The Innovation Civilization Podcast
00:30:44 · George Sivulka, Waheed Nabeel
[00:30:44] S1: Oh, okay. That's a good one. I would say one thing would be the geography of data centers. Okay. Where data centers are is very important. I think that the Trump Sam Altman $500,000,000,000 investment project Stargate is... Was really just a bit of a nothing burger where, yeah, obviously, everyone, they're probably gonna make that investment anyways, etcetera, etcetera. But as data centers and access to power, I e, like electric power to power those data centers becomes more and more scarce, the location of data centers and the geopolitical implications of the location of those data centers will become increasingly important in terms of who controls compute and who controls effectively AGI. So that'll be interesting. And then another thing that I think is a really interesting trend is training scaling laws. So the ability to have more data and more parameters and to have a more accurate model come out are very interesting. But inference scaling laws, which is the ability to run more models and have more accuracy, is a new type of scaling law that has been powering a lot of paradigms like test time compute that was pioneered at Hebia with our matrix product. And I think that will be a massive unlock for more accurate and more able and capable AI systems. [00:31:58] S0: And so I'm not an expert on this, but I've basically heard that NVIDIA is great at the former, but not the latter, which is inference. So do you feel like the inference market has a lot more players to come, basically, in that space?
00:31:58 · Waheed Nabeel, George Sivulka
[00:31:58] S0: And so I'm not an expert on this, but I've basically heard that NVIDIA is great at the former, but not the latter, which is inference. So do you feel like the inference market has a lot more players to come, basically, in that space? [00:32:14] S1: Yeah. I think that long term inference will be done on the edge. It'll be done on your computer or your iPhone or your Android. And I think that NVIDIA has a massive stranglehold on the training market, but probably less so on the on the inference market. [00:32:29] S0: Yeah. Okay. That's pretty cool. And last question, basically. So you've been following the the global headwinds that are happening around the world where it feels like I'm not just talking about the Trump tariffs and the effect on markets, but I'm more talking about the supposed retreat of The United States from global trade, so to speak, and just leaving people be in each of the countries be and changing an order that was established at Britain Woods, like, the nineteen forties and so on and so forth. Given what's changing in the world today, basically, what's your kind of feeling about either where the world is headed or just building companies in general in AI? I know it's a very broad question, but, yeah, any sort of thoughts? Have you been thinking about it, or do you not think about that at all, basically? [00:33:23] S1: I do. I I think it's really important to contextualize, like, the business implications of geopolitics as they unfold. It's a great time to be building a company in America, and I think that [00:33:33] S0: Golden age.
00:33:33 · Waheed Nabeel, George Sivulka
[00:33:33] S0: Golden age. [00:33:34] S1: Hopefully. It hopefully will be. But at the same time, I think that there's definitely going to be a variety of different stressors, especially we're a multinational corporation now and a global business. And I think that I really do care that no matter what happens, we can act quickly and and serve as many clients as we possibly can. Because my ultimate goal is to build a business for not for The United States or for other countries, but for all of humanity. And I think that... My my hope is that what... Whatever the global order unfolds into, we'll be able to do that. [00:34:05] S0: Yeah. I hope that you guys have got a tariff calculator agent you're shipping to. We have. I do. We're good. [00:34:11] S1: We we have really good ability to actually look at a company's portfolio or an investment firm's portfolio and impact and and calculate the impact of tariffs Yeah. On on that portfolio, which is quite interesting as an agent. [00:34:24] S0: Yeah. That's pretty cool, actually. One of the other things that's interesting to me is right now, there feels like people just pumping money into AI. You mentioned about Stargate project and so on and so forth. It feels a bit bubbly when it comes to AI investment. When you talk to your customers, like, how do you guys, like, measure value from... Or ROI from your AI investments, basically? What's... What are they thinking?
00:34:24 · Waheed Nabeel, George Sivulka
[00:34:24] S0: Yeah. That's pretty cool, actually. One of the other things that's interesting to me is right now, there feels like people just pumping money into AI. You mentioned about Stargate project and so on and so forth. It feels a bit bubbly when it comes to AI investment. When you talk to your customers, like, how do you guys, like, measure value from... Or ROI from your AI investments, basically? What's... What are they thinking? [00:34:46] S1: Yeah. I think there's a lot of people that believe that AI is a bubble right now because valuations, prices, amount of spend have all gone up, an amazing amount Tremens found in the last in the last few years, especially since the launch of ChatGPT. At the same time, we have customers that are really upselling and increasing the amount of Hebiah that they buy, which is our... Which is starting to show that, hey, there's actual real value that's coming out of these firms. Because the only... [00:35:13] S0: How do you charge them? Is it by seat? Or... [00:35:15] S1: Yeah. It's by seat. [00:35:16] S0: By seat. Okay. Cool. Sorry. [00:35:18] S1: It's... So it just goes to show that there's real value that that companies and customers are are realizing with AI. But you'll start to see that that shakes out across the industry. There's firms that that's not happening for, and it... The time will come where it's obvious and a leaky bucket can't fill. [00:35:32] S0: Okay. But there is no measurement of... Or ROI kind of metrics you guys are tracking or just generally?
00:35:32 · Waheed Nabeel, George Sivulka
[00:35:32] S0: Okay. But there is no measurement of... Or ROI kind of metrics you guys are tracking or just generally? [00:35:39] S1: Oh, we have users that are saving eight to twenty hours a week self reported. So that's almost [00:35:45] S0: half of [00:35:46] S1: a week per person per seat of Hebiah. That's good. We have processes that used to take a hundred twenty to two hundred hours that are now taking ninety to a hundred fifty hours on average, which are massive votes of confidence for time saved. But the real unlock is actually something that's impossible to quantify, which is that if it makes you a better investor, a better banker, a better lawyer, a better employee, the upside, I can't really say, okay. Or the only reason your assets are going up x percent or you were able to call the effect of these tariffs. [00:36:18] S0: Yeah. [00:36:18] S1: But we might be a really big part of that, and that's unquantifiable value. [00:36:22] S0: Yeah. Yeah. Yeah. I think you mentioned this before, but it was like the benefits of A. I. There's twofold, right? One is, Okay, how faster, better, cheaper can I do my current job better? Right. Or current tasks. And then there's the other part of what new things can I do, basically, with the additional capabilities? Right. So I guess what you're saying is that the second, the latter part, the new things is still unquantifiable right now. [00:36:50] S1: And that's really where you'll see the maximum amount of value grow. And that's an important vector for expansion.
00:36:50 · George Sivulka, Waheed Nabeel
[00:36:50] S1: And that's really where you'll see the maximum amount of value grow. And that's an important vector for expansion. [00:36:55] S0: Okay. Cool stuff. Brilliant. George, thank you so much for your time despite of your health today for joining us. You've been a great guest, and thank you so much, and all the best with Hebiah. Looking forward to an IPO soon. [00:37:08] S1: Oof. Knock on wood. I'm knocking. But, no. We've got a long way to go and a lot of software to build, and really great to to have this discussion with you. Yeah. Thank you so much for the time. Cheers. [00:37:24] S0: Thank you very much for listening to the Innovation Civilization podcast. If you love the podcast, please subscribe on all major platforms as well as please share it with your friends and family. Thank you very much for listening, and see you soon for the next episode.