Angie Quennell, Amazon Director of Global Finance and Governance Communications
What can customers look forward to in the coming years as far as delivery enhancements?

Andy Jassy, Amazon CEO: 
Fast, reliable delivery is an enduring priority for us, along with selection and low prices. It always has been, and it’s why we’ve invested so much the last couple decades in our fulfillment network, and we’ve always had a strong fulfillment network, but as we expanded so much in the pandemic, Remember we, we doubled the footprint of our fulfillment centers that we built in the first 25 years. We doubled it in about 18 months, and then we also built a last mile transportation network, the size of ups, largely in those 18 months as well. As we did that, we ended up with a much larger network that was different and had a lot more facilities and nodes and associated connections. And as such, it caused us to rethink everything in our fulfillment network. A good example of that is our US fulfillment center network. Historically, we had a national flat network, but as we have a lot more facilities and connections, we found that certain sub optimizations at prior scale such as not having inventory and a facility close to an end customer. It just got amplified at this new scale, this new elbow of the curve and scale. And so we made the decision to regionalize this network, and what we did was we created eight interconnected regions in smaller geographic areas to better serve customers through lower cost and faster speeds. Each of these regions has broad, relevant selection to operate in a largely self-sufficient way with some of the most meaningful and hard work that we had to do coming from optimizing the connections between this large amount of infrastructure. We rolled out this regional network in Q1 and we’re really pleased with the results. Thus far, we’ve got about 15% reduction in the distance Items are traveling from fulfillment centers to customers. 12% decrease in touches are how often a package is handled. 76% of the products customers are ordering are now in facilities within their region versus 62% a year ago. That’s a big difference in a short amount of time and. What we’re fighting is that customers are using same day delivery at a, at an increased rate, about 50% year over year. And in the first quarter alone, we saw it reaching nearly 26 million customers doing same day delivery. So shorter distances traveled means faster delivery speed, lower costs, and better results for the environment. And we’re on track this year in 2023 for our fasting fastest shipping speeds ever for prime customers. Now you’re seeing us continue to find ways to make it easier and easier for customers to get packages quicker. We have one day delivery now on more than 20 million items and growing same day delivery, which gets products to customers in just a matter of hours, is available in 90 metro areas with selection of hundreds of thousands of items. So we’re not close to being done working on improving the speed of delivery for customers because we know it matters to them so much. So a lot more coming.

Angie Quennell, Amazon Director of Global Finance and Governance Communications:
Your next question is, what is Amazon doing in ai?

Andy Jassy, Amazon CEO
We’ve been using machine learning and AI at Amazon for 25 plus years, and it’s been really deeply embedded into every business in which we operate. And you can see it all over Amazon. If you look at the personal recommendations in our e-commerce business that comes from ai, if you look at the pick paths that our associates and our fulfillment centers Pick items that those are AI algorithms. You look at Prime or our drones, or if you look at our ghost stores or if you look at xa, those are all fueled by machine learning and ai. And so it has been a very big part of our heritage for 25 years. Now, over the last nine months or so, you’ve seen this new inflection point for AI with the advent of generative ai and it’s just made it much more accessible to a lot more people and. We’re quite excited about it as well. I think it’s still very early days in generative ai, but moving quickly and we see it as having three key layers, if at the bottom layer. Every company that does generative, generative AI is gonna train models and it’s gonna run inference or predictions from those models on compute. And the key factor in that compute is gonna be the chips that’s in that compute. And to date, there’s been. Pretty significant scarcity in the amount of in the capacity of chips out there and in the options for customers on what they could run chip wise there. And so it’s why we’ve been investing for several years in our AWS business in building customized machine learning chips. And we have one that’s focused just on training that we call cranium. And we have one that’s focused just on inference that’s called inferential, and we’re on the second versions of both those chips. And the price performance combination for those training and inference chips is very compelling relative to alternatives out there. So we’re very optimistic that a lot of these models will be trained on these AWS training and inference chips. Then the middle layer of that stack really is around the large foundational models that people are gonna build generative AI on top of. And if you look at these if you look at these leading models, they take. Billions of dollars and many years to ultimately train, and there’s gonna be relatively few companies that wanna pursue that. What most companies tell us they want is they want to use somebody else’s very large foundational model and then customize it. On their own data, making sure that their own data doesn’t leak to the general model of being able to customize that foundational model with their own data for their own application purposes. And then being able to have all the same capabilities they’re used to using in the rest of their technology infrastructure plat platform. And that’s why we announced Bedrock, which really is foundational models as a service. And so in bedrock, We will expose some of our own large language models that we’ve been working on for several years that we call Titan.

But we’ll also give customers access to other third party foundational models like anthropics, like stability, ais like ai, 20 ones, as people will use lots of different models, and they’ll be able to use those as the foundation to then customize with their own data, with the same security, the same privacy, and all the same platform features they’re used to using and running their workloads in aws. That they can run with these bedrock models as well. And then that type layer of the generative AI set of solutions is really the applications that you’re gonna build on top of these foundational models. And they’re gonna be a lot of these, we think one of the high potential early ones is a code assistant, and that’s why we built code Whisperer that allows customers to use natural language to say things like, I want to build a video hosting website, and then Code Whisperer generates that code for them to just deploy. It’s a huge game changer with respect to developer productivity and it’s got a lot of early traction, but there will, we’re pursuing lots of applications ourselves in inside of Amazon. Every single one of our businesses is using generative AI to reinvent customer experiences. And while we’ll pursue a number of those applications, many more will be built by external companies and developers. And we’re quite optimistic that a lot of those applications will be built on top of aws. So it’s very early days in generative ai. It’s very high potential and and we’re investing quite a bit in it and expect to be a leader.

Angie Quennell, Amazon Director of Global Finance and Governance Communications
Thanks Andy. What can you tell us about the strategy for investing in Kiper and other new business ventures like it? Are they worthwhile?

Andy Jassy, Amazon CEO: 
We think about our big investments with a series of questions that we ask ourselves and the questions we tend to ask ourselves are, number one, if we’re successful, can it be big and move the needle for Amazon with good return on invested capital returns? Number two, is it being well served today? Number three, do we have a differentiated approach? And number four, Do we have some competence in that area? And if not, can we acquire it quickly? And if we like the answers to those questions, then we’ll pursue that big investment. And sometimes that process has led to pretty straightforward new areas for us. These were things like getting into other product categories besides books in the early days of Amazon or getting into different international geographies. But sometimes that process has led to. Inventions and investments that weren’t as initially straightforward. And AWS is a good example of that where there were a lot of people wondering why we were pursuing a technology infrastructure platform when we were really just a retailer. And a lot of people questioned it internally and a lot of people questioned it externally. But think about what a different company Amazon would be today if we hadn’t pursued aws. And Kiper to me is an example of one that might not have been predicted by people, but makes a lot of sense when you look at that structural outline.

There’s over 400 million plus households that today have little to no connectivity or broadband. And that’s all capabilities and technology access that we take for granted today. But just imagine what. People who can’t do so today could do with reliable connectivity. Being able to do things like taking online education courses or using financial services, or starting their own businesses, or doing shopping or enjoying entertainment. Or just think about businesses and governments that need that broader coverage to improve their operations. So Kiper will change this for many people. It will be much more affordable and higher performance than alternatives that are out there today we’re launching. Two prototype satellites to test the end-to-end communications network this year, and then plan to be in beta with commercial customers in 2024. We’ve released some of the technical details recently and already consumers, enterprises and governments are quite excited about us providing kiper. So we believe inventions like Kiper have a chance to change what’s possible for many millions of customers and to be a very good business for Amazon.

Angie Quennell, Amazon Director of Global Finance and Governance Communications
Your next question is, while Amazon Ads is a big Dr. Revenue driver for Amazon, how do you balance that with ensuring a great customer experience for shoppers?

Andy Jassy, Amazon CEO: 
Customer experience will always be paramount for us. Everything starts with the customer and our strategies and tactics evolve backwards from there. That’s always been true and always will be true. Most of our resource and time that we spend in advertising is spent on machine learning algorithms to ensure that the sponsor results people see on search results are relevant. To what they’re trying to find, and we track it rigorously and we’re constantly refining the algorithms. Advertising, of course, is not new in retail. Retailers have been doing this for many decades, selling shelf space or end caps or placements in circulars. But unlike physical retailers, we can tailor the results to be relevant to what customers are actually searching for, given what we know about. Buying behaviors. And because our ads are relevant for customers and much more so than you find in physical retail, customers click on those sponsor links a lot, which means they perform better for brands. And this is a key reason why our ads business is continuing to grow at a good clip, despite the fact that we’re at a very challenging time for ads heavy businesses. I think it’s also worth noting that we’re still in the very early days of working through thoughtful placement in our video, live sports, audio, and grocery products. It is just early days for us in advertising and we’re excited about it.

Angie Quennell, Amazon Director of Global Finance and Governance Communications
Next, a shareholder asks, international is so flaky with volatile currencies and administrative situations. Why are you even playing in that area?

Andy Jassy, Amazon CEO: 
We’re pretty excited about our international business. I think if you look at 2022 our international consumer segment was 118 billion in revenue and that growth re accelerated in Q1 of this year over Q4 of last year. And if you look at our international. Consumer business. It’s a combination of established geographies. These are places like the UK and Germany and Japan and emerging countries that we’ve launched over the last several years. Now, our established international businesses are big enough at this point that they’re not immune from macroeconomic issues. And you could see that over the last year. But if you look at the compound annual growth rates of these established international businesses from 2019 to 2021, they’re pretty remarkable. 31% in the uk, 26% in Germany, 21% in Japan. That’s pretty unusual. So we’re optimistic about this how those business will continue to evolve and grow as we emerge from the current macroeconomic environment. I think we also have a number of new countries that we’ve launched over the last several years, and these are places like India or Brazil or Mexico or Australia or lots of different countries in Europe, in the Middle East and Northern Africa. And what we find with these new geographies is that it takes a certain amount of fixed investment to get these locales going. And then you need a certain amount of revenue scale where the operating income and margin covers the fixed investment. But we’re. We’re really liking the trajectory. It looks very much for the most part on the same trajectory as what we’ve seen in the US in our longer standing established international countries, and we’re quite bullish on building a large, profitable international stores business. We’re on the right track.

Angie Quennell, Amazon Director of Global Finance and Governance Communications
Thanks Andy. This will be the last shareholder question. I’m cur, I’m interested in the company’s foray into health and think there’s tremendous opportunity in service and care, but I’m curious where you are thinking you are going to take this. Healthcare, the healthcare experience. Across the world, but

Andy Jassy, Amazon CEO: 
I’ll just start with the US is a pretty rough experience for customers. Most customers are not happy with it. And as such, our customers have been asking us for many years if we would provide a pharmacy solution. And it, if you think about it, it’s not that far an extension from what we do in retail. There’s some regulatory pieces that are different, but it has many similarities. And so we launched a pharmacy offering an Amazon pharmacy a few years ago. It’s off to a good start. It’s growing quickly and as we’ve. Built that business and as customers have gotten used to, to using us in the healthcare space, not surprisingly, they’ve asked us to help with more issues in their everyday healthcare experiences. And, if you think about for most customers, if you really wanna make a change and a meaningful impact in their healthcare experience, Primary care is right in the middle of that experience. And just think about how we’ve all experienced primary care over the last several decades. I think about telling my grandkids that the way it was that you had to call somebody three to four weeks in advance for an appointment, drive 20 minutes to the doctor’s office, park, wait, and reception for 15 minutes. Then get put in an exam room, wait in the exam room for a doctor for 10 to 15 minutes, then the doctor comes in, speaks to you for about five to 10 minutes, and then you drive 20 minutes to the pharmacy. That is our grandkids are not gonna believe that was the way that we got primary care. And it is not gonna be the case moving forward. It’s just not. And so we, we’ve been thinking about this for a while and we were experimenting with a business that we called Amazon Care, which was telemedicine. And we could not believe how much customers liked that customer experience. It turned out that we had the wrong business model for it, which is why we didn’t see it moving forward and shut it down.

But the Amazon care experience really opened our eyes to the fact that the experience was not gonna. Was not just gonna change over time, but that people wanted to change now. And we came across this company in One Medical who we were just so impressed with. If you look at the how strong and how easy to use their digital app is with all of your healthcare information readily available in the app, if you want to have a chat with a A medical practitioner you can easily do so if you want to have a video conference with a medical practitioner, you can do if you need to come in and see somebody. One medical has clinics all over the country, you can do so same day or next day very quickly. They ha all the one medical. Physical presences have relationships with health service providers in cities. So if you need to see a specialist, you can do so very quickly. And then whatever medication you need, you can use Amazon Pharmacy or another third party pharmacy if you like to get that quickly without having to travel 20 minutes back and forth to the pharmacy set. That’s a very different experience, and that’s why we became so excited about One Medical and decided to acquire the company. And I think that we believe together in Amazon and One Medical together that we have a chance to meaningfully change the customer experience for so many customers in what they experience every day with healthcare. And we’re very excited about that.