Latest / Automated with Brian Heater / Daniel Rausch on How Alexa Was Rebuilt for the AI Era

Transcript

Daniel Rausch on How Alexa Was Rebuilt for the AI Era · Automated with Brian Heater

• 0:00
  SPEAKER_00: Alexa's an incredibly complex system of many models and many technologies, and a bunch of them enforced guardrails.
• 0:06
  SPEAKER_00: She was uninfluenza.
• 0:07
  SPEAKER_00: I'm inspired by Amazon's desire to invent.
• 0:11
  SPEAKER_00: If it's anything, it's really an invention machine.
• 0:14
  SPEAKER_01: Does the speed at which things are evolving in large language models or anything on the AI side, does that ever kind of keep you up at night?
• 0:22
  SPEAKER_00: The speed in the space, I think, Frank, it feels like we're going fast, and then you you look back and say, oh my god, we're going even faster now.
• 0:28
  SPEAKER_01: Once these LLMs come along, is there a sense in which development and the way things are built really did have to kind of scrap everything and start from scratch?
• 0:38
  SPEAKER_00: Oh, for sure.
• 0:39
  SPEAKER_00: I mean, it's a total re-architecture of all of the technology behind Alexa to bring Alexa from this era of voice assistance into the era of foundational AI assistance.
• 0:48
  SPEAKER_01: There's something there you hear, Chat GPT comes along and hear like genuine horror stories about people that get a little too, you know, maybe emotionally invested in a system.
• 0:57
  SPEAKER_00: Alexa's cognizant that she is AI.
• 1:01
  SPEAKER_00: You would recognize the names of the other two chatbots that she was being compared to that were the best they did was only 60% of the time be willing to change the facts.
• 1:10
  SPEAKER_00: You just can't, it's irresponsible.
• 1:12
  SPEAKER_01: Like you can't build your own echo chamber around Alexa, is what you're saying.
• 1:29
  SPEAKER_01: Welcome back to Automated.
• 1:31
  SPEAKER_01: I'm Brian Heater, the managing editor at the Association for Advancing Automation.
• 1:35
  SPEAKER_01: Uh, this week we are chatting with Daniel Rausch, Amazon's vice president of Alexa and Echo.
• 1:40
  SPEAKER_01: It's a little bit different than the usual automated chat.
• 1:44
  SPEAKER_01: Thankfully, I have had a couple of decades as a hardware journalist to lean back on here.
• 1:50
  SPEAKER_01: The interview comes as Amazon is rolling out Alexa Plus, which is its version of uh the smart assistant for the LLM era.
• 2:00
  SPEAKER_01: There's uh a lot in here for everyone who's interested in generative AI, privacy, and how a corporation effectively rethinks a product from the ground up.
• 2:10
  SPEAKER_01: So uh thanks to Daniel, thanks to Amazon, thanks to you, as always.
• 2:14
  SPEAKER_01: Uh, if you're enjoying the show, please like, subscribe, check out our newsletter over at automated.fm.
• 2:21
  SPEAKER_01: All right, and with all of that, please enjoy this interview with Daniel Rausch of Amazon.
• 2:25
  SPEAKER_01: You know, we talk a lot about what's coming up next in automation on this show, but if you really want to see the future motion, you've got to be there in person.
• 2:33
  SPEAKER_01: Automate 2026 is where the world's leading innovators, builders, and dreamers come together to show you what's possible.
• 2:40
  SPEAKER_01: Robots, AI, machine vision, motion control, you name it, all automation under one roof.
• 2:46
  SPEAKER_01: And as part of Automate this year, the Humanoid Robot Forum brings together leaders, engineers, and researchers for a two-day deep dive into the real world development, deployment, and commercialization of humanoid robotics.
• 2:59
  SPEAKER_01: Register for free at automateshow.com to join us in Chicago June 22nd through the 25th.
• 3:06
  SPEAKER_01: We will see you there.
• 3:08
  SPEAKER_01: Are you aware that you have uh an IMDB page?
• 3:11
  SPEAKER_00: Someone with my name uh has an IMDB page, yes.
• 3:15
  SPEAKER_01: Okay.
• 3:16
  SPEAKER_01: It's a source of some confusion because I'm pretty sure that the last credit is in fact you.
• 3:20
  SPEAKER_01: It is smart home.
• 3:21
  SPEAKER_01: Oh, really?
• 3:22
  SPEAKER_01: But you you were not a child actor in Sweden uh during the 80s and 90s.
• 3:27
  SPEAKER_00: Uh my Swedish just isn't quite good enough, but um, but I'm working on it.
• 3:32
  SPEAKER_00: I'm working on it.
• 3:33
  SPEAKER_01: Okay.
• 3:33
  SPEAKER_01: But you did you did go to University of Massachusetts and you did study philosophy and history and minor in mathematics.
• 3:41
  SPEAKER_00: Is that all correct?
• 3:42
  SPEAKER_00: That's correct.
• 3:43
  SPEAKER_00: And yes, as a as a way to start my career studying mathematics and logic, yes, that's that was the foundation.
• 3:50
  SPEAKER_01: Oh, interesting.
• 3:50
  SPEAKER_01: I was I was gonna ask about that.
• 3:51
  SPEAKER_01: So so philosophy and history, that that was all part of the uh basis of math?
• 3:58
  SPEAKER_00: Yeah, I um deeply interested in sort of the what's known as the philosophy of science or philosophy of mathematics was sort of part history, part study of what um the role that mathematics and science has played, the way that uh scientific and mathematical revolutions happen.
• 4:17
  SPEAKER_00: So how you create whole new spaces in mathematics or logic or science.
• 4:22
  SPEAKER_00: Um and then when I went to graduate school, I branched off into studying uh formal symbolic logic and uh basically proof theory and how you build mathematical systems.
• 4:34
  SPEAKER_01: Yeah, as a as a creative writing major who you know now obviously covers uh very technical things that I do not, in fact, have a background in, I'm very interested in this intersection between the humanities and and the sciences.
• 4:49
  SPEAKER_01: Do you find that as your career has tended towards AI and as we're exploring these new spaces, that those foundational pieces that you were studying, like philosophy and history, have actually been useful?
• 5:06
  SPEAKER_00: I mean, for me, yes, for sure.
• 5:09
  SPEAKER_00: Um I ended up branching, I became an entrepreneur and worked with some partners on, you know, basically building a company when I was in graduate school.
• 5:17
  SPEAKER_00: Um and even back at its origin in my career, you know, I was very interested in the ways you could use computers and deterministic systems to create different outcomes for different participants in some software.
• 5:33
  SPEAKER_00: So uh computer adaptive testing became a big interest.
• 5:36
  SPEAKER_00: And uh I worked with some partners on a company that did educational technology.
• 5:40
  SPEAKER_00: Um, we were sort of early on in the process of figuring out that the best way to actually uh do assessment of someone's um understanding of material would be customized to that student.
• 5:51
  SPEAKER_00: And, you know, if you if a student's doing incredibly well, you give them progressively hard problems.
• 5:56
  SPEAKER_00: If they're doing poorly on a particular, you know, standard, state standard, say in science or math, and you give them progressively easier problems to find out exactly where the threshold of their understanding is.
• 6:07
  SPEAKER_00: Uh and at the time that was sort of a brand new approach.
• 6:09
  SPEAKER_00: Now, you know, my kids are taking college prep uh uh you know assessments, and every test sort of works this way that you take on a computer now.
• 6:17
  SPEAKER_00: But at the time it was it was brand new.
• 6:19
  SPEAKER_00: And building deterministic systems to do this is one thing, but building non-deterministic systems where you get different outcomes and have different experiences of a technology, basically the thread that is sort of goes through my whole career and certainly leads us to AI today.
• 6:35
  SPEAKER_00: So it's been an interest of mine, I think, pretty much from the beginning of my career as an entrepreneur and certainly has some direct foundation in some of the work I was doing in Modal Logic, which would be too exhausting for anyone to have to listen to any discussion about.
• 6:51
  SPEAKER_00: So maybe I'll just flag it and we'll move on from that.
• 6:54
  SPEAKER_01: Yeah, I'm I'm really curious, you know, as somebody, especially as somebody who has been covering um, you know, as you know, I was in the consumer electronics space for a really long time.
• 7:05
  SPEAKER_01: So I was covering um specifically like echo devices, I was covering Amazon Alexa, and I had sat in on a number of uh those um uh Alexa developer seminars, um, you know, and there's a lot of talk about how in the future AI is going to evolve and change things.
• 7:24
  SPEAKER_01: But once something, once these LLMs come along and really obviously a game changer in so many ways, I I mean, is there a sense in which development and the way things are built really did have to kind of scrap everything and start from scratch?
• 7:44
  SPEAKER_00: Oh, for sure.
• 7:45
  SPEAKER_00: I mean, I think the the key product challenges with something like building Alexa Plus are figuring out how to benefit from everything you've already built for customers and the things that they love.
• 7:57
  SPEAKER_00: They love talking to Alexa.
• 7:58
  SPEAKER_00: I mean, I forget how many tens of millions of times customers had, you know, asked Alexa to marry them or for science jokes or whatever.
• 8:06
  SPEAKER_00: I mean, the the the there's a beloved technology uh that not just doesn't just live on someone's laptop or phone, but people brought it Alexa into their homes and included Alexa in their lives.
• 8:17
  SPEAKER_01: How did you how did you deal with the the the marriage proposals previously?
• 8:21
  SPEAKER_01: I mean, you gotta let the person down gently, right?
• 8:24
  SPEAKER_00: We had some pithy answer, yes, that you know, like you know, about being an AI and how that's somewhat different, but it's flattering.
• 8:31
  SPEAKER_00: You know, I mean Alexa has personnel attributes like being humble and being pithy and loving science, and you know, like there's all kinds of things that go into responses to those things.
• 8:39
  SPEAKER_00: But it used to be a much more deterministic system, a rules-based system, right?
• 8:43
  SPEAKER_00: But you want to take advantage of everything that you had built before, the capabilities we manifested for customers, what woven deeply into smart, this fabric of customer smart homes, right?
• 8:52
  SPEAKER_00: We sold over half a billion Alexa-enabled endpoints, and customers connected way over half a billion, I don't even know the count anymore, smart home devices to Alexa.
• 9:01
  SPEAKER_00: So you're just you're woven into the fabric of a household of customers' lives, you're answering all kinds of questions.
• 9:06
  SPEAKER_00: You want to, you want to build on that, but from a tech, that's the product challenge, right?
• 9:12
  SPEAKER_00: To sort of continue that, but make Alexa just much more capable.
• 9:16
  SPEAKER_00: On the technical side, you're trying to make sure you're manifesting all those capabilities built entirely anew and really replace the core.
• 9:26
  SPEAKER_00: I mean, it's a it's a total re-architecture of what uh of all of the technology behind Alexa to bring bring Alexa from this era of sort of voice assistance into the era of foundational AI assistance, where she's now, you know, one of these foundational AIs that can is just wildly more capable and not just not just getting things done, but much deeper conversations, deeply personalized, you know, to each customer, incredibly smart on any, you know, can really have a conversation on any topic at all, and gets a ton of things done.
• 10:01
  SPEAKER_00: And that, you know, a lot of those things are what really differentiates Alexa, in particular the last one, getting things all the way done.
• 10:08
  SPEAKER_00: I'm sure we'll spend more time talking about it, but uh but it's a real challenge, and you're talking about you know an 18-month you know, incredibly hard effort build cycle to get that done once the models are really sort of capable enough.
• 10:22
  SPEAKER_01: Previously, let's take the marriage example, that was a case in which you had to really hardwire a response into Alexa, knowing that at some point somebody is going to ask to propose marriage to this device.
• 10:37
  SPEAKER_00: Correct.
• 10:38
  SPEAKER_00: It's a rules-based system that would give an answer to anything.
• 10:40
  SPEAKER_00: I mean, at the beginning we had, you know, an incredibly talented team of comedy writers helping us build a database of jokes and responses.
• 10:47
  SPEAKER_00: And now you're talking about a you know non-deterministic system.
• 10:51
  SPEAKER_00: There are over 70 uh models that back Alexa, that we choose for different tasks, that we train differently, that we fine-tune differently, that um we count on for different parts of the infrastructure.
• 11:04
  SPEAKER_00: Uh and really it's about influencing the outcomes of those things, having the right training data, having the right post-training, doing the right reinforcement learning, et cetera, to get the experience that you're looking for that's built on sort of um the tenets and foundations that you want for that experience, but it's incredibly flexible and dynamic.
• 11:23
  SPEAKER_00: So uh a very different approach to building that.
• 11:26
  SPEAKER_01: Yeah, and and and if Alexa hadn't had an answer before, you know, Alexa would tell you that they weren't able to respond, um, which is obviously a frustrating experience from a user standpoint, but is also, in a way, kind of a safety net for you, right?
• 11:41
  SPEAKER_01: I mean, you know, we just don't quite have an answer to that.
• 11:44
  SPEAKER_01: So we don't have to really worry about the response that the system is giving to that question.
• 11:51
  SPEAKER_00: Yeah, I would say, you know, one of our one of one of our early tenets as we realized the range of possible experiences we could build with Alexa Plus was no dead ends.
• 12:02
  SPEAKER_00: Like now we have an opportunity to build a product where maybe we can explain ourselves and help a customer understand where to get some job done that they have and a capability we haven't quite added yet.
• 12:12
  SPEAKER_00: But uh we'll at least be able to talk about it.
• 12:14
  SPEAKER_00: One of the things we realized when we when we built the no dead ends tenet was that we, you know, Alexa has to be an expert on herself.
• 12:21
  SPEAKER_00: She has to be able to tell you how to get anything done that she can do with her, uh, how to change a setting, you know, if if something's not quite working the way you expect yet, for example, or um, you know, one of the first things you encounter in the Alexa Plus experiences you uh that we that we learned as as we were in early access was Alexa offers to change her voice if you want, because the new voice uh is just much deeper, richer, expressive.
• 12:48
  SPEAKER_00: There's a much fuller range of things we can do, but there's a small but vocal subset of customers, quite small in fact, but quite vocal, uh, that want uh quote unquote the old voice or the original voice.
• 13:00
  SPEAKER_00: So she explains right away how to get that done.
• 13:02
  SPEAKER_00: And it's just part of her reflecting her expertise on herself.
• 13:05
  SPEAKER_00: So not only can you go off and then you've learned that you can change the voice to any of a range of awesome voices, but you also realize that you can ask Alexa what she can do and how to get something done with her.
• 13:17
  SPEAKER_01: Yeah, this this is uh an example that I go back to a lot, and actually I have one behind me, but you know, I don't know how I'm I'm sure you actually are pretty familiar with Anki when they came along and they had the Cosmo robot.
• 13:28
  SPEAKER_01: And in much the same way that you were uh employing comedy writers, they had hired a bunch of folks from uh Pixar and Dreamworks to animate the eyes on the robot.
• 13:38
  SPEAKER_01: With robots specifically in the home, it's very important to have a personality, right?
• 13:43
  SPEAKER_01: Um, and even like a Roomba, people will sort of project a personality onto those systems.
• 13:49
  SPEAKER_01: Um is the idea, at least in-house when you're creating an AI like Alexa, to have a neutral personality?
• 13:56
  SPEAKER_01: Is it what the user projects onto the AI?
• 14:00
  SPEAKER_00: So I think a few things.
• 14:01
  SPEAKER_00: I think there are foundational you know components in Alexa's personality for sure.
• 14:06
  SPEAKER_00: I mentioned one before, like Alexa, Alexa is actually from Amazon in Seattle.
• 14:11
  SPEAKER_00: She has sports team preferences, for example.
• 14:13
  SPEAKER_00: She roots for the Seahawks, and the mayor's opening day was yesterday.
• 14:16
  SPEAKER_00: Uh so this is this is troubling.
• 14:18
  SPEAKER_01: As a bay, as a Bay Area native, this is all very troubling to me.
• 14:21
  SPEAKER_00: Sorry, yeah.
• 14:22
  SPEAKER_00: Sorry, we'll move, we we should move on.
• 14:23
  SPEAKER_01: You might have to go with Gem and I am gonna be on it.
• 14:26
  SPEAKER_00: It's time to move on.
• 14:27
  SPEAKER_00: But those are those are very like specific point examples, but they're also just attributes of you know, of of Alexa's personality.
• 14:34
  SPEAKER_00: She's Alexa's empathetic, for example, uh, and has you know, has been from the beginning.
• 14:38
  SPEAKER_00: Um, those are core, but also we know that customers want to tune uh they want to tune you know the Alexa experience to what they want.
• 14:47
  SPEAKER_00: So we we recently launched personality styles that let you pick from a range of some customers want really short, terse answers.
• 14:54
  SPEAKER_00: So we have a brief personality style.
• 14:56
  SPEAKER_00: Uh the brief style can can be pretty funny because it's not just brief, but it's also just pithy and on point.
• 15:01
  SPEAKER_00: And a range of styles that goes through chill and other, you know, styles that might reflect a customer's preference better, all the way up to sassy, actually.
• 15:08
  SPEAKER_00: And they're, you know, what what we see right away is of course the the best measure of of a feature like that, you know, it's success is is uh is stickiness and customers engaging with it.
• 15:19
  SPEAKER_00: And we see over 90% of customers sticking with the style that they pick.
• 15:22
  SPEAKER_00: So we know that we'll keep developing more of those.
• 15:25
  SPEAKER_00: It's just it's emblematic of um the way the customers want to be able to tune the experience.
• 15:30
  SPEAKER_01: It strikes me, you know, again, as you mentioned the marriage proposal thing, and and I'm thinking especially as a lot as Alexa uh plus is coming along, the idea here is to have an ambient system, um, a system that is really, in a sense, kind of always around and can be actively engaged with, who you can have active conversations with.
• 15:55
  SPEAKER_01: And and that does become more and more personal and people tend to personify those systems more and more.
• 16:05
  SPEAKER_01: And when that happens, um you you do have to take more and more stock of the answers that you're giving to users.
• 16:14
  SPEAKER_00: 100%.
• 16:15
  SPEAKER_00: And you know, in fact, another part of Alexa's personality is she helps you prioritize your human relationships.
• 16:21
  SPEAKER_00: So, you know, one of the things that that you hear in an experience with Alexa, and this honestly, it's a reflection of how long we've been working in the space that we have a trust and responsible AI team that's like fully staffed and they're really honestly experts in the field.
• 16:37
  SPEAKER_00: And they're great at helping shape experiences so that I think what you're alluding to is sort of these power social type relationships that people are that could could build.
• 16:46
  SPEAKER_00: I mean, there's it's funny and pithy, uh, you know, and and interesting to respond to something like a marriage proposal, but it's but it's something to it's something to have tenants about too.
• 16:55
  SPEAKER_01: Yeah, I mean that that's a little glib, and and you know, and and but you know, obviously, like, and this always happens, you know, and obviously a lot of tech reporting tends to be sensationalistic, but you know, there's something there.
• 17:05
  SPEAKER_01: You hear Chat GPT comes along and you hear like genuine horror stories about people that get a little too, you know, maybe emotionally invested in a system.
• 17:13
  SPEAKER_00: Alexa is cognizant that she is AI, she's not a human being.
• 17:19
  SPEAKER_00: Uh helps you prioritize your human relationships.
• 17:21
  SPEAKER_00: And that's sort of one of our foundational trust components in the experience overall.
• 17:25
  SPEAKER_00: And it's been through years and years of learning.
• 17:27
  SPEAKER_00: I know others are coming to the space a little bit more recently, but I think the I think we've been at it for some time.
• 17:35
  SPEAKER_00: There was another interesting piece of reporting recently about how effective Alexa is at deflecting customers trying to change the news, forcing you know, the model, quote unquote, which you can do if you're really just experiencing a model directly.
• 17:50
  SPEAKER_00: Alexa is an incredibly complex system of many models and many technologies, and a bunch of them enforce guardrails.
• 17:56
  SPEAKER_00: Like Alexa passed, and I'm this, I'm not here bragging about it.
• 18:00
  SPEAKER_00: I want the whole space to advance to the stage.
• 18:02
  SPEAKER_00: She was uninfluenzable, and you would recognize the names of the other two chatbots that uh she was being compared to that were the best they did was only 60% of the time be willing to change the facts.
• 18:14
  SPEAKER_00: You just can't, it's irresponsible.
• 18:16
  SPEAKER_01: Like you can't build your own echo chamber around Alexa, is what you're saying?
• 18:21
  SPEAKER_00: Yeah, yeah.
• 18:21
  SPEAKER_00: We that at least in their tests, I'm not saying in no tests, and I'm not saying Alexa's perfect.
• 18:27
  SPEAKER_00: I'm just saying in a panel of tests by, you know, an authority on testing for influencibility.
• 18:34
  SPEAKER_00: And and the reason is because there are you either have these systems or you don't.
• 18:38
  SPEAKER_00: You either have systems that help you enforce guardrails on your AI behavior or you don't.
• 18:43
  SPEAKER_00: And I think the what I what I want out of that is not for Alexa to be differentiated in that way.
• 18:49
  SPEAKER_00: I actually want just want to see the industry catch up to where it needs to be.
• 18:52
  SPEAKER_01: Obviously, a lot of times these things ebb and flow.
• 18:55
  SPEAKER_01: And it seemed like for a while there the smartphone smart home was was ebbing a little bit and had an opportunity to really come back on the heels of of generative AI.
• 19:07
  SPEAKER_01: Um is there a sense in which you've had to kind of bring people back and and and sort of not re-educate them, but um sort of reintroduce them to Alexa and show them the ways in which these systems have advanced from where they were?
• 19:24
  SPEAKER_00: I mean, even for our most passionate ongoing users, uh, you know, it we we have to help these customers realize the potential of what you can do with this new technology.
• 19:34
  SPEAKER_00: You can imagine what my house looks like.
• 19:36
  SPEAKER_00: Uh, and I've got some very you know, avid, but but plenty critical uh customers in my own home uh that use that use Alexa all the time between my wife and my three kids uh or my parents at their place.
• 19:49
  SPEAKER_00: Um and it's for a habituated Alexa user that was using the original experience, you know, there's just as much to learn about the range of things you can do with the new experience and get into the habit of just speaking naturally.
• 20:02
  SPEAKER_00: You can speak exactly how we're speaking to each other right now.
• 20:05
  SPEAKER_00: But it's a learning process for anyone, whether it's someone who said, you know what, actually, these smart home technologies aren't for me, who knows how long ago, or someone who's been avidly using it.
• 20:16
  SPEAKER_00: The sea change in the in what you can do, how conversational it is, just like this.
• 20:22
  SPEAKER_00: Uh, ums and ahs and continuity across devices.
• 20:26
  SPEAKER_00: I do something to help with homework on Alexa.com, and then you study the quiz in the echo the next morning in the over the breakfast table.
• 20:34
  SPEAKER_00: I saw my daughter do this uh a couple weeks ago, you know, just to review the chemistry symbols that she needed help with before rushing out the door.
• 20:44
  SPEAKER_00: And it's because she's brought on these new behaviors.
• 20:46
  SPEAKER_00: She, you know, used Alexa.com to figure out which elements don't quite match their chemical symbol, which those are always the hardest to remember.
• 20:54
  SPEAKER_00: And I was reminding, I was reminded by listening to her do it, lead is PB, and you know, got all these symbols that vary from from the word that we use to denote the the element.
• 21:04
  SPEAKER_00: And and then she's just you know, out of habit using the echo in the kitchen to say, hey, which ones was I getting wrong?
• 21:12
  SPEAKER_00: Can you remind me?
• 21:13
  SPEAKER_00: Oh, antinomy, what you know, tin, whatever they were.
• 21:17
  SPEAKER_00: Uh, and then she's out the door and she does great on the quiz.
• 21:20
  SPEAKER_00: So they're they're all these new, you know, they become habits quickly because we're habit-forming creatures when something works.
• 21:27
  SPEAKER_00: Uh, but they're new habits.
• 21:29
  SPEAKER_01: Yeah, I if your parents are anything like my parents, and and most parents in general, that that's probably uh they've been quite a font of criticism for you know for the way ways in which these sim systems are integrated into uh their lives.
• 21:45
  SPEAKER_01: I mean, what's what's some of the feedback that you've gotten from your your own parents as far as Alexa and Echo in the home?
• 21:51
  SPEAKER_00: I mean, they love you know, there's there are things that they just love and are marveled by, honestly.
• 21:56
  SPEAKER_00: Uh and you know, okay, but what don't what don't They love like what's let me do the love first.
• 22:02
  SPEAKER_01: Okay, do the love.
• 22:03
  SPEAKER_01: Okay.
• 22:04
  SPEAKER_00: The love is like the photo reels of their grandkids, right?
• 22:08
  SPEAKER_00: I mean, just seeing, and and they they turned on, you know, maybe a month ago, we launched something called Photo Gallery Unlimited.
• 22:16
  SPEAKER_00: So they have that on their echo shows, and it's just a continuous stream of photos.
• 22:21
  SPEAKER_00: Uh, one of the features does the same kid, you know, one of mine, but from when they were five years old next to a similar pose, you know, more recently, like the AI can just generate these mashups and amazing uh tier rendering things for them that they just love.
• 22:39
  SPEAKER_01: It's so funny because I I've been in this industry long enough to remember when they were first trying to make digital photo frames a thing.
• 22:45
  SPEAKER_01: And it feels like maybe we finally got to a point where like they're actually a thing and actually useful and people actually want them.
• 22:51
  SPEAKER_00: Well, and now you can, you know, just being able to see it, bookmark it, favorite it, you just being able to talk to the device, ask about the photo, when's it from, et cetera, like it's an amazing experience.
• 23:03
  SPEAKER_00: They love that.
• 23:04
  SPEAKER_00: Um, my dad loves getting smart home devices all set up.
• 23:09
  SPEAKER_00: The front door is downstairs.
• 23:10
  SPEAKER_00: They don't get up and down the stairs quite as simply and easily anymore, bringing up the ring doorbell anytime someone's at the door and just being able to communicate that way, unlock the door, let someone in, et cetera.
• 23:20
  SPEAKER_00: Like that's it's so much easier now.
• 23:23
  SPEAKER_00: At the end of the first month with Alexa Plus, we see that smart home usage among people that were already smart home customers, meaning lots of devices everywhere.
• 23:33
  SPEAKER_00: It's up 50% at the end of the first month and it keeps growing.
• 23:37
  SPEAKER_01: And there's a little bit of there's gonna be a little bit of bias there, right?
• 23:40
  SPEAKER_01: Because those are early adopters generally, the people who know this is this is in that customer's home.
• 23:44
  SPEAKER_00: I'm just talking about the same, the same customer.
• 23:48
  SPEAKER_00: So that that's the baseline.
• 23:49
  SPEAKER_00: That's the the they've they're up 50% over the over their own baseline on average when you aggregate before.
• 23:55
  SPEAKER_00: But it's not because they're unlocking the door more or turning on and off the lights more themselves.
• 23:59
  SPEAKER_00: It's because everyone in the household can now use it because you can just speak naturally.
• 24:04
  SPEAKER_00: So the the change, and I saw this change in my parents' house.
• 24:07
  SPEAKER_00: Like one of the things they love about it is that now anyone there, the dog walker, doesn't have to learn everything, can just speak to the device.
• 24:14
  SPEAKER_00: We see the dog walker help me with my mom, you know, who wasn't the smart home hacker in that household, is able to use uh Alexa to control the devices in their home and drop in on the front door and do all kinds of things like that, where it just wasn't possible before.
• 24:28
  SPEAKER_00: So I think those are some of the delights.
• 24:31
  SPEAKER_00: You know, I get feedback on one of the great things about customers, it's something actually Jeff Bezos used to say they're divinely discontent.
• 24:37
  SPEAKER_00: So you're marveled by these new habits that you can form right away.
• 24:41
  SPEAKER_00: But then of course, you want, you know, you want more.
• 24:43
  SPEAKER_00: Uh they want more smart home device control, they want more uh, you know, integration with communications.
• 24:50
  SPEAKER_00: I'll get emails my dad'll send me through Alexa, a news summary that he loved or something, but he wants to be able to do more that way.
• 24:56
  SPEAKER_00: So I think uh the list goes on, but I think the sea change we see in behavior, in particular, just engagement and utility for customers when they turn on Alexa Plus is is what's most encouraging.
• 25:08
  SPEAKER_01: Does this um the this more sort of ambient way of engaging with the system um certainly in the mind of of customers, certainly in the mind of of critics and and skeptics, uh it raises more privacy issues.
• 25:22
  SPEAKER_01: But are are there more privacy issues that you as a company have to deal with?
• 25:26
  SPEAKER_00: I don't, I mean, I think uh I think what what you need to do in those areas is create the right foundation, and then you can keep leaning on it.
• 25:35
  SPEAKER_00: So I think one of the good news uh bits about building on this, you know, decade plus of having Alexa exist out in the world is that we had some of the foundational bits right.
• 25:50
  SPEAKER_00: Like our the transparency that you have over your data and the degree of control that you have over your data uh is like one of the foundational bits that customers want.
• 26:00
  SPEAKER_00: They want to see everything Alexa heard in the Alexa privacy dashboard.
• 26:04
  SPEAKER_00: And you can delete those conversations one at a time or all at once.
• 26:07
  SPEAKER_00: You can control uh how your data is used for training or not.
• 26:12
  SPEAKER_00: You can control like all of the things that um honestly, again, it's another space where it's coming up new to some uh to some players in AI and definitely not new to us.
• 26:23
  SPEAKER_00: And I think the foundation is strong, and so we just continue to build on the same foundation.
• 26:28
  SPEAKER_01: I'm actually curious how uh this sort of like transformation of Alexa itself is has already and will continue to impact the the hardware.
• 26:39
  SPEAKER_00: We are building the best Echo devices that we have ever shipped for customers.
• 26:44
  SPEAKER_00: Um I think the new, more capable Alexa Plus experience has just inspired us to really lean in on the hardware and create great uh, you know, Echo Dots have been great for a long time, but the Echo Dot Max is the best Echo Dot we've ever built.
• 27:01
  SPEAKER_00: It's you know, unbelievable sound.
• 27:03
  SPEAKER_00: People can't believe what comes out of this, you know, compact fist-sized speaker.
• 27:08
  SPEAKER_00: Um, you know, it basically gets double takes in how good it is.
• 27:12
  SPEAKER_00: And, you know, is it's inspired by one of the things we saw we started when we started Alexa Plus was again another thing that we didn't think could really grow.
• 27:21
  SPEAKER_00: Music streaming time, the time customers spend listening to music was went up by 25%.
• 27:30
  SPEAKER_00: And this is like not something you think could just flip a switch and grow 25%.
• 27:35
  SPEAKER_00: One of the biggest surprises for me is like I mentioned smart home up 50% after a month.
• 27:39
  SPEAKER_00: Would not have expected that.
• 27:40
  SPEAKER_00: Would not expect music listening to go up 25%, and it's because it's just that much more delightful to find a song.
• 27:46
  SPEAKER_00: You go through a discography and you know, you learn new facts about an artist you've always loved, but you didn't didn't know completely.
• 27:53
  SPEAKER_00: You can do a full biography and you can say, when were all these songs written?
• 27:58
  SPEAKER_00: Play them in the order that the artist actually published them and wrote them.
• 28:01
  SPEAKER_00: You can, I mean, you can do all kinds of things that you couldn't do.
• 28:04
  SPEAKER_00: I I love folk artists, I love Bob Dylan, and Alexa's even deeper.
• 28:08
  SPEAKER_00: You can geek out on Bob Dylan on the internet.
• 28:10
  SPEAKER_00: Like you can go, there's no bottom.
• 28:12
  SPEAKER_00: And Alexa's that deep, you know.
• 28:14
  SPEAKER_00: Uh, and so I learn things every time I talk to her about it.
• 28:17
  SPEAKER_00: So I think the devices are endpoints that are really inspired by this increase in activity.
• 28:25
  SPEAKER_00: The screens do much more for customers because Alexa can show you much more at this point.
• 28:31
  SPEAKER_00: Um, they're just better.
• 28:32
  SPEAKER_00: I mentioned Photo Gallery Unlimited.
• 28:34
  SPEAKER_00: That's a great way that I think we've taken the best the best Echo Show devices we've ever developed for customers and just really delighted customers even more in their homes with them with these impressive photo reels uh that just walk you back through intimate memories.
• 28:50
  SPEAKER_00: So there's a lot to say about the Echo devices and the degree of inspiration that we're getting from watching customers use Alexa Plus.
• 28:58
  SPEAKER_01: So, you know, you you answered that from the perspective, obviously, like, you know, we're we're making them better, which you know, I think has probably always been the goal, like generation to generation.
• 29:06
  SPEAKER_01: Like we want to make the sound better, we want to make the microphones better.
• 29:08
  SPEAKER_01: Um But what I'm curious about is uh, and you must just be rethinking the way people are engaging with them and the way that they're living alongside people in the home, you know, being more more ambient devices.
• 29:25
  SPEAKER_01: Maybe that means, I don't know, maybe that means they're hiding away a little bit more.
• 29:29
  SPEAKER_01: Um that that there perhaps needs to be a fundamental sort of rethink about the the hardware if there's such a sort of transformation happening on the AI side of things.
• 29:43
  SPEAKER_00: I mean, I think I think um from the core capability set perspective, I think actually there's not a fundamental rethinking.
• 29:52
  SPEAKER_00: Um the range of modalities that we present, like the options for customers, we've always wanted to represent a great device for any spot in your home, whether it has a screen or doesn't have a screen.
• 30:03
  SPEAKER_00: Could be a singing fish, for example.
• 30:06
  SPEAKER_00: Could be a singing fish, for example, but also could be a bedside clock.
• 30:10
  SPEAKER_00: Uh, also could be, you know, the great speakers to kind of weave together in a Alexa home theater for your living room, uh around, you know, around your fire TV that will form honestly the most affordable, best, you know, Dolby surround sound that is you can buy anywhere.
• 30:28
  SPEAKER_00: Uh, but then also you can talk to Alexa and have this you know exchange back and forth.
• 30:34
  SPEAKER_00: So I think I think the design points that the goal of having a device that can fit in to anywhere in your home, the different aesthetics, I think we are elevating our hardware uh in its sort of aesthetics and quality, but also in just enhancing the capabilities to be ready for Alexa Plus, which isn't just, hey, it has better sound, but it's also what's it mean when customers talk to this device, you know, two to three times more?
• 31:01
  SPEAKER_00: Not just in a transactional single exchange most of the time, but in fact in a conversation that goes on.
• 31:08
  SPEAKER_00: So when you're having two to three times more conversations and the conversations are much more back and forth, and you need to have tighter beam forming on a customer to make sure you're really focused in on the target and really great, you know, background noise reduction and echo cancellation so that you're able to really catch someone.
• 31:24
  SPEAKER_00: They they're getting help cooking, but they've turned to look at the pan, and then they're back and forth in the kitchen.
• 31:30
  SPEAKER_00: And you know, we see cooking with Alexa and using recipes is up over 500% of what it was before.
• 31:38
  SPEAKER_00: It was fine and it was transactional and turn taking before, but now it's sort of like people doing ingredient substitution and and going back to the recipe and hey, can you repeat step seven?
• 31:48
  SPEAKER_00: And what was that again?
• 31:50
  SPEAKER_00: And you know, you when you're when you're literally using Alexa to cook for 45 minutes, the hardware capabilities do need to be enhanced to support that whole end-to-end experience.
• 32:01
  SPEAKER_00: So it's just another example of like the enhancement that we're going through versus a C change difference on the on the hardware.
• 32:09
  SPEAKER_01: I wonder if this is somewhere too where perhaps um uh like another uh study like linguistics might come in handy as far as studying tones of voice, studying speech patterns, studying um uh volume, the way people are are responding and processing that as an AI in order to understand intentionality of the speaker.
• 32:38
  SPEAKER_00: Yeah, we have uh we have different ways that we understand a customer's intent that are lexical.
• 32:45
  SPEAKER_00: The content matters a lot.
• 32:47
  SPEAKER_00: You can determine a lot from what a customer says, uh, specifically in the words and phonemes and you know, utterances that accompany words, uh, you know, paralinguistic uh cues we would call that, but also in the timber of voice, right?
• 33:04
  SPEAKER_00: And in the cadence and the prosody we call that and the intonation that a customer uses when they express themselves.
• 33:10
  SPEAKER_00: And Alexa interprets all of that to understand that you know, when I checked on what happened in the ninth inning yesterday for the Red Sox, it Alexa knew that I was pretty, you know, like anxious about it, but also happy because we I had lost my internet connection.
• 33:27
  SPEAKER_00: Uh Red Sox were up one nothing last I heard, and they won the game.
• 33:31
  SPEAKER_00: So when Alexa gave me that information, she knew to be excited and happy.
• 33:35
  SPEAKER_00: She knows I'm a huge Red Sox fan.
• 33:36
  SPEAKER_00: She knows it's over.
• 33:37
  SPEAKER_00: You are a Boston guy, yeah.
• 33:39
  SPEAKER_00: Yesterday, yes, originally a Boston guy.
• 33:41
  SPEAKER_00: Uh and so that's a big deal for me.
• 33:44
  SPEAKER_00: And Alexa reflects that back to me in the experience.
• 33:47
  SPEAKER_00: So it does it matters a lot to be able to um, you know, tune the experience to a customer's input in those ways.
• 33:57
  SPEAKER_01: That's really interesting.
• 33:58
  SPEAKER_01: And and and I suspect, you know, to a certain extent, AI is is is dealing with a lot of this through through these massive models.
• 34:06
  SPEAKER_01: Um, but you were talking before about the comedians, and I was talking before about you know animators.
• 34:12
  SPEAKER_01: And and I'm wondering if if there are any other sort of, I guess, surprising fields in which uh you wherein you employ people in order to kind of build out these systems.
• 34:25
  SPEAKER_00: I mean, the the the range of brilliant people that we have uh you know working at Amazon is super, it's always humbling for me because I it just means that you've got a team that you can constantly learn from.
• 34:37
  SPEAKER_00: You know, we have we we definitely have experts in human and computer interfaces and design, right?
• 34:42
  SPEAKER_00: That's a that's a huge one.
• 34:44
  SPEAKER_00: While also honoring tenants, like we talked about before, that Alexa knows she's an AI, um, but that you want to make it a you know a very human-like exchange because that's the best interface for these technologies in many ways.
• 34:56
  SPEAKER_00: So deep experts in that kind of design, you know, are you pair them up with you know, crazy experts in industrial design and people in acoustic design, shaping the best speakers on the planet, which they I mean, honestly, uh the team that builds these echo speakers are they are they are the best audio engineers on the planet.
• 35:16
  SPEAKER_00: They are unbelievable technologists.
• 35:18
  SPEAKER_00: Um, and you you you get something that both fits into your home, presents an amazing interface in voice to a customer or in touch modality to a customer on a screen, and uh makes you do a double take because you're like, is that coming from that small device uh at the end of the day?
• 35:36
  SPEAKER_00: So and the range just goes on and on from there.
• 35:40
  SPEAKER_01: Yeah, I may maybe I'm off on this a little bit.
• 35:43
  SPEAKER_01: I'm I'm kind of curious if I obviously um, you know, uh Astra is a little bit sort of outside of your specific scope, but you know, I I cover robotics, as you know, I cover robotics quite a bit, and there's been a lot of interest in robots around the home.
• 36:00
  SPEAKER_01: And I and I'm wondering if this is sort of like shifting or or evolving people's expectations or or interests in what an AI or what smart home hardware can do around the home or what they can expect moving forward.
• 36:15
  SPEAKER_00: The age of AI is also inducing the a golden age in robotics.
• 36:22
  SPEAKER_00: You know, building control systems that move a robot arm over and then take a bolt out of a out of a cup and put it into an assembly.
• 36:33
  SPEAKER_00: Uh all of that was done the in the exactly analogous way to what we were talking about before, these rules, these deeply rules-based systems, uh, with you know, these very, very specific sets of control instructions.
• 36:47
  SPEAKER_00: Uh and robotics was doing that in this human-intensive, incredibly coding-intensive way for decades and sort of moving into maybe some new domains and spaces, but all governed by the amount of effort that it takes to invest in these very deterministic things.
• 37:07
  SPEAKER_00: And then you have to control all the conditions.
• 37:09
  SPEAKER_00: It means that, yeah, the bolt has to be exactly so in the cup every time, right?
• 37:14
  SPEAKER_00: And and has the assembly has to show up in exactly millimeter-specific spots so that the arm can do something.
• 37:22
  SPEAKER_00: And the age of AI has just made robotics blossom because all of the stochastic, non-deterministic flexibility that you get out of models, all the techniques that apply, like reinforcement learning, teaching a robot arm now that this way was bad, that way was good, so it can generalize not just that task, but generalize the use of the robot arm for any pick it up put it task.
• 37:50
  SPEAKER_00: Uh that's what's happening in robotics right now.
• 37:53
  SPEAKER_00: Um and it's it's because of the same underlying trends in what we're able to do uh with these systems.
• 38:01
  SPEAKER_01: So you so you got North America now is pretty much fully covered by Alexa Plus.
• 38:06
  SPEAKER_01: And as we're recording this, um uh I think the UK is a is a relatively new addition.
• 38:14
  SPEAKER_01: Um obviously, you know, you're doing these rollouts deliberately, um, and there are cultural differences even within the same language.
• 38:23
  SPEAKER_01: I'm wondering, like, for example, um, you know, as as American, oh, you're in Seattle, so that you know, you're pretty close to the to Canada up there, right?
• 38:32
  SPEAKER_00: I mean, you know, you could just I love skiing in Canada, for example.
• 38:35
  SPEAKER_01: Yeah, you could just you could just you know go drive up to BC.
• 38:39
  SPEAKER_01: Um uh in terms of like the cultural exchange and cultural differences between Alexa Plus in Canada and the United States, is is that a are there really that many differences?
• 38:55
  SPEAKER_00: I mean, absolutely.
• 38:56
  SPEAKER_00: I was actually, I was just a couple weeks ago, you meant you mentioned the UK, uh, or maybe, man, it feels like weeks ago, but it might have been a week and a half ago.
• 39:05
  SPEAKER_00: But uh we launched there.
• 39:06
  SPEAKER_00: I was there for the for the launch event, and just the the local specialization in uh obvious things like jargon and calling you know what we would call the trunk of a car a bonnet or you know, a legion of other ways that two people are separated by a common language, as a as the saying goes.
• 39:25
  SPEAKER_00: Uh Alexa's fluent with all of that.
• 39:28
  SPEAKER_00: But also, interestingly, you know, the UK has 40 different dialects and accents, and that's the most of any of any place on the planet that's speaking the English language within the UK.
• 39:40
  SPEAKER_00: And so uh one of the people that presented at our launch event was was a Cambridge-based scientist that we have helping lead the team there that uh makes sure that Alexa is uh you know sort of fluent and understands and is a specialist really in comprehension with that range of dialects and and different sets of jargon and idioms to express the same thing.
• 40:05
  SPEAKER_00: Because every customer, of course, that's the only customer that matters to us, is that we understand that specific customer.
• 40:11
  SPEAKER_00: But that means you have to do this incredible range of things to really bring a personal experience to everyone.
• 40:17
  SPEAKER_00: So the team works really hard on that uh so that it so that it fits into every house.
• 40:23
  SPEAKER_01: Are are are there any particular dialects like like like like I don't know, Jordi or or I don't know, maybe like French Canadian, like are there any things that have been like really tough nuts to crack?
• 40:37
  SPEAKER_00: Not not nothing that we haven't been able to, you know, kind of master with the new techniques available to us.
• 40:43
  SPEAKER_00: Um I think the range of languages in AI, you know, it the the the fluency and capability at a baseline of any of these models is represented by the amount of training data available in any particular language.
• 40:57
  SPEAKER_00: Um those those sets are filling themselves out, right?
• 41:00
  SPEAKER_00: The power of the internet is real and uh the availability of the training data uh to us is is is really diverse at this point.
• 41:08
  SPEAKER_00: But you do need you need sufficiency.
• 41:10
  SPEAKER_00: Uh so you do need enough in any given language that you're trying to support.
• 41:15
  SPEAKER_01: As we're recording this, so just yesterday I spoke to the CEO of Rhoda AI, and they're they're doing some really just you know incredible stuff with with physical AI.
• 41:23
  SPEAKER_01: One of the things that he said um is, you know, obviously there's like every day there's these incredible breakthroughs happening, there are these papers being published, and these these leads that you thought you had like suddenly get cut.
• 41:38
  SPEAKER_01: Obviously, in your position, you have to pay attention to all the research that's that's happening.
• 41:43
  SPEAKER_01: Things are moving in incredibly quickly.
• 41:47
  SPEAKER_01: I does the speed at which things are evolving in large language models or anything on the AI side, does that ever kind of keep you up at night?
• 41:57
  SPEAKER_00: I mean, if it keeps me up, it's it's it's through inspiration.
• 42:00
  SPEAKER_00: Um I, you know, our job, my job is to if we're behind in any given area, be inspired and catch up and exceed.
• 42:10
  SPEAKER_00: And, you know, that's the that's the great the great part for customers about the availability of all these different technologies.
• 42:16
  SPEAKER_00: Uh when we're in a head, maintain the lead, right?
• 42:18
  SPEAKER_00: That's the that's the purpose of of working on any of these products is to delight customers.
• 42:24
  SPEAKER_00: And I think um, you know, I'm inspired by uh Amazon's desire to invent.
• 42:32
  SPEAKER_00: Like if if it's anything, it's really uh it's an invention machine.
• 42:37
  SPEAKER_00: That's how I think about Amazon.
• 42:39
  SPEAKER_00: And as an entrepreneur, it's what it's what keeps me at Amazon and and sort of uh charging into work every day is how much we want to lean in on areas where we feel like we can do a ton for customers.
• 42:50
  SPEAKER_00: And just watching us build Alexa Plus with the urgency we've had around it and the amount of support that we get for doing that and continuously reinventing Alexa Plus even as we're going, uh it's so high.
• 43:01
  SPEAKER_00: So I the the the speed in the space, I think, Brian, you're right.
• 43:05
  SPEAKER_00: Like it feels like it feels like we're going fast, and then you you look back and say, oh my God, we're going even faster now.
• 43:11
  SPEAKER_00: And that's been happening now for some time, and uh it's inspiring.
• 43:16
  SPEAKER_01: Yeah, and obviously Amazon's got you know its in-house models.
• 43:19
  SPEAKER_01: You've got uh Nova, I think Nova 2 launched late last year.
• 43:23
  SPEAKER_01: You've you've had partnerships with companies like Anthropic, but I mean how how do you keep on top of everything that's been going on and how do you make sure that you are uh staying on top of those breakthroughs and incorporating them into a system like this?
• 43:40
  SPEAKER_00: We made a really conscious decision early on in building Alexa Plus that customers don't care what model you're using.
• 43:48
  SPEAKER_00: Like the the real secret about models is customers don't care.
• 43:53
  SPEAKER_00: They want to get something done.
• 43:54
  SPEAKER_00: Like the idea that you'll go to a drop-down menu and choose a number of something.
• 44:00
  SPEAKER_00: I mean, it it oh on the in maybe in the short term and for specialists, sure.
• 44:06
  SPEAKER_00: Absolutely.
• 44:06
  SPEAKER_00: Like you want, you know, the right the exact grip on your screwdriver and the this framing hammer.
• 44:12
  SPEAKER_00: Like, of course, for an artisan, someone who's really working through a big content problem, you know, producing uh, you know, research with a given model, say, of course it matters.
• 44:23
  SPEAKER_00: For most things that AI is gonna help customers do, the idea that they'll pick a number of a model or that they they think about it.
• 44:30
  SPEAKER_01: Especially Alexa, where it's like it's like what you know, the the idea behind a voice assistant is that this is the least friction possible, right?
• 44:38
  SPEAKER_00: You're just trying to get a job done.
• 44:40
  SPEAKER_00: You're just trying to help make customers' lives easier.
• 44:42
  SPEAKER_00: So an early design decision we made was first of all, we'll take whatever tools get the job done, and we'll always be able to choose the best tools if we don't attach ourselves to any one set of models.
• 44:55
  SPEAKER_00: We're not gonna make customers choose.
• 44:57
  SPEAKER_00: Alexa Plus has over 70 different trained models in it.
• 45:02
  SPEAKER_00: And for customers, it's one Alexa experience.
• 45:05
  SPEAKER_00: That's what matters.
• 45:06
  SPEAKER_00: They're just trying to get something done.
• 45:07
  SPEAKER_00: It's our job to choose the best tool.
• 45:09
  SPEAKER_00: And if there's a better tool tomorrow, we'll choose the better tool tomorrow.
• 45:12
  SPEAKER_00: Now, Amazon models do take the majority of traffic within the Alexa infrastructure, but we've got great partnerships with many folks.
• 45:20
  SPEAKER_00: If you want to know what models we're using, basically go to the Bedrock, you know, AWS Bedrock page, and you can see, you know, this incredibly long set of partnerships we have because different models are great at different things.
• 45:31
  SPEAKER_00: So that's that is, I think, one of the most important ways we quote unquote keep up is that we're we're we're we've built a product that fundamentally allows us to use the best tool for the job.
• 45:43
  SPEAKER_01: Yeah.
• 45:43
  SPEAKER_01: I I guess instead of um you know reading a paper and saying, oh, AI can do that now, it's more these are the um these these are the you know continued pain points or or or friction or these are the features that people are asking for, what you know, what's the best thing available in order to address those?
• 46:04
  SPEAKER_00: Correct.
• 46:05
  SPEAKER_00: And you know, the bet the best tools available, and also I think making some decisions like including access to Alexa Plus and Prime uh for Prime members unlimited, I think was a pretty critical one too, because even the best tools available, sometimes you have to pay an incremental 20 bucks or more a month.
• 46:22
  SPEAKER_00: If you're already a Prime customer, you don't run out of tokens on Alexa Plus.
• 46:26
  SPEAKER_00: You have unlimited access to an amazing uh AI, foundational AI assistant in your browser, and on an app on your phone.
• 46:37
  SPEAKER_00: Follow up by talking about the research you were doing on your Echo device in the morning over coffee.
• 46:43
  SPEAKER_00: I think it's um uh best tool for the job, but also right tool for the job might mean that you're already paying for a great foundational AI assistant, too.
• 46:54
  SPEAKER_01: Yeah, so we're we're getting close to time, so we we can end on this.
• 46:57
  SPEAKER_01: But you know, I was I was looking through some of the recent announcements, and uh I noticed there was some big news about a large data center in in Spain.
• 47:05
  SPEAKER_01: Um and I know that obviously uh you know climate has been a huge thing for for for Amazon.
• 47:12
  SPEAKER_01: Um there's been a lot of conversations around uh climate impact when it comes to AI and data centers and things like that.
• 47:19
  SPEAKER_01: Um it you know, how how are you how are you looking at that when it comes to um something like you know these large language models?
• 47:27
  SPEAKER_00: I mean, I think um if you know hopefully you've also read about the climate pledge uh taken on as a company.
• 47:35
  SPEAKER_00: I mean, we're the we're the largest purchaser of green power on the planet, uh, and that is just a documented fact.
• 47:41
  SPEAKER_00: And I I I believe strongly and support strongly our commitment as a company to the climate pledge, and we are ahead of schedule.
• 47:50
  SPEAKER_00: Uh you know, people, people, uh many companies have signed up for that same pledge, and I think it's you know, we have we have one planet, uh we have to take care of it.
• 47:59
  SPEAKER_00: And I think Amazon's deeply committed to that idea, whether we're talking about building data centers for to power generative AI or we're talking about fulfillment centers to get you your prime delivery this afternoon with you know in an in an hour or less.
• 48:13
  SPEAKER_00: Um, we're we're as committed to every aspect of that.
• 48:16
  SPEAKER_00: So and I it's just it's a it's a very um leading commitment.
• 48:22
  SPEAKER_00: And uh, you know, I'm humbled to work at a company that takes it so seriously.
• 48:26
  SPEAKER_01: Well, Daniel, thank you so much for taking the time.
• 48:28
  SPEAKER_00: I really appreciate talking to you, Brian.
• 48:30
  SPEAKER_00: Thanks a lot.
• 48:31
  SPEAKER_01: All right, thanks so much to Daniel and Amazon.
• 48:33
  SPEAKER_01: It's always good to dust those uh gadget blogging skills.
• 48:36
  SPEAKER_01: Thanks to you for tuning in.
• 48:38
  SPEAKER_01: Please like, subscribe, tell a friend.
• 48:41
  SPEAKER_01: We also have a newsletter with original features that drops every Thursday morning.
• 48:45
  SPEAKER_01: You could find that and more over at automated.fm.
• 48:49
  SPEAKER_01: And with all of that, we will catch you this time next week for another episode of Automated.