1 00:00:00,280 --> 00:00:03,640 Speaker 1: AI investments still on fire, but there is growing concern 2 00:00:03,680 --> 00:00:05,920 Speaker 1: that it may be a bubble. And there's also a 3 00:00:05,960 --> 00:00:07,560 Speaker 1: sense that maybe we're seeing a few things that are 4 00:00:07,600 --> 00:00:10,600 Speaker 1: quite similar to the dot com boom bust, vendor financing, 5 00:00:10,800 --> 00:00:13,480 Speaker 1: artificially inflated demand and so on. Sam Dickey from Fisher 6 00:00:13,480 --> 00:00:16,919 Speaker 1: Funds is with us. He Sam, what exactly is vendor 7 00:00:16,960 --> 00:00:18,800 Speaker 1: financing and why should investors care about this? 8 00:00:20,640 --> 00:00:24,680 Speaker 2: Yeah, it's quite normal in many industries, so when supplies 9 00:00:24,760 --> 00:00:28,160 Speaker 2: lend money to customers specifically to buy their own products, 10 00:00:28,240 --> 00:00:31,560 Speaker 2: essentially paying customers to be customers. And it's often used 11 00:00:31,560 --> 00:00:35,599 Speaker 2: in small business sales or equipment sales when the buyers 12 00:00:35,640 --> 00:00:40,520 Speaker 2: otherwise can't get credit easily. However, it is also used 13 00:00:40,600 --> 00:00:45,159 Speaker 2: at big turning points in technology when massive, bold investment 14 00:00:45,240 --> 00:00:48,760 Speaker 2: is required, and when suppliers start leading billions of dollars 15 00:00:49,120 --> 00:00:51,320 Speaker 2: to customers so they can buy their own products. It's 16 00:00:51,440 --> 00:00:55,280 Speaker 2: arguably artificial demand, and history suggests we should stand up 17 00:00:55,280 --> 00:00:55,920 Speaker 2: and pay attention. 18 00:00:56,520 --> 00:00:59,640 Speaker 1: Now, how does today's AI vendor financing compare to the 19 00:00:59,680 --> 00:01:00,520 Speaker 1: dot com bubble? 20 00:01:01,520 --> 00:01:05,280 Speaker 2: Some similarity? So in two thousand, let's use Lucent as 21 00:01:05,319 --> 00:01:09,520 Speaker 2: the poster child. It made networking in telco equipment required 22 00:01:09,520 --> 00:01:12,360 Speaker 2: for their early internet and mobile boom, and they and 23 00:01:12,400 --> 00:01:17,000 Speaker 2: others were lending to large telcos but primarily small startup 24 00:01:17,080 --> 00:01:20,440 Speaker 2: telcos and internet service providers to stoke demand for their product. 25 00:01:21,520 --> 00:01:25,119 Speaker 2: And the thing was those small telco customers had were 26 00:01:25,120 --> 00:01:29,680 Speaker 2: burning cash and had stretched balance sheets. Now today in Vidia, 27 00:01:30,040 --> 00:01:33,120 Speaker 2: AMD and oracle It are funding customers like open Ai, 28 00:01:33,520 --> 00:01:36,520 Speaker 2: the owner of ch GBT, and Corewe who's a sort 29 00:01:36,560 --> 00:01:38,880 Speaker 2: of data center provider, to buy their chips or space 30 00:01:38,880 --> 00:01:41,800 Speaker 2: and their data centers. So definitely some similarities. And for context, 31 00:01:42,600 --> 00:01:45,480 Speaker 2: in two thousand, Loocent and Co lean around twenty five 32 00:01:45,480 --> 00:01:48,680 Speaker 2: billion dollars, which was about one hundred and fifty percent 33 00:01:48,720 --> 00:01:51,120 Speaker 2: of their earnings back then. Today the number is greater 34 00:01:51,120 --> 00:01:54,240 Speaker 2: than one hundred billion, but it's a smaller percentage of 35 00:01:54,240 --> 00:01:57,400 Speaker 2: earnings because the balance sheets and cash flow generation of 36 00:01:57,440 --> 00:02:00,960 Speaker 2: the companies today, both on the on the vendor side 37 00:02:00,960 --> 00:02:04,320 Speaker 2: so in Video and Co, but also their big customers 38 00:02:04,320 --> 00:02:07,240 Speaker 2: like Meter and Google are significantly healthier than Loosen to 39 00:02:07,280 --> 00:02:10,480 Speaker 2: the startup customers, so it's not nearly as severe as 40 00:02:10,520 --> 00:02:13,720 Speaker 2: two thousand yet, And at one point in two thousand, 41 00:02:13,760 --> 00:02:17,040 Speaker 2: just for context, Loosen wasn't even selling equipment anymore. It 42 00:02:17,080 --> 00:02:18,960 Speaker 2: was just giving stuff away on credit and calling it 43 00:02:19,000 --> 00:02:21,680 Speaker 2: a sale. So there was some fraud back then as well. 44 00:02:21,800 --> 00:02:23,400 Speaker 1: Yeah, what are some of the warning signs that we're 45 00:02:23,440 --> 00:02:23,920 Speaker 1: seeing today. 46 00:02:25,080 --> 00:02:27,440 Speaker 2: I think it's apart from the circular nature of this 47 00:02:27,560 --> 00:02:30,000 Speaker 2: the money merry go round, it's the sheer scale of 48 00:02:30,040 --> 00:02:32,440 Speaker 2: the deal. So you and I talked last week about 49 00:02:32,440 --> 00:02:35,280 Speaker 2: the staggering deal where open ai signed a contract to 50 00:02:35,280 --> 00:02:39,520 Speaker 2: pay Oracle three hundred billion dollars over five years. Yet 51 00:02:39,560 --> 00:02:42,520 Speaker 2: open ai itself only has fifteen billion dollars in revenue 52 00:02:42,520 --> 00:02:45,280 Speaker 2: in total today and it just signed up to pay 53 00:02:45,320 --> 00:02:47,800 Speaker 2: one of its supplies three hundred billion dollars. And the 54 00:02:47,800 --> 00:02:51,280 Speaker 2: other one is AMD, which is sort of the big 55 00:02:51,560 --> 00:02:53,600 Speaker 2: LaGG a out or the pull man's and Video they 56 00:02:53,639 --> 00:02:56,840 Speaker 2: signed away ten percent of their equity to open Ai 57 00:02:56,880 --> 00:02:59,120 Speaker 2: as a customer, just so open Ai would buy lots 58 00:02:59,160 --> 00:03:00,799 Speaker 2: of its chips, so try and catch up to in 59 00:03:00,919 --> 00:03:04,000 Speaker 2: video and that. The final one to keep an eye 60 00:03:04,000 --> 00:03:05,720 Speaker 2: on is the fact that a lot of the credit 61 00:03:06,280 --> 00:03:09,480 Speaker 2: or the loans are being backed or collateralized by these 62 00:03:09,560 --> 00:03:13,320 Speaker 2: same Ai chips, which actually is quite reminiscent of how 63 00:03:13,400 --> 00:03:18,480 Speaker 2: Loosens customers used overinflated telecom spectrum licenses back in the day. 64 00:03:18,560 --> 00:03:20,560 Speaker 2: Is collateral for the debt? Right? 65 00:03:20,960 --> 00:03:23,960 Speaker 1: Well, what does this mean for investors? 66 00:03:23,960 --> 00:03:26,480 Speaker 2: No doubt that some frothy scigns, and even today you 67 00:03:26,520 --> 00:03:29,400 Speaker 2: saw the Bank of England and the IMEF warning people 68 00:03:29,440 --> 00:03:33,400 Speaker 2: about the risks of AI. What we don't know is 69 00:03:33,400 --> 00:03:37,119 Speaker 2: how much longer this exuberance can continue for and for now, 70 00:03:37,200 --> 00:03:39,680 Speaker 2: and this is important. The primary customers of these AI 71 00:03:39,800 --> 00:03:43,600 Speaker 2: chips are companies with incredibly strong balance sheets and cash generation, 72 00:03:43,720 --> 00:03:46,080 Speaker 2: and that is quite different than two thousand. But we 73 00:03:46,640 --> 00:03:47,840 Speaker 2: do need to keep an eye on this head. 74 00:03:48,160 --> 00:03:50,280 Speaker 1: Yeah, interesting stuff, Sam, Thank you very much, appreciate it. 75 00:03:50,280 --> 00:03:51,760 Speaker 1: Sam Dickie Fisher Funds. 76 00:03:52,360 --> 00:03:54,600 Speaker 2: For more from Heather Duplessy Allen Drive. 77 00:03:54,760 --> 00:03:56,160 Speaker 1: Listen live to news Talks. 78 00:03:56,200 --> 00:03:59,400 Speaker 2: It'd be from four pm weekdays, or follow the podcast 79 00:03:59,480 --> 00:04:00,480 Speaker 2: on iHeart Radio.