 From theCUBE Studios in Palo Alto in Boston, bringing you data-driven insights from theCUBE and ETR. This is Breaking Analysis with Dave Vellante. The AI heard around the world is put the machine intelligence sector back in the spotlight. When you squint beyond the press hype, the data shows that artificial intelligence is now the number one sector in terms of relative spending velocity. Normally market hype leads spending momentum, but the data suggests that spending activity and market penetration seem to be aligning with the hype. Now, while hyperscale cloud players are reaping the rewards, we think this is a rising tide that's going to lift all AI ships. Those both plainly in sight and others that may not be so visible. Hello and welcome to this week's Wikibon Cube Insights, powered by ETR. In this Breaking Analysis, we dig deeper into the AI space with spending data from ETR and one of the best minds in tech generally and AI specifically, Jeff Jonas, CEO, founder and chief scientist at Sensing. Jeff, good to see you, my friend. Hey, Dave. I don't know if you remember. It's been a while since we hung out. I mean, I've probably seen you, but last time we hung out was in New York City. It was after like a big day to NYC. We went to Spark Steakhouse and afterwards we had some drinks and so forth. We were recreating the John Castellano murder. I still have that photo. I'm laying on the ground. Yeah, you're in the chalk line. You were there, living through my photos. I'm like, wait a minute. You got to hang out more. You got to hang out more, man. Instigator. Well, thanks for doing this. Let's get into it. I'm really excited to be spending some time with you and sharing with our audience. So let's look at the sector momentum and spending and the pervasiveness. We got the survey of about 1700 IT decision makers. And after dominating the taxonomy during COVID, spending momentum dropped below the magic 40% mark. So this chart shows in the vertical axis, spending velocity or net score and it bottomed on October of 2022 is back on top. And then the horizontal is pervasion or presence in the data set, basically the end within a sector divided by the total end. And so you can see that the squiggly line kind of shows where AI was, it sort of was up top, dropped down below the 40% and now it's back. And so Jeff, the industry surrounding AI and machine learning obviously grown rapidly, significant hype around it, particularly in the case of generative AI with chat GPT. But given all this, how do you envision separating genuine progress and applications from inflated promises and the over expectations in the market? Where do you see this going? That's going to be huge, man. I think that's going to be a rocket ride. I think that little line that we see, there's just going to continue to just, it's just going to continue to zoom. I think, you know, after Watson beat Jeopardy in however long ago, it captured the world's imagination and tens of billions of dollars went into this, into the field of AI. It's going to be a hundred times that big. And the utility of LLMs is going to be, it is phenomenal. It's overestimated, but it's still huge. So I think it's straight up from here. Why do you think that IBM Watson didn't just run the table on this whole space? I think people over imagined all of his utility. I think that original algorithms, the stack was really tailored to actually beat the game Jeopardy. I remember being in Singapore and somebody was telling me they wanted to use those algorithms to do fluid modeling for tsunamis over the surface of the Singapore landmass. I'm like, that is the wrong use for that. So I think those original algorithms had a certain set of utility, they're more narrow. And I think the value of LLMs in the range of utility is going to be spectacular, but probably half as big as people actually think. All right, let's get into some of that. So if we dig into the spending profiles from what we just showed you and then zoom in to AI. So chatGPT was a catalyst for change in this spending profile that we're showing here. So let me break it down, the colors. The lime green is new customers for AI platforms. The forest green says customers are spending 6% or more relative to last year. The gray is flat spending plus or minus 5% and the pink spending is down 6% or worse and the bright red, which is nothing, is churn. So when you subtract the reds from the greens, you get something called net score, which is the percentage of customers that are spending more. And you can see that soaring in the blue line. And a yellow line is that N in AI divided by the N in all sectors, that N of overall 1700. And it's on a very, very steep rise in terms of, you know, it bottomed in October and then that was like a month before chatGPT. So in terms of technical advancements, how do you see the development of AI and large language models, generative AI, moving beyond supervised learning and LLMs? What do you think is most promising in terms of areas of research in this regard and potential impacts that they have on society? I think the really big thing that's happening is integrating diverse data. You know, the fact that I can go to chatGPT and ask for a board agenda and then have it reduce it to a rhyme and then reduce that to a haiku is because it's been trained over such diverse data. When researchers in Africa found termite mounds that have natural climate control and then they started working with people working on skyscrapers and building energy efficient skyscrapers. And then they started new innovations break out of that. There's so many new innovations and breakthroughs coming from better connected integration of diverse data. And we're gonna see this in a lot of different texts, graph databases, of course, what we've seen in LLMs, vector databases, traditional machine learning. My, I'm the symbols section on AI for entity resolution, integrating diverse data about people. But it's this collection of technologies to integrate diverse data that I think's really, and it's gonna accelerate what's happening in personalized medicine, climate improvements to climate change. The graph, as you mentioned graph, graph is interesting because you get this expressiveness with the graph database, but the way in which you query it is still kind of like old. Do you see that changing where you're going to get the expressiveness of graph with the query flexibility of like SQL? I think the way humans interface with computers is about to radically change with these LLMs. Like, you know, in my particular case where you've got a entity resolution algorithm that's got JSON into JSON out. Now you can literally do conversational entity resolution and say, hey, can you load a record that looks like this? It'll automatically load it. It'll, why do these two records come together? And you can get the answer in plain English. Imagine, you know, and so you package that type of qualitative human language and you put that on top of graph databases on top of entity resolution on top of anything. You're not going to do Jeff, tab, tab, Jonas, tab, June, tab, 22nd. You're going to literally just say it or type it. And so how we interface with computers is going to be, by the way, I've been thinking about this. You know how sometimes you want to change the behavior of some software. So you're having to look in settings and then look through the tabs. You're just looking for the one switch. Okay, that's going away. We're literally going to be just like, can you make it not do that anymore? He goes, yeah, I got that for you. And you have just dug way in and changed one setting. It's so true. You take a screenshot, you know, and you have to follow the map. It's just, it's terrible. But to your point, cloud is code and code is now language. Yeah, this is going to happen. So I just, I keep kind of telling people how fast this is going to happen. I want you to imagine, you're talking to somebody that you want to work with. He like, hey, are you a Mac or an Apple person? They're like, oh, I'm never really into computers. Like, excuse me? They're like, yeah, I just don't do computers. You're like, they go, like, I put in the work. Like no emails, nothing, no computers. Like, yep, it's going to be exactly this way in like a year or maybe less. You're going to be like, hey, what kind of, you know, what kind of AI assistant do you use? You know, you use JGPT, Bing, Bard, what do you got? They go, oh, I don't use assistance. You don't use assistance. You replay the recorded one hour meeting and then rewrite the meeting notes and summarize it by hand. I'm like, yeah, I put in the work. You're going to be like, oh. I'm old school. Yeah, I'm old school. All right, let's move on. The big three cloud players, they're obviously benefiting, but their positions have changed since chat GPT was announced, which is what you're showing here. And of course open AI has stormed the castle. So this shows sort of AWS in the upper left, Microsoft upper right, and then Google in the lower left and then open AI. Sort of pre and post chat GPT and we put in sort of the key performance indicators. And you can see the positions have switched. Like Microsoft, you know, basically cut the line on spending momentum on AI, you know, with its relationship with open AI. Open AI now dominates the combination, at least for now. You see AWS, that momentum picked up since they announced bedrock in the spring, but they went from number one to number three in terms of the spending velocity. They're all pretty high. You know, Google is Google. So Jeff, my question is, so you look at open AI as GPT-3 and now GPT-4, a lot of questions about responsible AI, how it's got potential misuse of powerful AI models, particularly when they become commercially available and they hallucinate. How do you think organizations should be thinking about, you know, this tension between broad based access and potential misuse, especially as these models become more capable? Well, you definitely want to be careful not submitting data that's going to become, you know, live in somebody else's logs and then be used for training. And now you've got some of your proprietary data in there. So there'll be a, there needs to be careful work on that. And there'll be more remedies from organizations to have run their own local models or have their data protected when they're using public models. It's going to be very interesting about what data was used to train what entitlements that people have to use that data. Are they using data that they had the rights to use? And I think that's going to be really interesting. There will be some benefits to those that are, you know, have a wider range of data. When I asked Chad GPT about who I am, it confuses me with the Iron Man movie because I'm new Iron Man triathlons. I mean, it's damn convincing. Like it tells me I met with John Favreau, the filmmaker in Vegas and the think tank. And I'm like, did I, I didn't think tanks and I meet filmmakers that I meet him. But, you know, I did some more research and called some people to see if these claims about me were true and it's not. But here's the thing about hallucinations is if you're, if you're complaining about these things, having hallucinations, you're already using them the wrong way. They're not designed for truth. Like literally they're qualitative, not quantitative. My, my litmus test is, if you were to ask the generative AI tomorrow, the same question. And if you get a different answer, if that's a problem, you're already using it the wrong way. So qualitative is, it is its best use, not truths. Yeah. At one point chat, GPT said that I started the cube with Jason Calcanis, which was kind of funny. I don't even know him. I know him, I guess he doesn't know me. John knows him though. So he's connected through Furrier. So maybe that's how, you know, chat, GPT, you know, hallucinated. But to your point, you don't use it for that purpose. You use it for ideation. You use it for, you know, summarization and the number of other things that it's really good at. And do you think, now, over time, is there any reason why something like the chat GPT of today can't go in that direction and be something that is designed to actually maybe ask questions if it doesn't know or basically hedge more. It's going to evolve, obviously. It's generative though. It's bones. In its bones, it's generative. I think that definitive paper on this was written by the Wolfam Alpha guy. He wrote like a 42 page paper on it. And I really love that paper because it kind of describes how it works. But it basically, you know, it already commits to a bunch of words. On Thursday, they went to the beach to, you know, then just like swim, sunbathe and it just rolls the dice to pick the next word. And then it goes, okay, I committed to Sunday. Now what? That's what generative does. And it's phenomenal. But that's, you can't use that for things that you need the same answer tomorrow. That's not the way to use it. And it doesn't have any source. You can't ask it about where did it get the information. When I ask it about all this Jeff Jonas Ironman Conflation, it finds quotes that I said in Wired Magazine or a blog post that I never said. It gives you links that aren't true. There's, it's sourced lists. You can't get attribution for it. And so again, I sound like I'm whining, but it's phenomenal. It's going to do all kinds of things. But if you need to get this, if you're wishing to get the same answer tomorrow and you needed explainable, then it's not going to be the LLM type of technology. It'll be something else. I mean, and just to the, you know, what we're doing, we have an AI we've been working on now for a long time that does interview resolution figures out who's who real time self-learning. They get the same answer tomorrow and it's explainable. So I don't know, it's going to get inserted. Things like this, you know, Wolfram Alpha gets inserted into chat GTP so it can do math. I think entity engine sensing or whatever will be used to call them anti-psychotics into these LLM. So we'll see things maybe integrated with them. And then these LLMs will put flowery words around it. But don't underestimate what else is coming. You know, I'm just saying the LLMs themselves that lose attribution, they don't know really the source are only going to do so many things themselves. That spells huge opportunity to me, you know, especially in the enterprise where you do need things that are explainable and with all the other enterprise features and characteristics that we know and love. So it's not just about the big three clouds and open AI, you know, as we're showing here, there are many others. You got players like Databricks have been visible. They just made that big acquisition and sort of turned a weakness into a strength a couple of weeks ago at their big conference. You got guys like Anthropic, Hugging Face is not shown here, but you know, they're prominent. You got guys like Data Robots or a lot of these AI specialists. Of course, IBM and Oracle, you know, driving into their own products. IBM of course, you know, with Watson but a lot of folks that are pushing AI into their platforms they don't necessarily show up in the spending data. So Jeff, with several companies and organizations working on similar AI technologies how do you think firms can differentiate their AI products or solutions from others in the market? What's going to be that unique value prop that firms should offer that's going to set them apart? Man, that is going to be tough. I'm speculating right now that there are companies out there that are getting term sheets but they're going to be out of business before they get their first around. Like that's how fast it's moving, you know? Warren Buffett calls it the moat. It's going to be hard to have moats in a lot of cases. You got to build things that everybody else doesn't have. Folks that think you can take three public, you know, kind of publicly available things wrap them together and call it unique. It's not unique. So if you want to have something unique you have to have, you better have real work. You have to have something that's really different and that's hard for others to replicate. So that is the big question. And I'll tell you the VCs as an LP of four funds you got to be really careful what you fund because things, they just, how are they going to be different? There's a lot of mystery around that, especially, you know, you can add an LLM, anybody can add an LLM to anything right now. So then you better have something different that's not the LLM. And by the way, lastly, you better add an LLM onto your widget as soon as you can or you're not going to be very interesting because if a company has money to spend, they're going to spend it on things that have AI and LLMs related to them versus something else because they've got to report it up to the leadership who's going to report it up to the board because the board's asking the leadership, how are we investing in this? So you got to get into the wave. So is data going to be that differentiator or is that going to be, you were talking before about IP leakage essentially but do you feel like that corpus of data is ultimately going to determine and define the moat? I think it's going to make a big difference. You know, when I ask a thing about who I am because they have, maybe it's because they have LinkedIn but it's a quality of understanding that is so much better. I do think, you know, he who has the most knowledge and can harness it is going to have the best advantage. And then how is that going to be liberated that data so that others can build those into their systems and what's the pedigree of that data? It's going to be a very interesting world coming. It's exciting. So it sure is. So Furrier and I talk about this a lot on our Q pod is, you know, what you try to compare with other waves that they're never identical and past is not prologue but you think about the internet, it benefited a lot of incumbents and yeah, there was a lot of disruption too but everybody was able to take advantage of it. How do you see this wave relatively? I think this wave is going to be 10 times bigger than the Watson wave. I mean, sorry, a hundred times bigger maybe a thousand times bigger. There are so many exciting ways to integrate LLMs into systems to create a better experiences for organizations using technology. So yeah, that's it man. I just, it's going to show up everywhere by the way and you don't even have to do it yourself. I mean, it's just going to be in your office. I mean, it's going to be in the app you're using, you know what the office is going to be in these tools you use, it's just being integrated, integrated left and right and the world's going to become more conversational. I think, you know, a weird thing is like cursive. You know, they don't teach cursive anymore in school and you have to wonder when things become really natural language oriented. You know, at what point do keyboards go away? Yeah, typing with our thumbs is probably not going to be the norm. You kind of referenced this earlier, that you know, you can jump on board. You got to apply LLMs like now, yesterday, months ago. So how should leaders think about investing in AI? What would your advice be? Well, what would be my first advice on that? Well, you know, there's a lot of, a lot of people are labeling things AI and it's a bit of a stretch, you know? So one thing you want to do is probably find that it's legitimate. You want to make sure that it's going to work within your compliance and you're like having repeatable answers. You want to use it in ways that are going to be responsible. You're going to have to be able to answer to the decisions that you're making. I think if it gets used, if LLMs get used qualitatively to package up and put the words around the messaging to the customer around the advice, it's not the actual core of the advice. I think it's probably a bit safer. And I think a lot of organizations are going to get their AI through partners that are specializing in AI. I mean, I'm trying to outsource entropy. You know, the second law of thermodynamics, the world's trying to break into small pieces, spread out and cool off. And you have to be careful where you spend your energy. We spend our energy, the food we eat and organizations to like hold things together. And why do organizations, no one's out there got an army of people doing AI on spell check and grammar checker. You plug in an API. That's the kind of thing I'm trying to do for entity resolution, who's who. And I think organizations are going to find lots of technology components that have AIs built in them that they can integrate into the workflows that they have or the products that they're building for the market. You know, you mentioned entropy. You think about security and randomness of data. You probably, I presume you could use AI to sort of measure essentially the degree of randomness in data to identify potential malware as just one of about eight billion examples. But do you think, I mean, the security guys seem to want to build this in themselves. I don't know, maybe bolt it on in a lot of cases. Well, if you can buy something and plug it in and like most organizations aren't building their own credit card settlement, you'd use Stripe. They're most organizations aren't building their own mobile comms, they would use a Twilio. So I definitely think, you know, you try to have some outsourced the entropy and have somebody else specialist, you know, plug it in. Hey, this thing about anomalies though, in big data, things that are rare, like one in a million happen a million times a day. So rare doesn't always mean interesting. And by the way, entropy might be winning. I don't know, like as I think about it, you know, these LLMs, they're creating more data. The more data I blogged about my thing with the Iron Man and then LLMs are going to end up reading that. And then they're going to insert it into their knowledge base. We're going to need a clear demarcation line between data that was like generated by humans and actual sources versus data just purely generated. You know what, that data will be what? February something 2023. Clear line, let's say on what data existed before that. You better know, man. You use second law of thermodynamics, doesn't it say entropy increases over time? You might be right. Yeah, that's what I mean. It's winning. It is, you know, I used to think in this entity business, you're trying to figure out who's who to, you know, like, hey, they're all the same people. Decades ago, I'm like, well, you won't need to do this for very long because everybody's going to speak common keys. Everyone shares the key. There are now more keys than ever. You have more handles about you. Entropy is winning. The universe has taken a tremendous way with it. I have to retitle this breaking analysis, entropy is winning. All right, last question. If you had to start a company today and this is your chance to plug sensing, but I would say other than the one that does entity resolution although you can work that into your answer, how would you go about thinking about what to start? Well, I would first say that you want to build a capital efficient company. This idea of just throwing more money on and going with entity metrics, how many people you have, how much money you've raised and growth at all cost. I think it's just basically out the window. I just don't think that's the future. I think the idea to start a company, you got to figure out unit economics and you have to be able to make money along the way and early. And I see some companies that are really good product people at marketing people and they figure out how to market anything. And you see some companies have really good engineering technology, but they know how to market it. And there's really the pairing of the two. If you can't figure, you can have an ATEC and not a great go to market and then you're just going to be one long. So anyway, capital efficiency and you probably have to be as good as marketing as you are at your tech. Well, to your point, I mean, the term zombie corns has come about, people were all celebrating, we're a unicorn, we're a unicorn. It's like they can't get their next round and there's somebody, you know, three kids who just got laid off from whatever company or disrupting them and it's all getting compressed so quickly, it's amazing. It is, we're at a golden age. This is so exciting. Jeff Jonas, you know, we got to hang out more. Love having you on theCUBE. You're an amazing guest. So thanks so much for spending some time with us. It's always fun. All right, I want to also thank Alex Meyerson who's on production, manages the podcast. Ken Schiffman too. Kristen Martin and Cheryl Knight, they helped get the word out on social media and our newsletters and Rob Hof is our editor-in-chief over at SiliconAngle.com, does some great editing. Remember, all these episodes are available as podcasts wherever you listen to search breaking analysis podcast. I publish each week on Wikibon.com and SiliconAngle.com. David.Valante at SiliconAngle.com. If you want to get in touch with me or at dValante for DM, talk to me on LinkedIn on our post and check out etr.ai. They got some great survey data focused on the enterprise tech business. This is Dave Valante for theCUBE Insights, powered by ETR. Thanks for watching and we'll see you next time on breaking analysis.