The information economy as a rod. The world is teeming with models reinvent every industry, industry. Inside Analysis is your source of information and insight about how to make the most of this exciting new era. Learn more and inside Analysis dot Comside Analysis dot com. And now here's your host, through Eric Kavanaugh. Yes, oh, yes, folks, Welcome to the future. Indeed it is July of twenty twenty three. The future is here already, It's
just not evenly distributed yet. That's the tagline for our TV show future Proof. Check local listings. We just got our own time slot in Washington, DC and Fairfax County. Pretty excited about that. And timing is everything, folks, and this show, the Inside Track, came together quickly. I heard some exciting news I guess about a week and a half ago that Bruno Aziza was going somewhere. Of course, he's been in Google the last few years, and we're like, all right, what's next with Bruno. It's
always exciting. It's always interesting. That I found out he's a newly minted partner at Capital G which means, Bruno, we're going to call you the g Man. I'm pretty excited about that. Bruno Bruno's everybody in this business. He's been around a long time. His passion is undying and amazing. So we're very excited about his new gig over there, and we decided to do a little series of shows about innovation in our space. And what a time to live in this space. Oh my goodness. I've seen lots of
trends in the last twenty five years or so. I've seen cool things come and go, but the size of this wave and the magnitude of what's changing is like nothing I've ever seen before. And of course AI is at the heart of all that. Generative AI is the topic du jour, and we're going to talk about what to do with it, what not to do with it, what gets us excited, what gets us concerned. There are a lot of things to talk about with generative AI. It is incredibly powerful,
but you do have to be can you expect from it? So you have an all star cash here, Bruno Aziza of courses with us from Capital g We have Yashaalytics these days and San Chief Mohan or Kamo, and I'll just say that, yes, this stuff is amazing, Yes we have to figure it out. My big concern maybe we talk about in segment two is explainability. If this stuff is going to be you are regulated and they're going to be a real issue, it seems to me. But let's go around the
room, get introductions and see who's excited about what. Bruno Aziza, congratulations on your new gig. We're all very excited for you. What gets you excited about Generative AI? Well, I think it's really always nice to talk to you. And so for those who don't know capital g it's Alphabet's independent
growth funds. So I'm still in the Alphabet family. Of course, if you have my email and contact information is still the same, you could reach out to me there For me that I've been in this did ai and analytics market domain for what twenty five years? I have we known each other, eric In, Yeah, we've all known each other for a long time. And so what's exciting to me is if you remember at the beginning of our career when we started, nobody cared about data. Right, It's a back
office topic, it's an expensive one. Why do I even need to have to store data and so forth? And I think what jen Ai is doing
is changing the game for that. Now the interface for data has changed everybody can work with data, which has of course advantages in terms of democratizing access to information and letting people kind of interact us questions and saying it rate around the answers around the data, but also has things that gotchas I think in the world and I know we'll talk about bolts, but in the world of gen AI, of course, if you have bad data, it's really going
to hurt you. So having high quality data I think is a big topic of course to look at. But if you look at all the forecasts in terms of opportunity in this space, it's there's got some really really good oportunities,
both from research firms. I think was it McKinzie recently that has said that the opportunities and the trillions right, it's almost as much as the UK's GDP and organizations like Walmart that are starting to work at very large scale with AI to drive productivity for their employees and then create compelling customer experiences for their
customers. I think in the case of Walmart, last week at the venture be Transform, we learned that approximately fifteen million, fifty million Walmart customers are interfacing in any way, shape or form where their conversational experiences and over two million, two million associates inside Walmart. I couldn't believe it myself. Had to check how many employees does Walmart have? They have two point three millions, so I guess it makes sense. But two million of them are using
this application called asks pees associates be more productive. So I'm really excited about it. But of course, like anything in data, we got to do it intentionally. We got to do it just like you said, Eric, in a very open manner, so it's transparent and that it's accurate, it's we removed the bias, and that it's also sitting on high quality data.
So these decisions that we're going to make out of the data that we can interact with are not going to be detrimental, right right, And we have to audit this stuff and know what happened to be able to report back what happened. You know, there are a lot of perhaps hurdles in the future to get over. At the moment, I think people are just trying to figure out exactly how to use this stuff, and there are tremendous case studies.
I mean, to me, a customer experience a number one fueling your chat pots fantastic because you think about if you could point these large language models or even small language models we'll talk about that, point them at the corpus of your customer interactions over the last two three years, for example, and in very little time you can have a foundation for a chat pot that will know most of what people have been talking about with your brand. So to
start off like that, I think is a pretty powerful storyline. That's a pretty compelling business case. But maybe we'll bring Sanji Mohan to comment on that. I think there are still a lot of miles to be driven figuring out where we're going to use this stuff and what the sort of rough edges are. But Sanji, what's your take? Thank you all right, so that you actually set me up really well, because what we have today is a game changer in the sense that I can use open AI and I can ask
all kinds of questions. It's really democratized AI. Anybody in the world can finally use AI. But there's one area that we still have to cross a hood, which is how do I use LMS on my enterprise data most Because once we can efficiently and effectively start doing that, then all of a sudden, I can do stuff that I have never even dreamt of. Today. I write sequel queries, I do keyboard searches or I do know. It's
on my structured data and my unsquerry. And not only am I getting the results from my database, but I'm using a large language model to give me some sort of recommendations, aggregations, some action plan things like that on top of what I just got. So this combination of keyword search and semantic search or similarity search is what is super exciting for me. Yeah, that's that's
good stuff. And I have played around quite a big with chat chypt and it does reflect back to you some remarkable depth on even some esoteric subject areas, because who cares. It's captured all this stuff that they've baked into this engine basically, and then you just give it prompts to see little fragments that
come out. Of course, it's designed to fuse vectors of information, so as the go to market from chat gpt chid to be on an email through sorry in a Twitter thread during the data breaks show, he's like, well, you know, remember the who listen nations are more of a feature as opposed to a bug right, Like, the whole idea is to generate new stuff. So the question then becomes what material do I provide you to generate
new stuff? And then how do I guide or train that? And we can get into the challenges maybe in the second segment, But I do believe that there is still a just whole host of value if we can figure out how to leverage it and how to kind of move things forward. And yashall bring you into comment on this year an old hand at this exciting enterprise software stuff. What's your takeaway on the upside of jen Ai and where things are going? Well, look, I kind of sit in the middle between what
you get excited about. Eric and let Sunjeeb talked about that. I mean, you do. I've known each other, not super super in depth like you and Bruno do, but we have no each other for quite a few years. I've been a CMO for twenty five plus the years, but a long time there's been like one thing that at the center of the universe for me in every organization I've been a part of, and that the better know who my customer is so that I can serve them in a meaningful way.
And there really is one fundamental problem with that, and that's the data that exists inside of the organization is fragmented and it's all over the place. Fundamentally,
it's inaccessible. And it kind of doesn't matter what type of new tools we buy, the type of things that we layer on top of it, what new technology comes about, like, it's just not in a format that makes it possible for me to be able to write the right kind of queries or even forget about like get the right kind of people that can write the right queries. We are now at a point where we'll call it jen Ai has opened up the opportunity for accessibility to truly understand who our customers are,
to be able to create meaningful experiences. And this is really what is most exciting in my mind right now, not months from now, not years from now, but like right right now. We can mechanize the organization of data inside of the organizations that we are a part of. We can use LMS to get acts into all of that data in a way that can help us derive the right insights to be able to create the best experiences for our current customers. And yeah, I don't know, I mean, I've been a
rapper a long time. I've got a bunch of great here. There is no time that I've been more excited about being able to utilize technology to create better experiences for my customers than right now, even back in the late nineties when the Internet was retarting, Like right now is an absolute golden age. It is so incredibly exciting. Yeah, I have to agree. And it is a reflection. That's the thing. These large language models reflect the world
around us, reflect the world that they have consumed. And of course they're fast evolving too, So these things are changing. As we talk about this on our show right now, advances are being made. What did one guy say, I'll throw it over to Bruno to comment on. He was talking about healthcare, and he was saying that today is the dumbest this engine will ever be, and it already knows more than any doctor has ever known. And so you start recognizing the power of this breadth of information. Yes we
have to vet things, Yes we have to be careful about things. That's sort of the next phase I think of evolution in these models, these foundational models, is getting focused models on particular industries, because when you can dive into an industry there's a whole lot of training you don't have to do, right, you get a lot of training on the box. When I think about that and think about the fact that we can now leverage that other eighty
percent of all the unstructured data. I mean, in our data analytics industry we've been focused on for a long long time. Now, well, guess what that eighty percent provides all the context that makes the rest of the twenty percent meaningful, And so now we can leverage that for you know, what do you think absolutely? I think using an industry like health scare. The Mayo Clinic is an example and organization that's pouring through medical records to help the
practitioners get more efficient. In another area, as the legal domain, I think a company like Accenture, you know, and they've been doing this for years by the way, right we're talking to the broad markets talking about it now, but many organizations that have high level of maturity within engineering have been working on this for a while. So Accenture, for instance, we'll work for them on this program called ALICE. It stands for Accenture Legal Intelligent Contract.
Over a million contracts added and every month into the system and using kept the patterns and really help your thousands. I think it's twenty eight thousand professionals globally to deliver faster value by pouring through all this instructure data and get to the bottom line. So there's industries that are going to benefit from it a lot faster than others. I think the main bit here and we touched it at the beginning as there's a lot of confusion right now, a lot of
hype. So the industry, I think needs to simplify their approach. Every customer I talk to as a council for figuring out okay, how do I power students And hopefully we'll talk about the in the second part of this show here. So there's a need for simplification, needs of an intentional and proactive approach with specifics this case that needs to be made. But right now there's some industries you could take advantage of this technology almost immediately. Again, if
they've got the right technology, things you can do. I call it a dynamic cliffs notes generator where you can take a one hundred throws over to you, dump it into chat, GBT or BARD or any one of these major financial models, and say give me the two page summary of what this document says and boom, it'll do it. Then you can say, all right, now drill into this part, give me more infraction about that boom.
I mean, it really is like the ultimate assistant. It's like an incredibly knowledgeable intern that make it a few things wrong here and there you have to check up on, but we'll do whatever you ask them to do and won't
complain. What do you think, Sage, Yeah, it's like you should treat or treat these llms like having a personal butler who has who have memorized all of Internet, and I can ask questions and then apply some of my logic and then certainly I have this powerful helper at my back end, call anytime I want. So, So where I see us heading And to the example Bruno give for legal professions, what we have today these lms that are trained on vast corpus of data. But in an enterprise, I mean not
need all that. I don't need my LM to have memorized all a stack overflow and read it and get hub. But maybe I've All I want is for it to summarize or find legal presidents of my fully years of legal practice. So where I see us heading and this is a bit controversial because we're not there yet. Is an ability to train my own enterprise model in my domain and pointed to my documents. So what I see happening in future is
censure. For example, will have a domain practice for legal, another one for supply chain, the one for marketing, and it'll say, you know, here is a blueprint of an LM and we can train it for you. Today, that infrastructure does not exist because you need thousands of GPUs which how to get the very expensive No one's got extra millions of dollars to spend
to train it. But over time, I see this expertise the infrastructure becoming more prevalent, so I will have this ability to train the model on the data I want, very specialized, so I call it small form factor LM, and then I can start throwing questions and have an amazing copilot or an assistant to help me. Yeah, that's a good point, and I think you're really honest something. And I refer to these as small language models, and that may not be the best term to use, but the point is
you don't need everything. I only need what I need, and you don't want to cloud up your view of the world with lots of extraneous concepts and information. I mean, this was the thing I'm kind of thinking to myself as I'm watching demos at the Data Bricks seven and granted, this very powerful stuff that they're building, very very thoughtful, and they're doing it in any very what's the word clear fashioned way, I guess is what I would gather
from that. But nonetheless, there's a lot of training that's going to have to take place. Where is the value in that if I have to spend six months training this huge model to my business, Well I kind of lost six months there. And is are things going to change by the time that six months goes by? I think so, and so I think you are going to see I mean, it was funny. Even Sam Altman came out and said the era of large language models is over. And he's the chat
GPT guy. Now. Someone I went and met at the Reuter's conference and often last week said, yeah, it was a very self serving statement by him. Don't listen to him. We think that's nonsense. We're gonna be fus whole ecosystem built around it, just like we saw in the days of head Dup, right, head Dup came out, everyone got excited about it, and this whole ecosystem grew up around it, sort of ins and pieces
and fits and starts, and of course had to kind of disappeared. I don't think the foundational models are going to disappear by any mean, but it is a warning to everyone out there. Be careful. We'll be right back. You're listening to inside and out For me, what if you could own a piece of the future. What if you could build your next castle not on sand, but on the bedrock of a modern blockchain ecosystem. The first
Internet gold rush made millionaires. The second wave is minting billionaires, but the third wave is just gathering now and anyone can get in on the action. Hop online to crowdpointtech dot com to learn how you can secure a foothold in the blockchain revolution. Whatever your passion, wherever you want to go in life, there's an opportunity awaiting you right now. Go to crowd pointech dot com to learn how the blockchain who will fuel the next generation of innovation in this
globally connected world. That's crowd pointech dot com, your trusted agent in an untrusted world. What's the longest running radio show in the world focused on data, DM Radio. Want to be a guest sometime, send an email to Info at dmradio dot biz. That's Info at dmradio dot biz. Can your IRA stand up to the next financial crisis that our top economists are saying is
at our doorsteps? By allocating a percentage of your IRA into physical gold and silver with a tax free rollover, you can diversify in safeguard your holdings from turbulent markets and economic downturns by putting your IRA back on the gold standard. Find out how to safeguard your STS with a tax free rollover with a Genesis Gold IRA, the only IRA that can hold physical precious metals. Call now
for your free Golden Silver report. Protect your IRA today with one simple phone call and learn how to qualify for up to ten thousand dollars in free silver called Genesis Gold Group Empowering Faith Driven Stewardship. Eight hundred and six four four eight six one one eight hundred six four four eight six one one eight hundred six four four eight six one one. That's eight hundred six four four eighty six eleven. Do you own a timeshare? We'll face the facts you made
a mistake. You made a bad perdit that continues illegally. Get out of your contract. Call my friends right now at the Timeshare Exit hotwayers who help good people like you get out of a timeshare contract that they just don't want. Don't throw away your money on maintenance fees. Use it for things you really want. We can help you end your time share contract and stop the money drain immediately. If you are ready to move on with your time share,
call our team right now. Cancel your time Share now with a free call. Eight hundred two eight nine oh four one three, eight hundred two eight nine o four one three eight hundred two eight nine oh four one three. That's eight hundred two eight nine zero four thirteen. If you served in the Marine Corps, by now you know about the contaminated water problem at Camp Lejune. If you were stationed or worked at Camp Lejune from nineteen fifty three
to nineteen eighty seven, you probably have a lot of questions. We have some answers. You could be entitled to compensation. Billions of dollars are being allocated to pay for damages to anyone stationed at Camp Lejune during that time. Unfortunately, it appears that officials may have known the contaminated water problem existed and did little to protect their men. The Semper five code was not honored.
If you or someone in your family has developed a serious illness in various forms of cancer, called this camp lu June legals back the clock and change what happened. But you can certainly call right now and learn your rights as a marine. Here's the number. Call eight hundred two five four three two one eight eight hundred two five four three two one eight. That's eight hundred two five four thirty two eighteen paid for by Legal Alert Line. Welcome back to
Inside Analysis. Here's your host Eric two on Inside Analysis talking to a whole host of experts. This is so much fun. We've got Yasha from Lytos Bruno from Capitol G and of course Sanji Mohana, good buddy from Sanjmo. And on the break we were chatting about accuracy and yes, accuracy matters. You have a fun side stool throw at you. One of my good buddies who's also a client, a guy who runs a company called Crowdpoint Technologies.
He was invited into Open AI two and a half years ago and he went there and talked to them and signed an NDA and said, no, no, you got it all wrong. You're gonna have a short term memory problem, you're gonna have a long term memory probably, just like outlining all these issues they're going to have. And so he didn't go with them, but
he couldn't talk about it because he signed an NDA. So he said, this is just what I'm in Texas because I just got out of the NBA's and he's got a concept that we'll take intell intelligence and the ideas you apply machine learning to trusted corporate data, so instead of just using whatever information was on the interwebs for the last ten years to show you what's going on. No, it's very focused, and that focus really matters in the enterprise SI,
but throw it over to you. Accuracy, governance, security, trust, these things must be in the enterprise world. We don't have the luxury of consumer media and not really caring about the truth. Right, Yeah, So I want to bust this out. The accuracy, which is what I'm going to talk about. Security and governance you cannot have. So you shouldn't put out there that is giving you people's you know, personal private issue. The information that would be a data privacy issue. So but let's talk about
accuracy for a second. We have to understand that lms are not designed. If you want accuracy, do a keyword search, do a sequel search. If I ask a question to my database and say show me all the top salespeople from Q one of twenty twenty three, it will give me the one the same information every single time. It's a very deterministic query. That's what
we're used to. But an LM is trained on probability. It's a statistics thing, and the answer can change depending upon you know, the situation where people are asking a lot of like there's a lot of world match and everyone's talking about some new incident. Next time I ask a question, is that the probability of the next world is different. Now it's a probabilistic thing. So so is that a problem? Yeah, of course, I mean if you cannot trust it. But the idea is that if we understand the way
LM's work, then we start using it for our own benefit. There's a question in QNA that business will get frustrated. Yes, they will get trusted if they don't understand what is the purpose of an RLM. The purpose of an LM is, let's say, to aggregate the information that I've just sent it in my prompt. If we go with that, you can now use LLM for what it's good at. It's a language model. You know.
There's that famous case where recently a near some attorney, now you know, he asked a question about some presidents and three cases came up and they were factually incorrect. Well, he was not supposed to trust that data out of the box it was. It's supposed to assist him, and he just presented it into the court and they don't out to be wrong. So that's a wrong use of an ELM. Yeah, no, that's right, And that's
what I think some of the big prominent journalists were doing as well. The guy from the New York Times whose chat gbtpot set I want to marry you leave your wife or whatever it's like, Well, you let it down the garden path. It's the point you were misusing the tool, and therefore you didn't get the desired effect. But you know, to kind of keep this on track and maybe I'll throw it over to Yasha. It is generative,
so it can get you eighty percent of the way there. This is I think the proto principle is always out there, and it seems very good at getting you eighty percent of the way there, to give you some structure, to give you some form you can work with, and then you kind of mold it like clay to get it where you want it to be. In that category, it's very very powerful. But just be careful about issues that must require accuracy deterministic value use. If you will just be careful about that
stuff and you should be all right. What do you think, Yoshia? Yeah, well, I think that the concept of copilot is maybe the big and most important takeaway here, right Copilot infers exactly what we need to understand, and that's that we're not trying to get the final answer. We're trying to get support to get to an answer that we feel competent with so that we can take control and take over what's next. You know. The question that was posed in the Q and A, I think is a good one.
Right Ep writes that given the publicity of LMS and gen AI, like, how does a business not get frustrated right about the lack of a readiness inside of the enterprise? And I think the question really comes down to where am I going to think about deterministic information inside of my business and apply LMS to it behind the concept of a copilot, so that I can apply gen
AI in a way that's meaningful for the business, right like you. Just like you said, you apply a model, you're going to get a result that's roughly eighty percentage accurate. And when you start too in that model in a way that is built behind the idea of a copilot, you can actually tune it up even better. Let me give you a really specific example. The business that I said and looks at zero party in first party data like
it's really really simple. It's all deterministic data and the idea of a copilot being able to look at data make a match of an inbound set of data into a customer profile. An LM tuned to deterministic data, you can get to a ninety seven percent accurate mapping in a copilot model every single time.
So again, it depends on where you want to look. So EP's answer to get ready inside of the business, look at deterministic data, look at your customer data and zero party first party data, and think about the copilot metaphor. Because that's really the way to get to the end result that you care about. Don't expect the right answer. You know what's interesting here,
I'm going to tie this together. This might be a bit abstract, but I remember when the mobile movement really took over, and it forced designers, why designers to rethink their entire layout, where the functionality lives, what you can get access to from which window, all that kind of stuff, and they it really forced because of the real estate constraints. I'll throw this maybe
to Yosha first and then to the other guests. It forced the developers to really think through workflow and what is necessary at this point in the workflow that is not necessary at a previous point or at a following point. Right. And so this copilot concept I think can be very compelling in terms of guiding
you down the path of what can be done. Because if you think about and I always pick on Adobe Photoshop just because in like the year, you know, nineteen ninety nine, you would have all these menus and all these options on the menus where at any point in time they're like eight thousand things you can do. Well, you know, no one's just going to figure
that stuff out. Especially photoshop, you have layers and diffusion layers and all this kind of stuff going on, But at any point in a given workflow, there's only so many things that you can do. And so in that sense, I think generative can kind of guide us through what we're complex workflows but can now be fairly simple, while also educating the worker in the process. I'll turn it up at Bruno first. What do you think about that idea of Bruno? I think it's a it's a great idea. It's a
great example as well. I think if you look at any adoption curve of any new technology, there's three big questions we need to answer, right, what is this? How do I use it? And so what? And I think you we've talked a lot about what is it? Right, and Zanjie just explained, Look, this is not magic, it's probability at scale, So that's what it is. The question I think a lot of customers
are asking is how do I use it? And so I think what the industry needs now is a lot of simplification, a lot of guidance on how do I think about using this amazing technology everybody's talking about. I came up with this acronym d ACS, which spells dates I guess, and it talks about what are the verbs you're looking for when you're using GENEI. The first one is draft or document Right, that's the d is answer or not invent
answer. See it's classified. There's a great job classifying, adit and summarize. But you got to think about, Okay, now that I have these use cases or these verbs, these jobs, I'm trying to use this technology for how do I find example across marizations. So I'll give it an example of Nfair. You probably go to Wayfair to look at a lot of you know, material you want to use, couches and so forth. They looked at, okay, how do I help people that are creating content my website.
That's an example of documentation or maybe creation of content of a body of data that already have and accelerate the productivity of my copywriters in the case of Wayfair, and they're looking to double triple the productivity of people creating MIDA data, if you will, off of data they might already have. You look at developers. You know, developers can benefit from JENA at least in three
different areas. This actually came out from McKinsey research. One is existing meaning if you're a coder today, you probably are going to a lot of manual, repetitive work to document your code. Well, McKinnon do that in about half the time. If you're a developer and you want to jumpstart, I think, yesha, I was talking about this idea of copilot. I like
the idea of companionship. You know, if you're a content creator or even a coder, often when you get started, you have a blank canvas, so you can play within gen AI and start getting started on how you know, what should I create and so forth, an accelerate update to existing code and so forth. In this case, you know what McKinsey were code refactoring, content refactoring at about two thirds of the time. And then finally,
I think the third example here we keep talking about Jeni. You know taking over the situation is more a human without that tool versus a human with the tool. And so an example is, you know, in McKinsey's example, they found that developers using GENI tools are twenty five to thirty percent more likely to complete complex task than people that don't have it. So you can now
kind of unleash imagination with the assistant of that theology. Interestingly enough, you know, we think about these as this nice element that if Eugenai tool effectively, it becomes this human companion that you can kind of run ideas with. But again, the invention comes from you. You shouldn't expect them to invent for you. You iterate with them. I think that's really kind of how because we confuse what they are and use them for those verbs I mentioned,
and how you can get value from them. Yeah, well guardrails, right, I mean, that's what you're getting here. And with the copilot stuff, I mean, and this is not terribly new, right, Google was finishing sentences in your emails like about a year and a half ago, so that was kind of an early indicator of what's coming down the pike. Right. So now I really like on the copilot thing, and maybe I shall throw it to you. We've got to It can help you understand where to
go. It's like having someone at your side all the time says, now you're going to turn left. Now you're gonna want to turn right. Oh, maybe you should do this, and you're gonna see these suggestions come out. But I'll ask you quite eving this more every day that the manifestation of AI will be in the form of suggestions and as long as you accept that's what they are, and then you take accountability for the action based upon the
suggestion, then you're fine. Now I'm the human who is accountable. What did someone say one of these events recently, The question was like, oh, what is responsible AI? And she says, there's no responsible AI, and there's not going to be. There are people. Only people can be responsible for decisions they take. And so that's where the accountability comes in. It's in the first it's not in the AI. But Yasha, what do
you think? Yeah, but okay, I think you brought up a that we should call back into when you talk about the paradigm shift that happened a popular out to consumers. Right we're going through that right now. But we have this really really difficult challenge to work out of the rapid like mass mass adoption of jenai that's happening right now is coming behind one single user metaphor and it's chat. So like everything that we're dealing with is behind what we've been
taught over the first of the last month and a half. We all have to work our way out of that. And it's like it's trivial that that sounds amongst the four of us right now and those that are listening like that is not trivial, Like we have all been taught that the way that we interact with jen Ai is through a chat interface, and it is not that. It is everything that you just said, Eric, it's everything the Bruner just said. It is about providing access into other information behind these models.
The user metaphor has to change. But that's going to take a lot of thought by different enterprises to be able to provide that kind of coopilot recommendation, whatever that experience needs to be. We're just kind of caught in this user experience trap right now. That's very interesting, a user experienced trap. I got to write that down. I think you're right, and I'll just throw one other interesting observation out there. One thing I thought that open Ai did
that was very compelling is the workstation in which you operate. And I've always wondered about Google and how they just don't use the real estate that you have on that Google search page. And now they do cool stuff with the visuals and sharing history of individuals and things of that nature, and of course you had I think it was Project Hummingbird maybe where it makes all these recommendations of what you think of what it thinks you want to look for. But nonetheless
it wasn't really a workstation. It was just a window into the Internet that has been indexed for me. Whereas what chat GPT has done in bar and these others now is create these workstation environments where you can kind of tell that you're at work now and you can kind of piece things together, because that's what you're supposed to do with this is use it as your workstation to piece things together and make sense of your business. Bruno, what do you think?
You know? I think sometimes the beauty of the UI is no UI, right, is to be really invisible to you. So I personally I really like the interface of Google Search. Maybe I'm a little biased on that one, but I do think that it gets out the way of everything else that could be busy here. So I don't know, I think you'll see. I mean, even Claude, I think just came out yesterday. So I think we're still in the experimentation phase of which is going to look like
a use case that's going to be the most helpful to me. I think y Aasha has got a great point. You had and thought about it, but you're right that we think about the interface today as being chat whereas we really should think about it much broader, where it's everywhere and it could now start pushing decision out to you so you can react to it rather than have
to ask questions to it. I'll be curious to see what people think about it, but I certainly think today lots of noise, lots of confusion, potentially a lot of misuse that when you turn this consumer AI trend into an enterprise AI trend, you're gonna have to figure out what the flamewoark case will adopted it. Yeah, now we're talking about generative AI and the power of data in the enterprise and around the world will be right back. You're listening
to Inside an Alpha. Do you own an annuity either fixed rate indext variable? Are you paying high fees and getting low returns? If so, Annuity General would like to have this free book to learn the pitfalls and mistakes of buying an annuity. The Annuity Dues and Don'ts for Baby Boomers contains the little known truths about annuities, like how to help reduce your fees and increase retirement income. And it's free. That's right free. As a bonus, we'll
also throw in a free annuity rate report just for calling. We researched over one thousand annuities and summarized rates and benefits from financially strong insurers. You get annuity dues and don'ts for baby boomers and the annuity rate Report, both absolutely free for calling Annuity General Today. Hurry supplies are limited. Call now eight hundred two four, five one six nine seven, eight hundred two four, five one six nine seven, eight hundred two four or five one six nine
seven. That's eight hundred two four or five sixteen ninety seven. Do you own a timeshare? We'll face the facts. You made a mistake, You made a bad purchase. A timeshare is not an investment. It's a money pit that continues forever. If you use your time share, that's great. But if you don't and you want illegally get out of your contract, call
my friends right now at the Timeshare Exit Hotline. They're an experienced team of lawyers who help good people like you get out of a timeshare contract that they just don't want. Don't throw away your money on maintenance fees, use it for things you really want. We can help you end your time share contract and stop the money drain immediately. If you are ready to move on with your time share, call our team right now. Cancel your time Share now
with a free call. Eight hundred two nine zero six seven O five eight hundred two nine zero six seven O five eight hundred two nine zero six seven O five. That's eight hundred two nine zero sixty seven. When a player's sudden cardiac event brought a national football game to a halt, it's shown a spotlight on the importance of CPR readiness. Now, with youth sports in full swing, the American Heart Association is rallying parents and coaches to be ready in
an emergency. To be ready, learn hands only CPR. It's a skill anyone can learn in minutes. Just visit Heart dot org slash hands only CPR. Hands only CPR is nationally supported by Elephant's Health Foundation. Each year, three hundred and fifty thousand Americans die from a cardiac arrest. When seconds matter most, CPR can be the difference in whether a friend or family member survives. That's why the American Heart Association is challenging every household to elect at least
one person to learn CPR. If you have ninety seconds, you can be your family CPR hero. Just watch the American Heart Association's hands only CPR video at heart dot org and become a hero. Do you need to get your hands on some extra money right now? Maybe twenty five thousand or more If you're a homeowner, Now is a perfect time to get hash out while homes in many neighborhoods like yours have gone up in value. You can use the
money for anything it's yours. You can buy an investment property, payoff higher interest at or make home improvements. If you need twenty five thousand, fifty thousand or more. Now is the time home values are up and so is your equity. We offer you a way to use it. No need to use your savings called New American Funding Now and see how much cash out you can get. Call eight hundred seven one h three seven three nine eightred seven
one h three seven three nine seven one h three seven three nine. That's eight hundred seven one h thirty seven thirty nine nmls sixty six oh six www, dot MLS, consumer XS dot org. This is not an offer or commitment to lend subject to borrow or improperty qualifications. Not all borrowers will qualify. Terms and conditions apply equal housing opportunity. Welcome back to Inside Analysis. Here's your host, Eric Tabanaugh. All right, let us and gentlemen back
here on Inside Analysis. We're talking to a whole bunch of experts here. We've got Yasha from Lytos Bruno from Capital G which is part of the Alphabet Group. But of course Sangmo and Yasha. You had a really good point in the break there around sovereignty data sovereignty. Obviously the Europeans are big on this stuff. You have to be careful about where the data is persisted, where it's processed, what your intentions were. We're starting to see more of
that in the US, certainly in California with's CCPA. And you said there is one cloud vendor that is promised it is not going to be naughty with your data. Tell us about that. Yeah, I mean, but let's even talk about at one level higher up, like everything that we think about in the enterprise, we have to think about behind our data strategy and our data strategy has to be cognizant of all of the complexities of consumer privacy laws. You mentioned CCP. We have like six new states in the US that
now have their own privacy a lot like there are. There is no slowing
down of consumer data protection, no slowing down. So when we think about the application of gen AI, like every single person watching and listening like you have to be thinking about how lms are going to have a relationship into consumer data privacy period like period right now, there is only one cloud provider that will kind of back up the fact that when you are utilizing their models and their model garden, the data that you input in the corpuses that you train
on aren't going to get reshared. And that's Google, It's protects Ai. There are lots of announcements that have just happened, and I suspect that over time others are going to follow because the acknowledgement of the importance is there. Right enterprises have to operate with confidence that they can work within the data sovereignty rules that are required. But as Google right now, So if you're going to go out and experiment right now and you want to have confidence, protects
AI is really the place to go. M that's a good point. And you do want to have confidence. Yeah, this is concept of what psychological security that they talk about some of these events these days. And the idea is that you want to empower your people with their creativity and their confidence. You want them to be confident enough to take chances to look for things. You know, as I'm sitting here listening to these experts, I'm wondering to
myself and you maybe throw it over you first and then Bruno. This whole concept around the hidden context in your data, the hidden meaning in your data. These language models, these foundational models can be extremely powerful once you trusted to put some of your corporate data in there to just look around and find stuff you think about discoverability. To me, that is the most important part of analysis as being able to discover what's happening. So companies like thought spot
for example, have a great UI that a great concept. You start throwing natural language query, I think all of that is going to open up to tremendous value because of these foundational models, because you'll be able to ask much more complex questions in just natural language of your existing documentation and then be able to kind of guide the conversation to understand how this fits into certain corporate regulations
or policies or directives. For example, you think about the marketing world where you have these themes that you want to deploy, So you could explore your data and look for where these themes map with something you already have. My point is we have this very powerful mirror now that we can point at our data to export in whole new ways that I think is going to be really
compelling. Sanchi, what do you think now you're mute? If I may give an example, a real life example, So every organization has to be with this problem that the customer man will renew. So today what we do is we create these machine learning models to determine customers chilling. For example. The problem is that the machine learning model can detect and predict which customers are at the risk of not renewing. But then what you know, we've all
these strategies we try. We kind of like throwing stuff on the wall and see if it sticks or not. So there is a huge potential where lms can now come to our rescue. And the way it works is a combination of predictive AI and generative AI. So so it's a two step processes. The first step is is because I remember, LM has got no clue about sorry I was going to use a bad word, has no clue about who
my customers are. So I ask a query to my database and I say, okay, predict tell me which customers are at the risk of our journey. Once it retreats that information, I can now put that as a prompt into an LM and I guess everything that you know on this data, give me some strategies of how I should go about And this is hugely power strained to do this kind of work. It can come up with some amazing recommendations that are even beyond a human beings capability, at least in a given amount
of time. So you can instantly get that. But this, by the way, what I just explain is what is going around in the market. We call it rag RAG, which is retrieval augmented generation. So this is a great use case of an ELM working on my because you have to know that these elms were the big ones were trained like open air once in September of twenty twenty one. They don't even know anything that's happened in this year, So how are they going to help an enterprise if they're so out of
date. So that's how these lms are now being used. Yeah, that's fascinating pronoun. You want to comment on that, Well, I'm going to see if I can alcronym sound. Gee. I love that rag. That's incredible. There's three things I would say. The first thing is the state of readiness today. As looking at the twenty twenty three I Readiness Report, twenty one percent of organizations stay there in production. Nineteen percent of them have no plans. So you can see in between there's a lot of people that
are experimenting, planning, trying to figure it out. And I think a lot of it is related to some of the issues we talked about, obviously quality, the input of the data. But a big one that you just talked about, Eric, which is really interesting, is the refinement of the understanding of the question. You refer to. Thoughts product company I'm very familiar with, came out of this product called Sage, where the benefit of it
is understanding the question itself. Now, what is the real application of that. You take an organization like seventy five percent of people that are ordering from when he's they do it through their drive through, and so that happens is Wendy's as this amazing menu and there's billions of possibilities that you can order. Everything is customizable, and so having a model that is able to understand and the refinement behind the question is what's going to make the answer really interesting.
You know, talking about acronym, I was trying to think about what would be the attributes in the ideal world of the best in AI type of application. And it's not as great as as an acronym as Sanji did, but let me try it on you. I called MTKAC, MTKC, so M and T. It has to be multimodal, multimodal and input and output photos, text, audio. It's got to be able to handle that because that's how we live. T Trustable why and we talked about the multiple dimensions.
Obviously there's bias, there's ability to trust the output. Trustable is a key notion that you know, we everything we do in data the foundation is trust. See is what Sanji was talking about. You know, if you give me insights from five years ago, not really helpful, right, So I need currency on the data. I need you know that your work of data that's recent enough a is for applied to a workflow. I think that's what yah I was saying earlier today, is that it's got to be applied specifically
to my workflow. And see it's contextual. You know how many of these experiences we've had actually even with other humans, where we have a conversation and we lost the flow of that conversation. So contextual is a big value if I think, if you hit all these dimensions, you end up having the real capabilities that you need to take your company to the next kind of generational
change here, but we're not there yet. We're at the beginning. Yeah, and I am looking forward to this the single instance future where you can point your corporate data at it, you trust this engine. Once you can do that, you can learn tremendous things about what's going on in your organization. I mean you'll learn things that you do never even knew existed. There are workflows that you didn't even know about that are getting the job done.
One of my favorite analysts friends from like twenty years ago, said something I thought that was funny. He goes, if there's something in a large organization that must happen. It is happening, you just don't know how it's happening. You have to find out how it's happening. And we're going to be able to kind of see. And it's really this nexus of things, this nexus of analytics, of good quality data, of foundational models, and of
things like process mining. To me, that is the real magic is to be able to understand what your processes really are, how long they really take. Because I think we're going to see a whole lot of optimization on actual
run time workflow with things. I think that's the big benefit that's going to come in the next couple of years, is this all sort of comes together, is we're going to figure out there are whole masses of work that we're doing that are just not necessary and shouldn't be done and we shouldn't be wasting
time on things like that. I think that's going to be one of the hallmarks out of this whole pandemic era is we all figured out quickly there are things that have to get done and there are things that do not have to get done. And when you're trying to get the business afloat or trying to stay afloat during difficult times. You do not want to be wasting time at all. So you want to be leveraging every tool in your toolbox. You
want to be leveraging every piece of data that you have. And I think that is going to be the hallmark of the future is being able to piece all of this together. And that's what these foundational models are good at, right, They're good at fusing vectors of information to create something new. And I think we're going to see an interesting array of solutions come out from all of that. But folks, you've been listening to Inside Analysis recording stop Legacy
ten fifty AM, Southern California, NBC Radionews dot com. Tahibo te Club's original pure powered Arco Superta helps build red corpusals in the blood, which carry oxygen to organs and cells. How organs and cells need oxygen to regenerate themselves. The immune system needs oxygen to develop, and cancer dies in oxygen. So the t is great for healthy people because it helps build the immune system, and it can truly be miraculous for someone fighting a potentially life threatening disease
due to an infection, diabetes, or cancer. The tea is also organic and naturally caffeine free. A one pound package of tea is forty nine ninety five, which includes shipping too order, please visit Tihebow Tea Club dot com. Tihebow is spelled tea like tom A h E E B L boyo. Then continue with the word t and then the word club. The complete website is to hibow Tea Club dot com or call us at eight one eight six one zero eight zero eight eight Monday through Saturday, nine am to five pm
California time. That's eight one eight six one zero eight zero eight eight tohebow Teclub dot com. Now here's a new concept, digital network advertising for businesses. Display your ad inside their building. If a picture is worth a thousand words, your company has going to thrive with digital network advertising. Choose your
marketing sites or jump on the DNA system and advertised with all participants. Your business ad or logo is rotated multiple times an hour inside local businesses where people will discover your company. Digital network advertising DNA a novel way to be seen and remembered. Digital network advertising with networks in Redlands and KAIPA call in the nine on nine area two two two nine two nine three for introductory pricing.
That's nine O nine two two two nine two nine three for digital network advertising. One last time Digital network advertising nine O nine two two two nine two nine three. With sixty years of fascinating facts, this is the Man from Yesterday and back in time. We go to this time in nineteen sixty five. Because of the success of ABCTVS Bewitched this past season, TV's copycats are concocting at least three new series with supernatural beings. The one that's on everyone's
mind is from the same folks who brought Bewitched. Screen Gems is going to present Eye Dream of Genie this fall, and from this time in nineteen seventy two, looks like CBSTV is going to put the Waltons on the fall schedule. It's about the life of a rural family in the South, set about nineteen twenty nine, and from this time in nineteen eighty three, Sesame Street releases a rock album titled Born to Add The Jacket is a takeoff of Bruce
Springsteen's phenomen Warned to Run album for kids like that. We were born with Moore at Man from Yesterday dot Com. It's time to make the tri City Center in Redlands a regular part of your weekly shopping experience. Tri City is home to a wide assortment of quality businesses, including the all new Ocean Aquatics. Check out their variety of exotic tropical fish along with fish food accessories and tanks of all shapes and sizes. The Tri City Center is located just off
of Alabama and the Tennessee exits in Redlands. Visit the Try City Center today and find out why it's called the mall with a heart. You've eaten lots of great food and lots of great food at restaurants. Cowboy Burgers in Fontana and now on Arlington and Riverside will fast become one of your favorites with their delicious, mouth watering burgers and breakfast spurn. Cowboy Burgers and Barbecue also serves fantastic smoke barbecue, baby back ribs, dry tip chicken, pulled pork sandwiches,
as well as lunch and dinner plates. Everything is made from scratch, including their delicious side dishes like cole slap, potato salad, barbecue beans, and much much more. Check out their rich, decadent chocolate brownies. Hi, I'm food critic Allenborgan, and you can dine in, take food out, or have them cater your next special event. I highly recommend Cowboy Burgers and Barbecue at their new location at five five seven three Arlington Avenue in Riverside.
Just look them up on the Internet. That's Cowboy Burgers and Barbecue, Happy eating and perfect for the holidays. Cowboy Burgers and Barbecue is also available for catering. That's Cowboy Burgers and Barbecue in Fontana and now in Riverside on Arlington. Do you love to dine out and travel, then join me the number one food critic, Allenborgan, host of the Let's Sign Out TV Show
on PBS every Saturday at four pm. For the new Let's Sign Out Radio Show on CACAA ten fifty am, along with my costs Bobusi, we will be reviewing local restaurants, interviews with culinary professionals, review new cooking products, cookbooks, plus so much more, including giving away free gift certificates. That's the new Let's Sign Out Radio Show right here on CACAA ten fifty am every Saturday at four pm, Happy Eating. For several years, KAA has been
marketing the Youngevity brand of nutritional and personal care products. Our experience with Youngevity has been one hundred percent positive, so we are pleased to recommend them to you. Regarding nutritional supplements, we recommend pollen Burst in the berry flavor and tangy Tangerine two point zero in the tablet form. For regularity issues, we recommend three day cleans, and for personal care, we recommend morning hydration cream.
You can shop online for Youngevity at www dot KCAA team dot com, or you can order by phone by calling eight hundred and nine eight two three one nine seven and tell customer support that you are part of the case AA team. Youngevity is an American company based in San Diego. Call Youngevity at eight hundred nine eight two three one nine seven and ask about monthly autoship that allows you to buy Youngevity products at wholesale prices. That number again eight hundred
nine eight two three one nine seven. CACAA Radio has openings for one hour talk shows. If you want to host a radio show, now is the time. Make CACAA your flagship station. Our rates are affordable and our services are second to nine. We broadcast to a population of five million people plus We stream and podcast on all major online audio and video systems. If you've been thinking about broadcasting a weekly radio program on real radio plus the Internet,
contact our CEO at two eight one five nine nine ninety eight hundred. Two eight one five nine nine ninety eight hundred. You can skype your show from your home to our Redlands, California studio, where our live producers and engineers are ready to work with you personally. A radio program on CACAA is the perfect work from home avocation in these stressful times. Just TYPEKCAA radio dot com into your browser to learn more about hosting a show on the best station in
the nation, or call our CEO for details. Two eight one five nine nine ninety eight hundred. Listinga CACAA Litlinda at one oh six point five FMK two ninety three CF Burrito Valley, NBC News Radio. I'm Chris Garagio. The massive heat wave continues across the western US. The National Weather Service as Phoenix reached one hundred and eighteen degrees Saturday, shattering the previous daily high record of one sixteen set back in two thousand and six. It was also the
twenty third consecutive day of temperatures above one ten in Phoenix. A record breaking temperature of one oh six was reached in Salt Lake City yesterday. Excessive heat warnings are being extended across the Southwest at least through midweek. At least six people are dead following a series of weekend shootings in the Chicago area. The violence began Fry, when a man was fatally shot during an armed robbery and
found inside a vehicle in an alley. Since then, more than twenty others were injured in the shootings, with the victims ranging from the