility on our website as well too. So like people come on our website and be try out different bots to see what fits for them. And [00:08:00] what's great about it is that if.

They don't like a bot or they are like they want to try something else. We have like different ones that you can try with like different voices, different media, different styles, personalities. We also have this thing called role play. So it's like an AI role play. So for example, you can talk, like talk through a story, like getting dinner with one of your, one of our creators, you can get wine recs from like my bot, for example, you can get food recs from another, like creator.

And yeah. And yeah, and I, for example I really, I'm really passionate about wine and I'm also WSAT 2 certified, and so people like to talk to my bot and we just, and the AI just feels in that It's basically like a whole like wine book behind the scenes and you can just always get that information in my voice.

So yeah, it just depends on like different people's use cases. We definitely see like people, some people who just want to talk to somebody. We see people who want like more like a professional like education from these these bots delivered in their voice of their [00:09:00] favorite influencer.

Yeah. And some people just asking more about the influencer or creator themselves oh, like, where are you from? What do you like to eat? What are your favorite brands? And all and this AI, we pull we pull in the context from the creators themselves to formulate a personalized response to these fans and users.

So it's, yeah, so It just depends. And we also have retargeting as well too. So that's also like automatically done where if someone hasn't talked, used the bot in a while, it will come out and spark a conversation about something else or something. And it's also easily to pivot from one combo or one topic to another, depending on what the user says.

So yeah,

Joshua Schoen: what are the users getting out of this? Like, why do they want it? Why would they want to talk to a bot when they know it's a bot?

Maggie Sin: Yeah, so yeah, so well, it's The conversations people have is pretty close to a human. So [00:10:00] we've spent like over seven months of fine tuning here. So we have there's like a long context window.

So the bot actually remembers what like was previously said in the conversation. It's personalized message. And also for some influencers, like the voice that it delivers is almost a one on one match with your fans love. And like the I would say that the one thing that's very different is like these balls are 24 seven, which isn't human like.

In a way like people like it because it's like mentally it feels like you're talking to the real person, and you're getting someone to respond to you 24 seven. And yeah. And also you're able to like, get like personalized responses, which in fact, like if you talk to someone like on Instagram, or even just like to a friend, they may not even respond to everything you said, or may not even read your message correctly.

And so this bot is has promises that like quality conversation, remembering who you are and like what you like, what you have said previously. And yeah, and I would say There's like people in this is real and people do find it's they just [00:11:00] want someone to talk to.

So if they message someone, they may not respond. But these bots are always there for them. And we've seen like messages where people are like, I'm sad. I just want someone to talk to are like, I'm depressed. And thankfully Thanks to the, AI LLMs that are made, like they, they're very generally positive, uplifting, supportive.

And they're like, like when they feel like they need, like this message like needs like a mental health professional, it will say it back to the person on like right then and there, instead of waiting for this person to wait till the next day to be like, Oh no, are you okay? Like from someone else.

And yeah, I think like all this, just like some like just having something being there all the time and emulates a real person is like actually very beneficial for like users and fans and it's something that they would want to use.

Joshua Schoen: In some ways, it's. You're creating a friend who can actually be there, remember, details about you, talk to you about things and not actually ever blow you off.

[00:12:00] And if you're really not feeling it they can point it out to you and tell you where, that you probably need, start needing to get help.

Maggie Sin: Yeah. Exactly. Yeah. And in some ways it's better than a human, right? Like not everyone there's a lot of Like humans, like we most words in the dictionary are actually like negative and some people they bond on more like negativity and that kind of puts people in a worse mental state, which were like a lot of people aren't aware of.

But AI and like this intelligence is able to discern like what's going on with that person and move this conversation that's negative to a more positive. And we see that happen. So like the LLMs, the AI and LLMs we use, like when someone talks to them about I'm not feeling well, like the LLM will respond tell me more, like you want someone to talk to and then when it gets serious, it's Defer to a mental health professional.

Joshua Schoen: Yeah. And how does it capture the voice of the individual creator? How do you get set [00:13:00] up when you're doing that process?

Maggie Sin: Yeah. Thanks to a lot of generative AI voice platforms, there are now that's very possible. So all we need to do is have a voice recording. We ask the creator to do it in a place of like minimal background noise.

If there's like microphone like. Products like use it. And they just send over like a 10 minute voice recording and we feed it into so we use 11 labs. So we feed it in there and it that training data uses like over like thousands of voices and they have different accents available too.

And we feed it in there and then it's able to like, spit out the words you want to say, the sentences. And yeah, like sometimes like the creator, has like a very unique accent. And therefore we asked them to accentuate the accent, talk slowly when they like send us for voice recordings.

And sometimes it does mean that they have to send us more than 10 minutes. Like we have a creator who would send us like 30 minutes because we want to capture like their unique accent. Yeah.

Joshua Schoen: So in some ways you're actually having a real life conversation with another [00:14:00] person.

Maggie Sin: Yeah it's pretty realistic.

Yeah. And yeah, a few of our creators like promote like they have videos like they make like voice videos on their Instagram and TikTok and they're like, AI bot. This is sounds like one on one match like to their voice and their fans love it because it's it's they want to hear their voice.

They want to hear them as if it's like they're talking to them.

Joshua Schoen: And what do you think is the future for Mycompanions and platforms like it? Is it, are we all going to be talking with bots? What does that actually mean if we're not talking to other human beings and end up just talking to other AI systems?

Maggie Sin: Something wrong with that? No, I'm kidding. Yeah, so I think AI is our revolutionary. We see it everywhere, right? It's like the way we look for information, seek someone to talk to summarize a document, like it's just changing the way Consumers and also businesses work. So I think the [00:15:00] future of okay, so the future of creator economy is going to be huge because I think there's so many use cases for them to like creators to like are used for people to use the creator's voice and their like likeness to deliver a message or the weather report, or then we tell you to do last for.

Task for today, et cetera. So it's a massive market. There's lots of platforms here and lots of businesses like trying to like, make AI a bigger use case for just everyone. And so I think the future. For specifically the creator economy. It's in the future brands. I think it's going to be like pretty much creator driven.

It's viewed as one of the largest markets for sole entrepreneurs. It's a massive market. And even like greater businesses like businesses created by creators are outperforming like tried and true individual brands like for example like Mr. Beast is chocolate brand is out competing Hershey's.

And there can be a creative for everyone in every industry and people are [00:16:00] spending billions on content every year. And therefore, I think that native creative companies will be worth like billions of dollars and a lot of this industry can be that is like service intensive is can be automated via AI.

Therefore. I think that yeah I think the future is going to be like AI delivery, like people's messages or to do lists, weather reports, all these like use cases that they want to hear their favorite information. Yeah, just because. Of, people just like these people are like fans of these people.

They want to hear it from these specific people. And we already see that in like the parasocial relationships that we see since like for decades now. Like people like to listen in when a specific person is speaking. There's like role models out there. Yeah. And the AI technology is just becoming so good.

And Our technology at Mycompanions like we're a middleware layer that works across like various different models. We provide context to the [00:17:00] models, and then we can, we're able to emulate the exact create like creators style and content. So that's really cool. It's like basically AI clone.

 So I think AI will be very it's just going to get bigger and people would actually prefer to use AI due to its reliability, responsivity, and also how much money it saves businesses and even creators time and the, and also if you tune it well you would drive the cost lower and it's just more efficient.

Joshua Schoen: So Maggie, what does this kind of mean for society? How if I'm a user, like, how can I be sure My information or what I'm saying to this bot isn't being used for, helping target further advertising or sell products to me and that I can have an open conversation with a companion.

Maggie Sin: Yeah, that's a great question. And say chat safety and privacy is a high priority of our company. So chats are [00:18:00] very safe on our site. We don't let any third parties access any conversations on our platform. All data is private and highly secured, and you can just use our products without supplying your name or any information about yourself actually.

And we. Yeah, so it's basically a conversation just between like you and the creator, and we're very conscious about data integrity and also like making people feel secure when they use our platform.

Joshua Schoen: And what do you think is the future for Mycompanions and. Where do you think it'll go?

Maggie Sin: Yeah. So I think this is a very growing industry and a lot of creators are very interested in AI and as they're learning more about it, they want to get more and more into it as there are a lot of benefits to these creators.

So we are. We have a self serve bot creation process right now on our website and we just launched that and basically any creator can just apply for a AI twin of [00:19:00] themselves and like you know start getting earning money quickly and keep their fans engaged or they can even like just contact me to learn more how to do it.

Second thing is we are partnering with like big creator platforms. And having integration with these companies. So to reach like more creators and to really get the word out there that like these creators can start a new, like business venture with us, it's very easy and it benefits both them and their fans.

Joshua Schoen: And sorry, one second. And where and how can people sign up right now if they want to be a creator or if they want to be a fan?

Maggie Sin: Yeah. So our website is www. Mycompanions. ai. Creators can sign up for a bot there. There's like a whole page and it's also in the nav and fans can also explore creator bots there. People who know creators can also refer creators and they can get a net revenue share.

We have a very attractive offer right now. And there's also a [00:20:00] contact us option there to email me directly if you have any questions.

Joshua Schoen: Awesome. Thank you so much, Maggie. And it's been great having you and learning more about Mycompanions and the use of AI in our everyday lives.

Maggie Sin: so much, Josh. I'm so happy to be able to share more about it and talk more about Mycompanions and AI.

This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit workingitwithai.substack.com

Welcome to Episode One. I still haven’t decided what to call this podcast, but I am speaking with some of the top leaders in tech and VC focused on the future of work. Still a work in progress, so don’t mind the scaffolding!

This episode provides a deep dive into the evolving role of generative AI in legal technology, exploring both its potential and the prudent safeguards necessary to harness its benefits responsibly.:

My guest is Ryan McDonough, Head of Development at KPMG Law

Ryan shares his journey from a software developer to leading development teams in legal technology, emphasizing innovation and efficiency in legal processes.

Main Discussion:

- Ryan discusses the integration of generative AI in legal tasks, highlighting its effectiveness in automating routine tasks like document summarization, which allows lawyers to focus on more complex, value-adding activities.

- He explains KPMG's approach to developing large-scale solutions that improve. service delivery across their global network, aiming to handle legal tasks more efficiently without overwhelming staff.

Key Insights on Generative AI:

- Generative AI's role in legal tech is more about aiding with tasks rather than replacing jobs, shifting focus from manual labor to supervisory and analytical roles.

- The technology is especially beneficial for tasks like lease reviews, where it can automate information extraction and basic inquiries, significantly reducing manual effort and time.

- Ryan underscores the importance of implementing smart systems to direct tasks to appropriate tools, optimizing resource use and ensuring accuracy.

Challenges and Considerations:

- Ryan addresses the hype around generative AI, clarifying its current capabilities and limitations. He stresses the need for clear instructions and frameworks to guide AI applications to prevent errors or misinterpretations.

- The conversation covers the necessity of building robust guardrails around AI to manage risks associated with data privacy, energy consumption, and misuse of AI-generated content.

Future of AI in Legal Tech:

- Ryan envisions a future where AI tools are seamlessly integrated into legal workflows, enhancing productivity but still requiring human oversight for quality control and decision-making.

Conclusion:

- Ryan is open for further discussion on LinkedIn, where he shares additional insights through his blog and other resources.

This episode provides a deep dive into the evolving role of generative AI in legal technology, exploring both its potential and the prudent safeguards necessary to harness its benefits responsibly.

This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit workingitwithai.substack.com