1 00:00:00,040 --> 00:00:02,000 Welcome back everyone. Today on the Joseph Carlson 2 00:00:02,000 --> 00:00:05,240 Show, Amazon, the company that competes with everyone, is once 3 00:00:05,240 --> 00:00:07,680 again competing with seemingly everyone. 4 00:00:08,080 --> 00:00:12,280 They launched Amazon Nova, our new generation of foundation 5 00:00:12,280 --> 00:00:14,760 models. This is a direct competitor to 6 00:00:14,760 --> 00:00:17,360 ChatGPT. It's advertised to have many 7 00:00:17,360 --> 00:00:20,360 different versions. You have Nova Micro, Nova Lite, 8 00:00:20,360 --> 00:00:24,400 Nova Pro, Nova Premiere, Nova Canvas, and Nova Real. 9 00:00:24,680 --> 00:00:28,080 This model has all different capabilities, and you can pick 10 00:00:28,080 --> 00:00:29,800 which one best serves your needs. 11 00:00:29,800 --> 00:00:32,880 But this is a clear attempt from Amazon to match Chat GPT's 12 00:00:32,880 --> 00:00:35,880 capabilities and surpass them with distribution. 13 00:00:35,880 --> 00:00:38,240 Amazon has deepened their partnership with Anthropic to 14 00:00:38,240 --> 00:00:40,440 accomplish these goals, and some of the things that they're 15 00:00:40,440 --> 00:00:43,320 accomplishing are pretty impressive even by modern AI 16 00:00:43,320 --> 00:00:45,960 technologies. Amazon looks like it's closed 17 00:00:45,960 --> 00:00:47,960 the gap with its closest competitor. 18 00:00:48,040 --> 00:00:51,760 The stock is now at 220, up another 2.39% today because 19 00:00:51,760 --> 00:00:54,320 investors are being reminded that Amazon is a dominant, 20 00:00:54,320 --> 00:00:56,560 aggressive combative technology company. 21 00:00:56,600 --> 00:00:59,400 Not only do we have the report that Amazon launched Nova 22 00:00:59,400 --> 00:01:02,400 competing directly with Chachi PT, closing the gap with 23 00:01:02,400 --> 00:01:05,480 generative AI, but we also have another report at the same time 24 00:01:05,480 --> 00:01:09,200 that Amazon's creating its own homegrown AI supercomputer will, 25 00:01:09,200 --> 00:01:13,000 quote, lower the cost of AI as it seeks to build an alternative 26 00:01:13,160 --> 00:01:15,680 to NVIDIA. So Amazon is now competing with 27 00:01:15,680 --> 00:01:18,800 Chachi BT at the same time that they're competing with NVIDIA. 28 00:01:18,800 --> 00:01:20,600 This is a reoccurring theme in investing. 29 00:01:20,600 --> 00:01:22,960 Investors forget about established companies and chase 30 00:01:22,960 --> 00:01:25,080 whatever is the flashiest object at the moment. 31 00:01:25,600 --> 00:01:29,160 In this case, investors have forgotten that Amazon is a 32 00:01:29,160 --> 00:01:32,080 dominant player in AI. We have examples here of them 33 00:01:32,080 --> 00:01:35,360 now using their AI analytic skills to partner with Moody's 34 00:01:35,480 --> 00:01:37,480 to make Moody's Analytics more powerful. 35 00:01:37,480 --> 00:01:40,440 During the keynote presentation, Jassy, the CEO of the company, 36 00:01:40,440 --> 00:01:43,800 goes through a litany of examples of how Amazon is using 37 00:01:43,800 --> 00:01:46,360 their generative AI and practical applications. 38 00:01:46,600 --> 00:01:49,520 Some of these are literally game changers for Amazon. 39 00:01:49,520 --> 00:01:52,440 They have incredible practical applications that are going to 40 00:01:52,440 --> 00:01:55,760 be implemented over the next year as Amazon continues to 41 00:01:55,760 --> 00:01:57,480 climb. I think it's good to put this in 42 00:01:57,480 --> 00:02:00,120 context. What is all of this AI stuff 43 00:02:00,120 --> 00:02:01,800 mean for the future of this company? 44 00:02:02,000 --> 00:02:04,760 How will it bolster Amazon's earnings capability? 45 00:02:04,960 --> 00:02:07,320 We're going to be going over all of it in this episode. 46 00:02:07,400 --> 00:02:09,080 Now, of course, we have a lot of news to get to today. 47 00:02:09,080 --> 00:02:13,000 We have Salesforce rising 9% on the day after reporting earnings 48 00:02:13,320 --> 00:02:15,920 yesterday. Why is the stock up 9% when the 49 00:02:15,920 --> 00:02:19,000 company reported earnings that are mostly in line and the 50 00:02:19,000 --> 00:02:21,720 guidance they gave was also mostly in line? 51 00:02:21,720 --> 00:02:23,880 Well, I'll give you a hint. It's another story that has to 52 00:02:23,880 --> 00:02:27,160 do with a changing AI narrative. We also have Texas Roadhouse 53 00:02:27,160 --> 00:02:29,960 receiving free advertisement from former NFL players. 54 00:02:30,080 --> 00:02:33,720 And we have United Health executive, not the CEO, but an 55 00:02:33,720 --> 00:02:37,840 executive of the company, Brian Thompson, that was shot dead 56 00:02:38,080 --> 00:02:41,400 near Manhattan Hotel. This was a premeditated targeted 57 00:02:41,400 --> 00:02:44,440 attack and we'll be looking at all the details in this episode. 58 00:02:44,440 --> 00:02:46,600 Now, as we jump into this episode, I first want to point 59 00:02:46,600 --> 00:02:50,040 out that if you try Qualtrim today, you'll get the entire 60 00:02:50,040 --> 00:02:52,600 month for free. We offer a free month trial 61 00:02:52,600 --> 00:02:55,480 through the month of December. You can join by going to Patreon 62 00:02:55,600 --> 00:02:58,840 dot com slash Joseph Carlson. Once you join the Patreon, 63 00:02:58,840 --> 00:03:01,960 Qualtrim is included along with a lot of other content like 64 00:03:01,960 --> 00:03:05,360 exclusive episodes, portfolio updates, all sorts of different 65 00:03:05,360 --> 00:03:06,720 goodies. So if you want to try that out, 66 00:03:06,720 --> 00:03:08,600 you can as well. There's nothing to lose now. 67 00:03:08,600 --> 00:03:11,560 We start off with the main story, which is Amazon's huge 68 00:03:11,560 --> 00:03:13,640 push into artificial intelligence. 69 00:03:14,000 --> 00:03:17,160 It seemed like it was going very slowly and then all of a sudden 70 00:03:17,160 --> 00:03:19,800 it's happening very quickly. But for those of us paying 71 00:03:19,800 --> 00:03:22,640 attention, this should not come as a surprise. 72 00:03:23,120 --> 00:03:25,720 When I look at Amazon, it's not a company that I hold in my 73 00:03:25,720 --> 00:03:28,840 passive income portfolio. So I don't hold Amazon in the 74 00:03:28,840 --> 00:03:31,000 tech category or the financial category. 75 00:03:31,400 --> 00:03:34,480 I own a lot of Amazon in my secondary portfolio, which is 76 00:03:34,480 --> 00:03:36,800 the Story Fund. The Story Fund's a portfolio I 77 00:03:36,800 --> 00:03:40,040 track primarily on my secondary channel, which is Joseph Carlson 78 00:03:40,040 --> 00:03:42,800 After Hours. Amazon is the second largest 79 00:03:42,800 --> 00:03:45,600 position. It's a $92,000 position with 80 00:03:45,600 --> 00:03:48,160 $34,000 in the green. This is one that I've been 81 00:03:48,160 --> 00:03:51,320 saying for over a year now is undervalued. 82 00:03:51,400 --> 00:03:54,640 I believe that Amazon has been perpetually undervalued for 83 00:03:54,640 --> 00:03:57,920 quite some time, especially coming out of 2023. 84 00:03:58,200 --> 00:04:01,320 This company was not getting a lot of credit, but it seems like 85 00:04:01,320 --> 00:04:04,120 that's starting to change and that's typically what happens 86 00:04:04,120 --> 00:04:06,560 with narratives. Stocks are driven by cash flows 87 00:04:06,560 --> 00:04:08,960 in the long term and narratives in the short term. 88 00:04:09,280 --> 00:04:12,040 Narratives can give you opportunities to buy into stocks 89 00:04:12,040 --> 00:04:15,200 at good prices when there's a negative narrative on a good 90 00:04:15,200 --> 00:04:17,720 stock that will generate good cash flows in the future. 91 00:04:18,000 --> 00:04:21,240 That was the case with Amazon. This is a company that had an 92 00:04:21,240 --> 00:04:25,080 ongoing narrative that it was not a major AI player. 93 00:04:25,080 --> 00:04:28,240 Over the past two years, Amazon has been answering that question 94 00:04:28,240 --> 00:04:31,000 of what are they doing in artificial intelligence? 95 00:04:31,400 --> 00:04:34,080 And part of their answer is Amazon Nova. 96 00:04:34,320 --> 00:04:36,920 They call it the new generation of foundation models. 97 00:04:36,920 --> 00:04:39,840 Our new Amazon Nova models are intended to help with these 98 00:04:39,840 --> 00:04:42,960 challenges for internal and external builders and provide 99 00:04:42,960 --> 00:04:46,120 compelling intelligence and content generation while also 100 00:04:46,120 --> 00:04:48,160 delivering meaningful progress on latency. 101 00:04:48,240 --> 00:04:50,320 Inside Amazon, we have around 1000 Gen. 102 00:04:50,440 --> 00:04:53,360 AI applications in motion and we've had a bird's eye view on 103 00:04:53,360 --> 00:04:55,880 what application builders are still grappling with. 104 00:04:55,920 --> 00:04:58,320 Our new Amazon Nova models are intended to help with these 105 00:04:58,320 --> 00:05:00,640 challenges for internal and external builders. 106 00:05:00,680 --> 00:05:02,280 That's a lot of very area they covered there. 107 00:05:02,280 --> 00:05:05,600 They're building all these Gen. AI applications and they put it 108 00:05:05,600 --> 00:05:08,680 into this model called Nova Nova's being broken up into four 109 00:05:08,680 --> 00:05:11,240 main categories. You have Nova Micro, the text 110 00:05:11,240 --> 00:05:13,800 only version that's very low cost, low latency. 111 00:05:14,080 --> 00:05:16,920 You have Nova Light. It's one step up, a very low 112 00:05:16,920 --> 00:05:20,040 cost multi model that is lightning fast for processing 113 00:05:20,280 --> 00:05:24,000 images, videos and text inputs. Then you have the step up from 114 00:05:24,000 --> 00:05:28,160 that Nova Pro, a highly capable multi model with the best 115 00:05:28,160 --> 00:05:31,360 combination of accuracy, speed and cost for a wide range of 116 00:05:31,360 --> 00:05:33,960 tasks. Then you have the most powerful 117 00:05:33,960 --> 00:05:38,080 1 Nova Premiere, the most capable of Amazon's multi models 118 00:05:38,080 --> 00:05:41,520 for complex reasoning tasks, for use of the best teacher for 119 00:05:41,520 --> 00:05:44,840 distilling custom models. It's going to be available early 120 00:05:44,840 --> 00:05:46,960 next year. And then you have Nova Canvas 121 00:05:46,960 --> 00:05:50,000 and Nova Reel. Canvas is for generating images, 122 00:05:50,000 --> 00:05:53,520 Reel is for generating video. So they also have those video 123 00:05:53,520 --> 00:05:56,560 and image generation as well. Overall, what Amazon created 124 00:05:56,560 --> 00:06:00,200 with Nova is a comprehensive model for generative AI. 125 00:06:00,280 --> 00:06:02,880 They had one part of the keynote where they actually put in the 126 00:06:02,880 --> 00:06:06,160 text prompts so you could see what the prompt was and what the 127 00:06:06,160 --> 00:06:09,000 reels have generated. They still look AI, but they 128 00:06:09,000 --> 00:06:11,920 look pretty impressive. Many of these are just as good 129 00:06:11,920 --> 00:06:15,120 as I'd see in ChatGPT, and Amazon is offering these at much 130 00:06:15,120 --> 00:06:18,240 lower or more competitive prices than existing competition to 131 00:06:18,240 --> 00:06:20,640 power all this generative AI. Amazon isn't working by 132 00:06:20,640 --> 00:06:22,200 themselves. They continue to deepen their 133 00:06:22,200 --> 00:06:25,520 relationship with Anthropic, which has so far been considered 134 00:06:25,520 --> 00:06:29,320 2nd place to Chachi BT, but over time, at least right now, it 135 00:06:29,320 --> 00:06:32,160 seems like many of the capabilities from Anthropic are 136 00:06:32,160 --> 00:06:34,920 a very close second place. The relationship is pretty 137 00:06:34,920 --> 00:06:37,040 simple. Amazon is giving them credits to 138 00:06:37,040 --> 00:06:39,960 run all of their processing on Amazon's web services. 139 00:06:39,960 --> 00:06:42,960 Amazon in return gets equity and influence over Anthropic. 140 00:06:43,000 --> 00:06:45,440 So they're they're both swapping different services they use. 141 00:06:45,440 --> 00:06:48,480 Now all of this is really cool. Amazon's going to gain a lot of 142 00:06:48,480 --> 00:06:50,840 additional customers with Amazon Nova. 143 00:06:51,160 --> 00:06:53,680 There's already so many companies using Amazon Web 144 00:06:53,680 --> 00:06:55,840 Services. So leveraging that already 145 00:06:55,840 --> 00:06:59,440 existing platform and then pushing Amazon Nova seems like a 146 00:06:59,440 --> 00:07:01,440 no brainer. But one part of this that I 147 00:07:01,440 --> 00:07:05,040 think needs to get highlighted again is that inside of Amazon, 148 00:07:05,120 --> 00:07:08,520 they're already using 1000 Gen. AI applications. 149 00:07:08,960 --> 00:07:12,200 They're already in motion. And this is the AI story that's 150 00:07:12,200 --> 00:07:15,560 being left out. Amazon so far by investors has 151 00:07:15,560 --> 00:07:19,560 been considered a second tier, third tier AI company, one that 152 00:07:19,560 --> 00:07:22,480 doesn't really get included in the conversations of companies 153 00:07:22,720 --> 00:07:25,360 that will benefit dramatically from artificial intelligence. 154 00:07:25,400 --> 00:07:27,800 This narrative is changing and it will continue to change 155 00:07:27,800 --> 00:07:31,040 rapidly in 2025. How I know this narrative is 156 00:07:31,040 --> 00:07:33,720 going to change is because if you've been following Amazon 157 00:07:33,720 --> 00:07:37,480 very closely, you'll know that they've been migrating AI into 158 00:07:37,480 --> 00:07:39,560 every single aspect of their company. 159 00:07:39,920 --> 00:07:44,000 Every part of it, from the movie business, Amazon Prime, to their 160 00:07:44,000 --> 00:07:47,160 retail business, to their servers to their advertisement. 161 00:07:47,480 --> 00:07:51,480 It is transforming Amazon, and this isn't some vague promise of 162 00:07:51,480 --> 00:07:53,560 the future without any specifics. 163 00:07:53,720 --> 00:07:57,120 Amazon gives detailed specifics of how this is transforming the 164 00:07:57,120 --> 00:07:59,480 company, starting with customer service. 165 00:07:59,760 --> 00:08:02,200 So take customer service. We have a retail business with a 166 00:08:02,200 --> 00:08:05,560 few 100 million customers they occasionally need to contact. 167 00:08:05,560 --> 00:08:09,080 Customer service, he mentions we have a retail business with a 168 00:08:09,080 --> 00:08:12,480 few 100 million customers. He throws out a few 100 million 169 00:08:12,480 --> 00:08:15,000 customers. Like it's just a side note, but 170 00:08:15,000 --> 00:08:18,120 just take that into account. Amazon's retail business has a 171 00:08:18,200 --> 00:08:22,880 few 100 million customers. They own 40% of online shopping. 172 00:08:22,960 --> 00:08:26,320 So anything that moves a needle with customer service at all, 173 00:08:26,760 --> 00:08:29,280 increasing customer service response time, customer 174 00:08:29,280 --> 00:08:32,760 satisfaction, lowering cost of customer service, that is 175 00:08:32,760 --> 00:08:35,840 massive for Amazon's business. Occasionally need to contact 176 00:08:35,840 --> 00:08:38,480 customer service. The vast majority of them prefer 177 00:08:38,480 --> 00:08:41,039 to do it in a self-service way so they can do it quickly and 178 00:08:41,039 --> 00:08:44,440 take care of it themselves. And we had built a chatbot many 179 00:08:44,440 --> 00:08:47,920 years ago and it of course used machine learning, but it had 180 00:08:47,920 --> 00:08:51,120 static decision trees and the customers had to enter a lot of 181 00:08:51,120 --> 00:08:55,200 words before you got answers. So a couple years ago we rebuilt 182 00:08:55,200 --> 00:09:00,200 this using generative AI and so now it's much easier for 183 00:09:00,200 --> 00:09:02,640 customers. So imagine I ordered an item a 184 00:09:02,640 --> 00:09:05,080 couple days ago. I get on the chat bot, the new 185 00:09:05,080 --> 00:09:09,520 chat bot, we know who you are, that you ordered a couple days 186 00:09:09,520 --> 00:09:12,480 ago, what you ordered, where you live. 187 00:09:12,720 --> 00:09:15,680 And we can predict in this model if you're contacting us just a 188 00:09:15,680 --> 00:09:18,840 couple days later that you might be contacting us about a return. 189 00:09:19,240 --> 00:09:22,320 And so when you start to tell us that, we quickly can tell you 190 00:09:22,640 --> 00:09:25,040 where the nearest physical location at a Whole Foods or 191 00:09:25,040 --> 00:09:26,960 somewhere else is that you can return that item. 192 00:09:27,520 --> 00:09:30,840 And then the model is also smart enough to be able to predict 193 00:09:31,080 --> 00:09:33,880 when you're getting frustrated with it and you might need to be 194 00:09:33,880 --> 00:09:35,720 connected to a human for resolution. 195 00:09:36,200 --> 00:09:39,920 Now this chatbot before we re engineered it already had very 196 00:09:39,920 --> 00:09:42,440 high customer satisfaction. But since we added the 197 00:09:42,440 --> 00:09:47,400 generative AI brain to it, 500 basis points better customer 198 00:09:47,400 --> 00:09:49,840 satisfaction, that's practical AI. 199 00:09:50,280 --> 00:09:54,040 Increasing customer satisfaction by 500 basis points. 200 00:09:54,160 --> 00:09:57,880 When you scale that out on Amazon's full retail business, 201 00:09:57,880 --> 00:10:01,040 that is incredible. The amount of savings and time 202 00:10:01,040 --> 00:10:04,040 and hassle for Amazon, the amount of money saved scales 203 00:10:04,040 --> 00:10:06,120 across hundreds of millions of customers. 204 00:10:06,720 --> 00:10:09,600 So even these little incremental changes from Gen. 205 00:10:09,680 --> 00:10:13,080 AI have massive impacts on the margin of Amazon. 206 00:10:13,120 --> 00:10:14,920 And of course, this doesn't just end with customer. 207 00:10:14,920 --> 00:10:19,640 Service or take sellers, we have about 2 million sellers who sell 208 00:10:19,640 --> 00:10:24,040 in our retail store worldwide. It's over 60% of the units that 209 00:10:24,040 --> 00:10:27,560 we now sell. And the way they get a product 210 00:10:27,560 --> 00:10:30,480 onto the website is they have to fill out this very long form. 211 00:10:30,840 --> 00:10:33,240 And the reason there are so many fields is we're trying to make 212 00:10:33,240 --> 00:10:36,080 it easy for our customers to navigate and understand what the 213 00:10:36,080 --> 00:10:38,320 products are. But it's a lot of work for 214 00:10:38,320 --> 00:10:41,160 sellers. And so we rebuilt a tool. 215 00:10:41,160 --> 00:10:44,440 We basically built a brand new tool using generative AI such 216 00:10:44,440 --> 00:10:48,760 that now sellers only have to enter a few words or they can 217 00:10:48,760 --> 00:10:51,600 take a picture, or they can point to a URL. 218 00:10:52,000 --> 00:10:55,760 And then the the tool fills in a lot of those attributes. 219 00:10:55,760 --> 00:11:00,400 It's much much easier for sellers and we have over 500,000 220 00:11:00,400 --> 00:11:02,400 sellers now using our generative AI tools. 221 00:11:02,400 --> 00:11:04,680 So we have AI integration with the customers. 222 00:11:04,680 --> 00:11:08,200 We also have AI integration with the sellers, making their job 223 00:11:08,200 --> 00:11:11,320 easier, making it so that more people want to sell on Amazon. 224 00:11:11,480 --> 00:11:15,320 This continues on with practical example after practical example. 225 00:11:15,320 --> 00:11:17,720 The amount of scale that Amazon has whenever they make one of 226 00:11:17,720 --> 00:11:20,080 these iterative changes is massive. 227 00:11:20,160 --> 00:11:22,280 And of course this also has different applications like 228 00:11:22,280 --> 00:11:24,520 their warehousing. Look at inventory management. 229 00:11:24,880 --> 00:11:27,560 So think about the scale of problem we have to solve. 230 00:11:27,560 --> 00:11:31,480 In our retail business, we have over 1000 different buildings or 231 00:11:31,480 --> 00:11:35,800 nodes as we call them. And everything we do is 232 00:11:35,800 --> 00:11:40,120 optimized to get the right product in a fulfilment centre 233 00:11:40,120 --> 00:11:42,560 or building close to the end customer to save on 234 00:11:42,560 --> 00:11:45,960 transportation time, which means we get it to you faster and we 235 00:11:45,960 --> 00:11:49,600 do it for lower cost. And so that means anyone point 236 00:11:50,080 --> 00:11:55,120 we have to understand what's in that fulfilment centre, what's 237 00:11:55,120 --> 00:11:58,160 the inventory levels of each item, which items are being 238 00:11:58,160 --> 00:12:02,080 ordered and at what rate do we have more capacity in that 239 00:12:02,080 --> 00:12:04,560 fulfillment center? Do we need to move inventory 240 00:12:04,560 --> 00:12:07,440 around to other fulfillment centers to balance the network? 241 00:12:08,120 --> 00:12:12,440 And so we've used transformer models to solve these problems 242 00:12:12,440 --> 00:12:16,360 and make forecasts. And already our long term demand 243 00:12:16,360 --> 00:12:20,240 forecasting transformer models has improved that accuracy by 244 00:12:20,240 --> 00:12:23,080 10%. And then we've also improved the 245 00:12:23,080 --> 00:12:25,720 regional prediction accuracy by over 20%. 246 00:12:25,960 --> 00:12:27,880 Those are big gains at our scale. 247 00:12:28,120 --> 00:12:31,720 He mentions that these are big gains at their scale, and again, 248 00:12:31,720 --> 00:12:35,600 scale is a meaningful word here. If Amazon was a very small shop, 249 00:12:35,600 --> 00:12:39,080 then 10 or 20% gains doesn't really hit margins quite as much 250 00:12:39,280 --> 00:12:42,320 as a company at their scale. But they're already using Gen. 251 00:12:42,400 --> 00:12:45,400 AI and the logic behind it to help solve these tough questions 252 00:12:45,600 --> 00:12:48,400 behind logistics. Any efficiency in logistics 253 00:12:48,400 --> 00:12:50,200 means cheaper prices for customers. 254 00:12:50,200 --> 00:12:52,200 They're more competitive than the competition. 255 00:12:52,200 --> 00:12:55,400 It means a bigger market share for Amazon, better returns for 256 00:12:55,400 --> 00:12:57,560 the investor. Or think about robotics. 257 00:12:58,920 --> 00:13:02,920 We have over 750,000 robots roaming our various fulfilment 258 00:13:02,920 --> 00:13:05,280 centres and they have all sorts of AI in them. 259 00:13:05,280 --> 00:13:08,080 But I'll, I'll give you the example of Sparrow, which is a 260 00:13:08,080 --> 00:13:12,640 robotic arm that does resorting. And so if you got a chance to 261 00:13:12,640 --> 00:13:16,240 zoom out of our fulfilment centres, it's really an 262 00:13:16,240 --> 00:13:19,480 operation that's constantly taking items from lots of 263 00:13:19,480 --> 00:13:22,520 different disparate parts and aggregating them into 264 00:13:22,560 --> 00:13:25,040 containers. So we optimize the capacity we 265 00:13:25,040 --> 00:13:26,440 have and the conveyance that we have. 266 00:13:27,120 --> 00:13:32,080 And So what Sparrow is doing is it's taking items from one bin 267 00:13:32,080 --> 00:13:34,000 and it's aggregating them into another bin. 268 00:13:34,400 --> 00:13:38,400 And So what the generator of AI needs to do in Sparrow is it has 269 00:13:38,400 --> 00:13:42,360 to tell them what's in the first bin, what item do we want them 270 00:13:42,360 --> 00:13:45,080 to go pick up? It has to discern which item is 271 00:13:45,080 --> 00:13:47,720 which. It has to know how to grasp that 272 00:13:47,720 --> 00:13:50,760 item given the size of it and the materials and the 273 00:13:50,760 --> 00:13:54,320 flexibility of that material. And it has to know where in the 274 00:13:54,320 --> 00:13:59,560 receiving bin it can put it. And so these are, these are all 275 00:13:59,560 --> 00:14:04,320 inventions that are critical to us changing the processing time 276 00:14:04,800 --> 00:14:06,920 and the cost to serve our customers. 277 00:14:06,920 --> 00:14:11,200 So we have about 5 of these brand new robotics inventions 278 00:14:11,200 --> 00:14:14,560 that we've pulled together in our Shreveport, LA fulfilment 279 00:14:14,560 --> 00:14:16,520 center just a couple months ago that we launched. 280 00:14:16,840 --> 00:14:20,760 And already we're seeing 25% faster processing times. 281 00:14:21,080 --> 00:14:24,320 And we believe we're going to have 25% lower cost to serve 282 00:14:24,320 --> 00:14:27,480 during the holidays because of these AI inventions and our 283 00:14:27,480 --> 00:14:29,600 robotics. Amazon already has the best 284 00:14:29,600 --> 00:14:32,200 online retail customer experience by far of any 285 00:14:32,200 --> 00:14:34,040 company. They are the easiest to work 286 00:14:34,040 --> 00:14:36,800 with, the fastest to ship, the best return policy. 287 00:14:37,040 --> 00:14:39,600 Overall, they treat their members the very best out of any 288 00:14:39,600 --> 00:14:42,800 company online, and this is only going to get better. 289 00:14:42,960 --> 00:14:45,760 The distance between Amazon and their secondary competitors is 290 00:14:45,760 --> 00:14:48,840 going to widen. But we're also seeing altogether 291 00:14:48,840 --> 00:14:52,320 brand new shopping experiences that we're able to generate and 292 00:14:52,320 --> 00:14:54,840 invent with generative AI. So a few examples. 293 00:14:54,840 --> 00:14:57,920 I'll start with some agents. Let's start with Rufus, which is 294 00:14:57,920 --> 00:15:01,280 our shopping agent. So if you're going to buy an 295 00:15:01,280 --> 00:15:04,960 item and you know what you want, I would argue that there isn't a 296 00:15:04,960 --> 00:15:09,560 better experience than ordering it on Amazon, having it shipped 297 00:15:09,560 --> 00:15:13,440 very quickly to your home. However, if you don't know what 298 00:15:13,440 --> 00:15:17,240 you want and you're trying to decide, you can obviously do it 299 00:15:17,240 --> 00:15:19,000 at Amazon. Many of you do bless you and 300 00:15:19,000 --> 00:15:22,240 thank you for doing that. However, you can know it's, you 301 00:15:22,240 --> 00:15:24,960 know, and you do it through browse nodes and you do it 302 00:15:24,960 --> 00:15:27,360 through recommendations that we make and you do it through 303 00:15:27,360 --> 00:15:30,160 customer reviews. But there's something nice when 304 00:15:30,160 --> 00:15:32,640 you don't know what you want about going into a physical 305 00:15:32,640 --> 00:15:35,600 store and asking the salesperson, you know, telling 306 00:15:35,600 --> 00:15:38,240 them what you're thinking about and having them ask narrowing 307 00:15:38,240 --> 00:15:42,000 questions and then pointing you to maybe the couple items that 308 00:15:42,400 --> 00:15:45,160 you want to consider. And then you look at those items 309 00:15:45,160 --> 00:15:47,440 and you don't have all the data in front of you and you ask that 310 00:15:47,440 --> 00:15:49,200 salesperson, well, what about this? 311 00:15:49,200 --> 00:15:51,320 What about that? They could answer it quickly if 312 00:15:51,320 --> 00:15:56,040 they don't walk away. And then you know you have the 313 00:15:56,040 --> 00:15:59,640 ability to make those decisions and what you want quickly. 314 00:16:00,040 --> 00:16:04,440 And what we're trying to do with Rufus is we're trying to make 315 00:16:04,440 --> 00:16:07,800 that experience even better. So with Rufus, you can go to any 316 00:16:07,800 --> 00:16:11,000 product detail page and instead of going through the plethora of 317 00:16:11,000 --> 00:16:15,280 information on that detail page, you can ask any question and 318 00:16:15,280 --> 00:16:16,760 Rufus will answer it really quickly. 319 00:16:17,200 --> 00:16:20,080 Rufus will make comparisons for you across products and 320 00:16:20,080 --> 00:16:22,960 categories. It'll make recommendations you 321 00:16:22,960 --> 00:16:26,080 can make really, you can ask really broad questions for 322 00:16:26,080 --> 00:16:28,280 recommendations and it'll ask narrowing questions. 323 00:16:28,280 --> 00:16:30,440 So it gets it really what your intent is. 324 00:16:31,120 --> 00:16:34,680 If you say to Rufus, hey, I want that same golf glove that I 325 00:16:34,680 --> 00:16:36,120 always get, can you find that for me? 326 00:16:36,120 --> 00:16:38,520 The one I've ordered, Rufus will find it for you. 327 00:16:38,880 --> 00:16:41,840 You can say to Rufus, give me the order status of, of the 328 00:16:41,840 --> 00:16:43,760 items that haven't been delivered yet and you'll get 329 00:16:43,760 --> 00:16:46,440 that too. And one of the nice things about 330 00:16:46,440 --> 00:16:51,560 Rufus relative to to physical salesperson is that Rufus is not 331 00:16:51,560 --> 00:16:54,200 going to take another job at another retailer or it's not 332 00:16:54,200 --> 00:16:56,360 going to start working in another profession. 333 00:16:56,760 --> 00:17:00,080 Rufus is going to be there with you all this time, getting to 334 00:17:00,080 --> 00:17:03,240 know your intent and your interests and what you want 335 00:17:03,240 --> 00:17:05,280 better and better. Rufus was something that was 336 00:17:05,280 --> 00:17:07,640 launched that at first seemed unclear. 337 00:17:07,640 --> 00:17:10,680 It seemed like a little helpful tool, but I at first did not 338 00:17:10,680 --> 00:17:13,079 really understand the intent of this product. 339 00:17:13,240 --> 00:17:15,760 But it's clear that Amazon is designing Rufus to be an AI 340 00:17:15,760 --> 00:17:19,119 agent, to simply be a place where you can have a mini 341 00:17:19,119 --> 00:17:23,440 version of ChatGPT or Amazon Nova within the Amazon app. 342 00:17:23,520 --> 00:17:27,160 And again, these examples with Amazon implementing AI are 343 00:17:27,160 --> 00:17:29,400 literally nonstop. They just go on. 344 00:17:29,600 --> 00:17:31,800 And we're just talking about the retail business here. 345 00:17:31,840 --> 00:17:34,960 Jassy continues on giving examples of how generative AI is 346 00:17:34,960 --> 00:17:37,360 being used in Alexa. Of course, this is going to be 347 00:17:37,360 --> 00:17:40,120 dramatically improved. They're intertwining all their 348 00:17:40,120 --> 00:17:41,840 aspects of their business with Alexa. 349 00:17:42,280 --> 00:17:44,800 But it doesn't end here. He gives more and more practical 350 00:17:44,800 --> 00:17:47,240 applications. He moves on from discussing the 351 00:17:47,240 --> 00:17:50,720 impact of agents to now discussing different features on 352 00:17:50,720 --> 00:17:51,840 Amazon's app. There's a. 353 00:17:52,240 --> 00:17:54,280 A feature we have called Amazon Lens. 354 00:17:54,600 --> 00:17:58,920 So let's say that you're at a friend's house and you see a 355 00:17:58,920 --> 00:18:01,280 planter they have that you admire because this happens to 356 00:18:01,280 --> 00:18:05,560 be very often and you want to know where that planter is from. 357 00:18:05,560 --> 00:18:07,600 And you ask your friend. My friend doesn't know. 358 00:18:08,080 --> 00:18:12,440 What you can do today is you can plug into a search engine like 359 00:18:12,440 --> 00:18:18,840 Amazon or somewhere else, planter, hanging macrame, maybe 360 00:18:18,840 --> 00:18:20,840 you'll get a decent answer. Probably not. 361 00:18:21,120 --> 00:18:24,920 Instead, you can use Amazon Lens and you can take a picture of 362 00:18:24,920 --> 00:18:27,880 that item. And what Amazon Lens is doing is 363 00:18:27,880 --> 00:18:32,160 using computer vision and then a multimodal model underneath it 364 00:18:32,160 --> 00:18:34,960 to do a search query that leads you right to the right search 365 00:18:34,960 --> 00:18:36,840 result on Amazon where you can buy it easily. 366 00:18:36,920 --> 00:18:39,360 It's really magical and cool. These are features that are 367 00:18:39,360 --> 00:18:42,280 going to become standard practice in the future as 368 00:18:42,280 --> 00:18:45,320 customers learn about them and become, I'm convinced that they 369 00:18:45,320 --> 00:18:48,160 actually work well, they will adopt these new features. 370 00:18:48,360 --> 00:18:51,040 Shopping behaviors have changed over the past five years 371 00:18:51,040 --> 00:18:52,960 dramatically. They're going to continue to 372 00:18:52,960 --> 00:18:56,840 change with the advent of AI, and Amazon is leading this 373 00:18:56,840 --> 00:18:59,640 online shopping AI integrated future. 374 00:18:59,720 --> 00:19:02,880 We've all had this experience where, you know we're buying a 375 00:19:02,880 --> 00:19:06,160 shirt and you don't really know if that brand runs large or 376 00:19:06,160 --> 00:19:08,600 small, whether you'd be a medium or large in that shirt. 377 00:19:09,080 --> 00:19:12,400 What we've done is we've built a large language model that takes 378 00:19:12,560 --> 00:19:15,600 all the sizing relationships between in the many, many brands 379 00:19:15,600 --> 00:19:19,160 that we have and compare which ones run like each other, which 380 00:19:19,520 --> 00:19:22,520 ones run larger or smaller. And then we're able to look for 381 00:19:22,520 --> 00:19:25,360 customers, what you've bought before and when you're at a new 382 00:19:25,360 --> 00:19:28,320 brand, we can make the right recommendation for what size you 383 00:19:28,320 --> 00:19:30,800 really should order. Very handy, very practical. 384 00:19:31,040 --> 00:19:33,440 This is something that I've already used on Amazon and it 385 00:19:33,440 --> 00:19:35,800 works perfectly. They already know based off of 386 00:19:35,800 --> 00:19:39,000 your order history what size you are, and then they know exactly 387 00:19:39,000 --> 00:19:41,360 what size of this product is going to fit you based on 388 00:19:41,360 --> 00:19:43,960 different reviews. So this is another pain point 389 00:19:43,960 --> 00:19:47,160 from ordering clothes online is getting the sizing right. 390 00:19:47,280 --> 00:19:50,440 So if Amazon can cut down on the massive cost associated with 391 00:19:50,440 --> 00:19:53,760 large amounts of returns by getting the size correct, that's 392 00:19:53,760 --> 00:19:56,440 another thing that can improve the margins of the retail 393 00:19:56,440 --> 00:19:58,280 business. And then if you look at what 394 00:19:58,280 --> 00:20:03,320 we're doing in Prime Video, we have a very deep partnership 395 00:20:03,320 --> 00:20:05,680 with the NFL. We've built something together 396 00:20:05,680 --> 00:20:10,080 over the years called NextGen Stats and we collect 500 million 397 00:20:10,080 --> 00:20:12,440 data points every season and then. 398 00:20:12,560 --> 00:20:16,080 We've built AI models on top of that and so you can see some of 399 00:20:16,080 --> 00:20:18,080 the features we've built. We've built something called 400 00:20:18,560 --> 00:20:23,840 defensive alerts, which shows you which defensive player might 401 00:20:23,840 --> 00:20:26,800 blitz the quarterback, puts a circle around it, changes the 402 00:20:26,800 --> 00:20:29,480 viewing experience. Or we can look at different 403 00:20:29,480 --> 00:20:33,240 formations and sets and detect where the defence may be 404 00:20:33,240 --> 00:20:34,840 vulnerable. And so we have a defensive 405 00:20:34,840 --> 00:20:38,840 vulnerability feature where we can highlight for viewers where 406 00:20:38,840 --> 00:20:43,280 the offence should attack. These change the experience for 407 00:20:43,280 --> 00:20:46,440 fans. And so these are by the way. 408 00:20:46,800 --> 00:20:49,000 This is actually astonishing, what they're doing. 409 00:20:49,000 --> 00:20:51,600 They look at the layout and formation of the football 410 00:20:51,600 --> 00:20:54,880 players, and then they kind of identify what the most likely 411 00:20:54,880 --> 00:20:57,720 play is to occur. They're predicting the future 412 00:20:57,720 --> 00:21:00,640 with generative AI and then sharing that with the viewer so 413 00:21:00,640 --> 00:21:02,520 they know where to better pay attention to. 414 00:21:02,560 --> 00:21:05,640 If Amazon gets this right, this is another thing that separates 415 00:21:05,640 --> 00:21:08,520 them from the competition. This will be difficult for other 416 00:21:08,520 --> 00:21:11,600 companies to implement at the scale that Amazon's already at. 417 00:21:11,640 --> 00:21:14,840 Amazon Prime Video is already a world class asset. 418 00:21:14,880 --> 00:21:18,240 In fact, now there's reports that 70% of the people that sign 419 00:21:18,240 --> 00:21:22,000 up for Amazon Prime sign up because of Amazon Prime Video. 420 00:21:22,320 --> 00:21:24,560 So as they're adding more value to this, they'll get more 421 00:21:24,560 --> 00:21:28,400 customer acquisitions as well. The applications that Amazon has 422 00:21:28,400 --> 00:21:31,200 to implement AI into their business are enormous. 423 00:21:31,200 --> 00:21:34,400 The opportunity here is larger than most other companies. 424 00:21:34,720 --> 00:21:38,040 So when investors disregard Amazon or don't consider it an 425 00:21:38,120 --> 00:21:41,600 AI company, I leave confused. I don't understand that thought 426 00:21:41,600 --> 00:21:43,840 process. Amazon has every reason to be 427 00:21:43,840 --> 00:21:46,280 considered one of the top AI companies in the world. 428 00:21:46,320 --> 00:21:49,240 Not only are they intermingling AI into all of their business, 429 00:21:49,360 --> 00:21:52,080 they're competing with Chachi, PT with Amazon Nova. 430 00:21:52,280 --> 00:21:56,000 They have massive cloud scale with Amazon AWS, but they're 431 00:21:56,000 --> 00:21:57,800 also competing on the hardware front. 432 00:21:57,880 --> 00:22:00,920 Amazon announced that they're building a supercomputer, a new 433 00:22:00,920 --> 00:22:03,360 server powered by homegrown AI chips. 434 00:22:03,360 --> 00:22:06,040 This is under Amazon Web Services and it's going to be 435 00:22:06,040 --> 00:22:10,080 called the Ultra Cluster, a massive AI supercomputer made-up 436 00:22:10,120 --> 00:22:13,320 of hundreds of thousands of homegrown Tranium chips. 437 00:22:13,440 --> 00:22:16,080 This means that Amazon is the one designing these chips. 438 00:22:16,200 --> 00:22:18,800 This means that Amazon is the one designing these chips. 439 00:22:18,880 --> 00:22:21,960 The chip cluster will be used by the AI startup Anthropic. 440 00:22:22,040 --> 00:22:25,440 It'll be located in the US. It'll be ready in 2025. 441 00:22:25,480 --> 00:22:28,000 It'll be one of the largest in the world for training AI 442 00:22:28,000 --> 00:22:30,520 models. When the AWS CEO was talking 443 00:22:30,520 --> 00:22:33,880 about this new chip cluster, he also mentioned that Apple is 444 00:22:33,880 --> 00:22:35,480 using it. Apple's one of the biggest 445 00:22:35,480 --> 00:22:37,880 customers of Amazon's custom AI chips. 446 00:22:37,960 --> 00:22:40,320 Apple said that we have a strong relationship and the 447 00:22:40,320 --> 00:22:43,320 infrastructure is both reliable and able to serve our customers 448 00:22:43,320 --> 00:22:45,080 worldwide. Now, of course, when Amazon 449 00:22:45,080 --> 00:22:47,400 announces that they're creating their own chips and customers 450 00:22:47,400 --> 00:22:50,480 like Ale are choosing them, this pits them a little bit against 451 00:22:50,480 --> 00:22:54,120 NVIDIA, a company that is a current partner of Amazons. 452 00:22:54,560 --> 00:22:56,560 Quote Today. There's really only one choice 453 00:22:56,560 --> 00:22:59,080 for the GPU side, it's just NVIDIA. 454 00:22:59,440 --> 00:23:02,040 That's what Matt Gurman, the chief executive of Amazon Web 455 00:23:02,040 --> 00:23:04,000 Services, said. We think that customers would 456 00:23:04,000 --> 00:23:06,120 appreciate having multiple choices. 457 00:23:06,200 --> 00:23:09,320 Amazon's trying to break Nvidia's monopoly to become a 458 00:23:09,320 --> 00:23:11,360 duopoly. A key part of Amazon's AI 459 00:23:11,360 --> 00:23:15,000 strategy is to update its custom silicon so that it can not only 460 00:23:15,000 --> 00:23:18,240 bring down the cost of AI for its business customers, but also 461 00:23:18,240 --> 00:23:20,840 give the company more control over its supply chain. 462 00:23:21,200 --> 00:23:24,320 That could also make AWS less reliant on NVIDIA. 463 00:23:24,560 --> 00:23:27,360 NVIDIA, of course, has made a fortune by selling their GPU's 464 00:23:27,360 --> 00:23:30,840 to mostly big tech customers, which make up over 50% of their 465 00:23:30,840 --> 00:23:33,320 current revenue. And Amazon's motto of your 466 00:23:33,320 --> 00:23:35,920 margin is my opportunity seems to be at play here. 467 00:23:36,120 --> 00:23:38,320 They want to drive down the margins of NVIDIA. 468 00:23:38,440 --> 00:23:41,840 It's going to take time for Amazon to make a lot of gains in 469 00:23:41,840 --> 00:23:44,400 the GPU market. Even though the use of Amazon's 470 00:23:44,400 --> 00:23:47,400 GPU's is much more narrow, the scope is smaller and the 471 00:23:47,400 --> 00:23:50,800 application is more specific. This will drive down costs for 472 00:23:50,800 --> 00:23:53,200 GPU's overall. When Amazon starts creating 473 00:23:53,200 --> 00:23:56,720 alternatives in any capacity to NVIDIA, that's going to drive 474 00:23:56,720 --> 00:23:58,480 down costs. So whether we're looking at the 475 00:23:58,480 --> 00:24:01,400 hardware aspect or the software aspect, something that I've been 476 00:24:01,400 --> 00:24:04,680 saying for over a year now with Amazon is that it is one of the 477 00:24:04,680 --> 00:24:07,400 leading AI companies. It is going to be a force to be 478 00:24:07,400 --> 00:24:10,160 reckoned with and investors are starting to recognize this. 479 00:24:10,360 --> 00:24:12,280 This is a narrative that has been changed. 480 00:24:12,480 --> 00:24:16,000 Amazon has been branded as one that's behind an AI or not 481 00:24:16,000 --> 00:24:18,600 really part of the conversation. And over the next year, we're 482 00:24:18,600 --> 00:24:20,240 going to continue to see that change. 483 00:24:20,480 --> 00:24:23,000 And as that happens, I'm going to be remaining fully invested 484 00:24:23,000 --> 00:24:25,160 in this company. Now moving on, another stock in 485 00:24:25,160 --> 00:24:27,160 My Portfolio that's going through a bit of a 486 00:24:27,160 --> 00:24:31,440 transformation is Salesforce. I purchased into Salesforce over 487 00:24:31,440 --> 00:24:34,160 the past six months and this one continues to rise. 488 00:24:34,200 --> 00:24:36,160 Today being the best day since I bought it. 489 00:24:36,240 --> 00:24:39,600 Salesforce stock is currently up 9% on the day after reporting 490 00:24:39,600 --> 00:24:42,680 earnings yesterday after close. There's banks raising their 491 00:24:42,680 --> 00:24:45,320 price targets. Analysts across the board are 492 00:24:45,320 --> 00:24:47,440 all becoming more excited about this company. 493 00:24:47,520 --> 00:24:50,520 Salesforce basically did exactly what they said they were going 494 00:24:50,520 --> 00:24:52,600 to do. If we look at their EPS and 495 00:24:52,600 --> 00:24:56,160 revenue estimates, they actually missed narrowly on the EPS. 496 00:24:56,440 --> 00:24:58,520 If we look at the estimate compared to dismissed, they 497 00:24:58,520 --> 00:25:01,040 missed by two or three pennies. And if we look at the revenue, 498 00:25:01,040 --> 00:25:02,880 they did be on their revenue estimates. 499 00:25:02,880 --> 00:25:05,120 So they reported 1/4. That was in line with 500 00:25:05,120 --> 00:25:07,320 expectation. We can turn to the guidance, but 501 00:25:07,320 --> 00:25:10,600 the guidance was also roughly in line with expectations. 502 00:25:10,600 --> 00:25:14,400 So why is Salesforce up 9%? The majority of this gain is 503 00:25:14,400 --> 00:25:18,080 because of a shift in narrative, A Salesforce similar to Amazon. 504 00:25:18,360 --> 00:25:21,800 Investors are getting more on board that this company is an AI 505 00:25:21,800 --> 00:25:23,760 company. Salesforce just released their 506 00:25:23,760 --> 00:25:27,000 new product called Agentforce, and within the first week of 507 00:25:27,000 --> 00:25:30,960 business, Salesforce's CEO said that they had over 200 customers 508 00:25:30,960 --> 00:25:33,960 sign up for the service. Agent forces Salesforce's new 509 00:25:33,960 --> 00:25:38,320 branding for their AI agents. These are better than chat bots. 510 00:25:38,320 --> 00:25:41,120 They're chat bots that can do very complex tasks. 511 00:25:41,360 --> 00:25:44,200 In fact, they can kind of do entire human tasks. 512 00:25:44,600 --> 00:25:47,760 The big selling point for Agentforce is that it will lower 513 00:25:47,760 --> 00:25:51,720 the intensity of work required for their company's customers. 514 00:25:51,800 --> 00:25:54,360 They're promising remarkable efficiency gains through 515 00:25:54,360 --> 00:25:56,440 Agentforce. We've gone through a phase of 516 00:25:56,440 --> 00:25:59,000 generative AI and we've seen what that has done, the 517 00:25:59,000 --> 00:26:01,800 efficiencies it's gained. But agents are this entire new 518 00:26:01,800 --> 00:26:04,320 concept. It's like the next level of AI. 519 00:26:04,520 --> 00:26:07,960 And in an interview today, Sam Altman described what he thinks 520 00:26:07,960 --> 00:26:11,360 is going to be the biggest change in 2025, the next 521 00:26:11,360 --> 00:26:14,520 iterative development that will change people's perspectives to 522 00:26:14,600 --> 00:26:16,160 what AI is capable of. Sure. 523 00:26:16,160 --> 00:26:20,840 I expect that in 2025 we will have systems that people look 524 00:26:20,840 --> 00:26:23,720 at, even people who are skeptical of current progress 525 00:26:24,080 --> 00:26:27,560 and say, wow, that I did not expect that. 526 00:26:27,560 --> 00:26:29,960 That does change what? Like what? 527 00:26:33,000 --> 00:26:35,360 Agents are the thing everyone is talking about, I think for good 528 00:26:35,360 --> 00:26:37,080 reason. You know, this idea that you can 529 00:26:37,080 --> 00:26:40,880 give an AI system a pretty complicated task, like a kind of 530 00:26:40,880 --> 00:26:44,280 task you'd give to a very smart human that takes a while to go 531 00:26:44,280 --> 00:26:47,160 off and do and use a bunch of tools and create something of 532 00:26:47,160 --> 00:26:49,560 value. That's the kind of thing I'd 533 00:26:49,560 --> 00:26:52,160 expect next year. And that's like a huge deal. 534 00:26:52,360 --> 00:26:54,160 We talk about that like, oh, you know, this thing is going to 535 00:26:54,160 --> 00:26:57,440 happen, but that's like, if that works as well as we hope it 536 00:26:57,440 --> 00:26:59,640 does, that can that can really transform things. 537 00:26:59,640 --> 00:27:02,680 And Sam Altman is right. If this is correct, if there are 538 00:27:02,680 --> 00:27:05,720 agents that can accomplish these complex tasks, that's going to 539 00:27:05,720 --> 00:27:08,560 be incredibly transformative and it's going to be very valuable 540 00:27:08,560 --> 00:27:12,000 for the companies that have the distribution system to sell 541 00:27:12,000 --> 00:27:15,480 these agents to their customers. Well, what company has a massive 542 00:27:15,480 --> 00:27:19,760 amount of existing customers, maybe over 150,000 customers? 543 00:27:20,120 --> 00:27:22,920 Salesforce does. What company has customers data 544 00:27:22,920 --> 00:27:26,760 in almost every segment of their business, in sales, in service, 545 00:27:26,760 --> 00:27:29,280 in platform, in marketing and integration? 546 00:27:29,680 --> 00:27:32,520 Salesforce does. Salesforce as a company has more 547 00:27:32,520 --> 00:27:36,280 customers and more customer data than almost any company on 548 00:27:36,280 --> 00:27:39,280 earth. They are the CRM, the customer 549 00:27:39,280 --> 00:27:42,280 relationship management tool for hundreds of thousands of 550 00:27:42,280 --> 00:27:44,440 companies. They already have all of that 551 00:27:44,440 --> 00:27:45,840 data. They already have all of those 552 00:27:45,840 --> 00:27:48,440 connections. They have all the sales funnel 553 00:27:48,440 --> 00:27:51,120 existing. All they need to do is press a 554 00:27:51,120 --> 00:27:54,760 button and sell agents to them. That is exactly the vision that 555 00:27:54,760 --> 00:27:57,600 Marc Benioff sold in this latest earnings report. 556 00:27:58,000 --> 00:28:02,120 He gave practically a 20 minute dissertation on what agents are. 557 00:28:02,400 --> 00:28:06,080 If you want the most in depth, comprehensive view of agents, 558 00:28:06,400 --> 00:28:08,840 just go listen to the last Salesforce earnings call. 559 00:28:09,080 --> 00:28:12,880 He goes on and on for well over 20 minutes describing this 560 00:28:12,880 --> 00:28:15,800 transformative technology. His opening remarks, I actually 561 00:28:15,800 --> 00:28:18,480 believe may have set a record for the longest opening remarks 562 00:28:18,480 --> 00:28:21,640 of any CEO in an earnings call. Although out of everything that 563 00:28:21,640 --> 00:28:24,240 he said in the opening remarks, I believe the most illustrative 564 00:28:24,240 --> 00:28:27,720 example of agent force was given during the Q&A. 565 00:28:28,240 --> 00:28:30,200 He gave an example of healthcare. 566 00:28:30,200 --> 00:28:32,640 Let's go ahead and listen to the story that Marc Benioff shares 567 00:28:32,640 --> 00:28:34,600 about healthcare. I just got a call. 568 00:28:34,600 --> 00:28:38,280 From my hospital telling me that I'm coming in to get scheduled 569 00:28:38,280 --> 00:28:41,320 for another MRI and incredible service. 570 00:28:41,320 --> 00:28:44,920 You know, Ruben knows that this great relationship with UCSF and 571 00:28:44,920 --> 00:28:47,680 they're, they, they pay a lot of attention to me. 572 00:28:47,680 --> 00:28:49,120 I pay a lot of attention to them. 573 00:28:49,560 --> 00:28:53,240 Incredible organization. And they had a lot of questions 574 00:28:53,240 --> 00:28:58,000 for me about getting me ready for my next MRI and, and at the 575 00:28:58,000 --> 00:29:00,720 end of the call and this kind of preoperative care and there'll 576 00:29:00,720 --> 00:29:02,840 be another call post operative and so forth. 577 00:29:02,840 --> 00:29:06,440 And I, I was just saying to myself, wow, what did that cost 578 00:29:06,440 --> 00:29:08,480 them? And they just don't have enough 579 00:29:08,480 --> 00:29:10,720 people as it is. Their doctors are already burned 580 00:29:10,720 --> 00:29:14,200 out, their nurses are burnt out. I've talked to the folks all the 581 00:29:14,200 --> 00:29:16,040 time. You know, there's a lot of 582 00:29:16,040 --> 00:29:18,640 pajama time for all these doctors is what they call it, 583 00:29:18,640 --> 00:29:20,960 because they're all working late at night with their families 584 00:29:21,280 --> 00:29:23,240 trying to get through all of their messages. 585 00:29:23,720 --> 00:29:27,560 Everybody's maxed out at UCSFI. Mean it's an incredible org, but 586 00:29:27,560 --> 00:29:32,080 I, I, I can't believe, you know, the amount that they have to do 587 00:29:32,080 --> 00:29:34,200 with such a actually limited workforce. 588 00:29:34,760 --> 00:29:36,720 And then, you know, I got this call. 589 00:29:36,720 --> 00:29:40,000 I said, you know, they kind of know all of this already about 590 00:29:40,000 --> 00:29:42,480 me. They've got all of my data, they 591 00:29:42,480 --> 00:29:46,440 have all of my care, they have my family history, they've got 592 00:29:46,440 --> 00:29:49,880 all my scans. They can have an agent do this 593 00:29:49,880 --> 00:29:52,880 work. And that call probably cost them 594 00:29:52,880 --> 00:29:57,000 $100 and it didn't have to happen. 595 00:29:57,000 --> 00:30:00,040 And I think that we could have done the call for probably about 596 00:30:00,040 --> 00:30:03,880 $1.50. And I think that that is the 597 00:30:03,880 --> 00:30:06,400 message to our customers. That's the sales pitch. 598 00:30:06,440 --> 00:30:08,960 It's that simple. Right now, humans are doing 599 00:30:08,960 --> 00:30:12,760 things that cost $100 per call. When you factor in a human 600 00:30:12,760 --> 00:30:16,080 salary, they're scheduling their healthcare, everything you need 601 00:30:16,080 --> 00:30:20,040 to do to hire a person to return calls, to answer all these 602 00:30:20,040 --> 00:30:23,080 different inquiries. There's a lot of cost associated 603 00:30:23,080 --> 00:30:25,360 to it. Salesforce created a product, 604 00:30:25,360 --> 00:30:28,160 these agents that they can distribute with their massive 605 00:30:28,160 --> 00:30:31,320 customer base and the sales pitches, It's going to save them 606 00:30:31,400 --> 00:30:35,040 a lot of money, a lot of money on returning monotonous calls 607 00:30:35,040 --> 00:30:38,120 that agents can figure out. The technology is finally here 608 00:30:38,120 --> 00:30:40,040 to solve it. When you have a margin 609 00:30:40,040 --> 00:30:44,600 difference between costing $1.50 or $100, that is an easy pitch. 610 00:30:44,720 --> 00:30:47,440 The only thing that remains is for customers to be convinced 611 00:30:47,640 --> 00:30:50,400 that this product is worth it. That's why they're implementing 612 00:30:50,400 --> 00:30:54,760 things like calculate your ROI right on Salesforce's website. 613 00:30:55,200 --> 00:30:58,280 This is a tool they developed to help customers understand their 614 00:30:58,280 --> 00:31:00,680 potential savings. They can put in the number of 615 00:31:00,680 --> 00:31:03,880 customer service employees, the average annual salary of those 616 00:31:03,880 --> 00:31:05,680 employees. You can punch in the number of 617 00:31:05,680 --> 00:31:08,960 conversations those employees take on everyday, and then you 618 00:31:08,960 --> 00:31:11,520 have a dial here to calculate the percentage of support 619 00:31:11,520 --> 00:31:14,880 conversations shifted to Agentforce over three years. 620 00:31:15,360 --> 00:31:18,840 If you get to 50%, here's a chart of the savings. 621 00:31:18,960 --> 00:31:23,920 Year 1 you save $200,000. Year 2 you save $283,000. 622 00:31:24,200 --> 00:31:28,480 Year three you saved $405,000 with only a small team of 20 623 00:31:28,480 --> 00:31:31,520 customer service agents. After three years, you've saved 624 00:31:31,560 --> 00:31:35,120 $886,000. Salesforce is also changing the 625 00:31:35,120 --> 00:31:37,040 way the customers are paying for this product. 626 00:31:37,360 --> 00:31:41,280 Instead of paying a flat license or a fee or a per seat license, 627 00:31:41,600 --> 00:31:43,600 customers pay for it through consumption. 628 00:31:43,840 --> 00:31:46,880 The customer doesn't pay for Agentforce unless they're saving 629 00:31:46,880 --> 00:31:49,880 money with Agentforce, but this also is beneficial for 630 00:31:49,880 --> 00:31:52,520 Salesforce. Salesforce knows that AI is 631 00:31:52,520 --> 00:31:55,480 replacing many different seats or different employees at 632 00:31:55,480 --> 00:31:58,840 companies, so having a consumption based model protects 633 00:31:58,840 --> 00:32:02,600 them from that dynamic change. Every single customer in Cory 634 00:32:02,600 --> 00:32:06,280 they can complete, they can get a dollar or $1.50 and that's a 635 00:32:06,280 --> 00:32:09,000 savings for their customer and its money directly in 636 00:32:09,000 --> 00:32:11,880 Salesforce's pocket. All Salesforce has to do right 637 00:32:11,880 --> 00:32:14,840 now is push this product to their customers. 638 00:32:15,160 --> 00:32:16,920 They have to run the sales pipeline. 639 00:32:16,960 --> 00:32:20,160 That is why Salesforce is hiring 1000 people to sell this product 640 00:32:20,160 --> 00:32:22,560 to their customers. They want to get the word out. 641 00:32:22,560 --> 00:32:25,000 They want to let their customers know we have this new product 642 00:32:25,000 --> 00:32:26,760 that can stand to save you a lot of money. 643 00:32:26,760 --> 00:32:29,360 Marc Benioff also noted that within the first week of this 644 00:32:29,360 --> 00:32:31,640 product being launched, they already have 200 customers 645 00:32:31,640 --> 00:32:34,480 signed on and they have conversations with thousands 646 00:32:34,480 --> 00:32:35,880 more. So if you're wondering why 647 00:32:35,880 --> 00:32:39,000 Salesforce is bouncing up after earnings that seem to be in line 648 00:32:39,000 --> 00:32:42,160 with expectations, it's because investors again, are getting 649 00:32:42,160 --> 00:32:45,240 more on board that this company is not going to be destroyed by 650 00:32:45,240 --> 00:32:47,760 artificial intelligence. In fact, it's going to be one of 651 00:32:47,760 --> 00:32:50,960 the biggest beneficiaries. Unlike Amazon, the biggest 652 00:32:50,960 --> 00:32:54,000 benefit that Salesforce currently has is their already 653 00:32:54,000 --> 00:32:57,320 established distribution system. This knowledge and data that 654 00:32:57,320 --> 00:32:59,880 they have around hundreds of thousands of customers is 655 00:32:59,880 --> 00:33:02,480 incredibly useful for when they they want to sell a new product 656 00:33:02,640 --> 00:33:04,440 like Agentforce. Now moving on. 657 00:33:04,440 --> 00:33:06,080 I saw something that I thought was funny. 658 00:33:06,080 --> 00:33:09,800 In fact, I was sent this by a couple people, but it's JJ Watt, 659 00:33:09,880 --> 00:33:13,640 a former NFL player and he gives a huge endorsement for Texas 660 00:33:13,640 --> 00:33:16,640 Roadhouse on Twitter. He tweeted Texas Roadhouse 661 00:33:16,640 --> 00:33:19,920 simply does not miss ever with a picture of the meal that he 662 00:33:19,920 --> 00:33:22,360 ordered. Now this tweet with this picture 663 00:33:22,360 --> 00:33:28,120 got 1500 comments, 1100 retweets, 37,000 likes and it 664 00:33:28,120 --> 00:33:31,520 got viewed 2.4 million times. This is what they column 665 00:33:31,520 --> 00:33:35,120 business earned advertisement meaning that Texas Roadhouse. 666 00:33:35,120 --> 00:33:38,880 In order to make this incredible endorsement from a former NFL 667 00:33:38,880 --> 00:33:42,160 player, probably the best person that you could partner with to 668 00:33:42,160 --> 00:33:44,880 endorse a company a restaurant like Texas Roadhouse. 669 00:33:45,320 --> 00:33:48,280 He endorsed it for free. He just went out and he enjoys 670 00:33:48,280 --> 00:33:50,040 the meal. He said that he got the meal in 671 00:33:50,040 --> 00:33:52,360 under 20 minutes. He was in and out of the door. 672 00:33:52,360 --> 00:33:54,920 The service was incredible. He gave it this glowing 673 00:33:54,920 --> 00:33:57,080 endorsement. This is one of the examples of a 674 00:33:57,080 --> 00:33:59,800 company that just offers a great product. 675 00:34:00,120 --> 00:34:02,440 The endorsements are in the customers themselves. 676 00:34:02,800 --> 00:34:05,600 The reason that we don't see any national advertisements, it's 677 00:34:05,600 --> 00:34:08,080 the same reason you don't see Costco advertising. 678 00:34:08,360 --> 00:34:10,960 The product sells itself so well through the word of mouth. 679 00:34:11,040 --> 00:34:13,440 I think things are going to continue on for Texas Roadhouse. 680 00:34:13,440 --> 00:34:15,480 There's so many people that love this restaurant. 681 00:34:15,480 --> 00:34:17,159 They're opening up new locations. 682 00:34:17,400 --> 00:34:20,400 Even though the valuation is a bit higher, it's traded up some. 683 00:34:20,639 --> 00:34:22,199 I'm still along the ride for this one. 684 00:34:22,320 --> 00:34:24,679 Now moving on, we get to some news that's rather shocking. 685 00:34:24,760 --> 00:34:28,360 A United Health executive, Brian Thompson was just killed. 686 00:34:28,360 --> 00:34:31,080 He was shot down in a premeditated attack. 687 00:34:31,080 --> 00:34:33,600 Now there's video of this clear assassination. 688 00:34:33,760 --> 00:34:35,120 I'm not going to show it on this channel. 689 00:34:35,120 --> 00:34:38,400 It's all over X and Twitter and online you can find if you 690 00:34:38,400 --> 00:34:40,719 really want to watch it. But it is disturbing. 691 00:34:40,760 --> 00:34:44,639 It shows him being gunned down in a premeditated fashion by 692 00:34:44,639 --> 00:34:46,199 someone that looks like a professional. 693 00:34:46,280 --> 00:34:48,800 The assassin has a jacket on with a hood up covering his 694 00:34:48,800 --> 00:34:51,080 identity. He pulls out a handgun with a 695 00:34:51,080 --> 00:34:54,520 silencer and shoots at point blank range multiple times, 696 00:34:54,840 --> 00:34:57,520 ensuring that he's dead. This is a clear cut assassin 697 00:34:57,600 --> 00:34:59,680 fascination. Now the Wall Street Journal is 698 00:34:59,680 --> 00:35:02,920 reporting that a manhunt is underway for the suspect, who is 699 00:35:02,920 --> 00:35:05,840 lying in wait for the executive. So as of right now, they haven't 700 00:35:05,840 --> 00:35:07,760 caught this assassin. He's still on the loose. 701 00:35:07,760 --> 00:35:09,360 In fact, they don't even know his identity. 702 00:35:09,640 --> 00:35:13,400 He fled on foot after shooting outside the Hilton Hotel and 703 00:35:13,400 --> 00:35:16,800 then the suspect rode on an E bike to Central Park where he 704 00:35:16,800 --> 00:35:19,000 was last seen. Police said that he planned the 705 00:35:19,000 --> 00:35:22,760 attack, but they don't know why. So as of right now, we don't 706 00:35:22,760 --> 00:35:25,120 have a clear motive. An important thing to note here 707 00:35:25,120 --> 00:35:27,560 is a small caveat. A lot of different news 708 00:35:27,560 --> 00:35:31,360 organizations are reporting that Brian Thompson was the CEO of 709 00:35:31,360 --> 00:35:34,360 United Health. He wasn't the CEO of the entire 710 00:35:34,360 --> 00:35:36,280 company. He was the CEO of a smaller 711 00:35:36,280 --> 00:35:39,080 business owned by United Health. So he was an executive at the 712 00:35:39,080 --> 00:35:42,240 company, but he wasn't the CEO. Thompson, who was 50, was 713 00:35:42,240 --> 00:35:45,200 believed to be going to United Health Annual Investor Day early 714 00:35:45,200 --> 00:35:47,800 to help set up. So to get this straight, this 715 00:35:48,080 --> 00:35:50,800 executive that works for a company that works for United 716 00:35:50,800 --> 00:35:54,360 Health was just going to an annual investor day to help set 717 00:35:54,360 --> 00:35:58,120 things up and he was targeted by an assassin early in the morning 718 00:35:58,360 --> 00:36:00,640 in broad daylight in New York City. 719 00:36:00,920 --> 00:36:02,760 This is truly shocking and disturbing. 720 00:36:02,880 --> 00:36:04,880 Now the police, of course, are trying to find this guy. 721 00:36:04,880 --> 00:36:07,920 They've already put up a bounty for the assassin offering 722 00:36:07,920 --> 00:36:11,200 $10,000 for anyone that has any type of information. 723 00:36:11,200 --> 00:36:13,920 This whole event is just tragic and unsettling. 724 00:36:14,400 --> 00:36:17,720 Ryan Thompson, From every bit of information we can gather on his 725 00:36:17,720 --> 00:36:21,360 history, at least now it's hard to find any possible motive for 726 00:36:21,360 --> 00:36:23,400 someone wanting to hurt him. There's already news 727 00:36:23,400 --> 00:36:26,080 organizations that have dug through his public history, 728 00:36:26,080 --> 00:36:29,120 looking at every challenges he's faced, everything that he's done 729 00:36:29,120 --> 00:36:32,000 with his work and private life. There's nothing distasteful, 730 00:36:32,000 --> 00:36:35,160 objectionable, questionable. It seemed like he was all around 731 00:36:35,160 --> 00:36:37,320 a good person, his mother-in-law said. 732 00:36:37,320 --> 00:36:39,280 The only thing I can say is that he's a good man. 733 00:36:39,280 --> 00:36:41,800 I can't say anything else. We're still in shock. 734 00:36:41,920 --> 00:36:44,320 This seems like a totally meaningless, senseless attack. 735 00:36:44,440 --> 00:36:46,160 The family must be devastated as well. 736 00:36:46,280 --> 00:36:49,080 He's also married, so his wife must be completely devastated 737 00:36:49,080 --> 00:36:51,200 right now. All around a completely horrific 738 00:36:51,200 --> 00:36:54,240 event and hopefully we find the scumbag that did this. 739 00:36:54,240 --> 00:36:56,520 That's going to be it for now. See you in the next one.