It's always listening, kind of, if you have a device. Hello everyone and welcome to AI Literacy, your podcast about artificial intelligence. We're your hosts. I'm Anna-Regina Entes. And I'm Victoria Rugley. Thank you for tuning in to a new episode of AI Literacy. We're here today with Noelle Silver. So Noelle, you're a multi-award winning technologist. You've built some of the first features of Amazon's Alexa, which millions of people are using today. You're a Microsoft MVP.
valuable professional in AI and you're the founder of three organizations, the AI Leadership Institute, Love Lancers, and the organization Women in AI, where you empower women in tech. And now you've recently joined Electrify. And next to all of that, you're a mother of four children. That's really incredible. Thank you so much for being here with us and actively making time to talk to us today. We are really excited to have you here at AI Literacy. Yay!
I am so excited to be here. Thank you for reaching out. I'm just so excited. This is such an honor really because literacy, especially around technology, is so important. I think a lot of people have dreams of using technology. And don't realize how easy or accessible it is today, more so than any other time in history. So I'm very excited to be here and part of this. So to start off, we would like to ask you how your career path started and what you did before, Alexa.
skills and before you ended up being an advocate for female empowerment and how it came by. Absolutely. So I've actually, I guess I've always been in tech, but it was timing is everything. I think there's a lot of women right now. right now who are thinking about like, oh, what can I do? Maybe they're just getting out of college. Maybe they're in college. Maybe they're even just finishing high school and they're deciding whether college is even a good idea. And we're in this time right now.
where we have this huge need for a certain type of technologist, specifically in the area of data analytics, data science, data engineering. 20 years ago, 20 plus years ago, we were in this spirit of Y2K, right? We thought way back then. The world was going to end when we hit this new number. It obviously, it didn't end. But there was a very similar momentum in the industry that said,
Anyone and everyone who has an aptitude for technology, come, we'll teach you. We need as many bodies as possible. Very similar to today around data. And so I, at that time, went to Barnes & Noble. Oh, good days of bookstores. One day we'll be back there again. But I went to Barnes & Noble and I started picking up books and I literally learned to code by...
doing exercises in books. I'm now a firm believer in like learning by doing because I didn't go to, you know, my university education was not specifically around computer programming or anything that would solve Y2K issues, which was all. around software engineering. It was all, I was in avionics at Embry-Riddle. I was on aviation school and I ended up, of course, not doing anything in aviation, though I still love to fly.
But that's how I got into it. Super exciting. And I think one of the key lessons I learned from it was that if you can find something that you like to do, which I did like technology, and that love for technology started. I was like six and my dad introduced me to science fiction and Star Trek and Star Wars and all these movies that inspired me. And so in the back of my mind, I was always like, man, it'd be cool to like build something that you could talk to or it would be cool to.
live in a spaceship or be in space and so I always had this kind of technical love for technology and so when this opportunity came I was like I could probably code Like, how hard is that? And I ended up learning it. My career has kind of been a whirlwind path since then. So you went, basically you went from tech giant to tech giant then. Yes. Yes. Which was not intentional.
No, not at first. So what ends up happening, right? I went to IBM. I was there for a long time, over 12, like about 12 years. And what ended up happening is that during that time, a new way of building software emerged called service oriented. architecture or microservices as we now call it. But back then, we were taking monolithic applications that were built that way on purpose and trying to change them. At that time, it was this term called enterprise application integration.
And we got paid to try and like glue the spaghetti of code together. It was a mess. But I thought it was fun. It was almost like I grew up doing jigsaw puzzles and crosswords and Sudoku, right? So my brain enjoyed the complexity of like lots of systems trying to talk to each other. And this was part of the kind of, I guess. in the early 2000s.
we started thinking about how do we make those big blobs of code more reusable? How do we break them up and make them more accessible? And that's where web services were born. But after that, companies, it just became a natural evolution, right?
First, we needed to break up our code. Then we needed to think about how do we actually deploy our code? Like, where should it go? I went to Red Hat because that became exciting for them. I then went to VMware because they started thinking about virtualization. And I just kept... going where the next big like wave of new
thinking happened to be. And that's what actually brought me to Amazon. I was recruited from VMware to Amazon to lead their AWS solutions architecture and training organization. And that was really fun because We were teaching people, again, like this art of thinking about new things and thinking about things in a new way. So we were trying to teach the world the cloud.
And then Alexa was born. And Alexa really changed everything, obviously, for my whole career. And cloud is still there. And I love cloud infrastructure and being a solutions architect for that. But Alexa really, it gave me another opportunity to join something where no one knew the outcome, right? Today, I'm glad I did it, but it just as easily could have gone the other way where Alexa was a device someone used once.
no one really liked and ended up dying. But it turned out that I got to be part of this velocity of a product that went from, you know, a bunch of beta testers to hundreds of millions of users. And it was a really exciting ride. You have those two organizations and you're still working full time, Electrify, and you're still doing thousands of podcasts and TikTok and all of this. How do you manage all of this? then we as women here yeah you cannot succeed with
both family and career and it's impossible. So like, what's your response on that? Yes. So I have always been a big believer in boundaries. Number one, right. So identifying what you want and it is possible to want. Like you can have whatever you want. It's just identifying what that is and then being willing to accept trade-offs. So for example, I wanted to have four children and I wanted to have an amazing career. The trade-offs is that I can't be home with my kids.
all day long every moment i can't homeschool them like there are certain things i can't do But I still get to have my dream of having a family. But now I have to share that. Right. So I often say I can have my cake and eat it, but I can't eat the whole cake. Right. Like there's some balance that happens. So one thing.
I do is that I set an intention in the beginning of every day, and I write down three sentences. And they might be a sentence, it might be a paragraph, depends on how much time I have. But one is, what do I personally want to do today that would make me feel fulfilled? Like personally, and this usually ends up being, I want to make sure I have dinner with my family or I want to make sure my kids, you know, that I.
play at the playground or go on the swing set with the kids or I go to the pool or whatever it is. I set a very specific intention for that day. And then I set a very specific intention professionally for that day. And I say, OK, well, today I want to make sure that I.
meet this client agreement or finish this document. And then the last thing I do is I write, and this is more of like an aspirational statement, but I write down what is my financial goal, right? What am I doing this all for? And so this is...
is when I combine usually my personal and professional together to say, I want to, you know, I always say something about, I want to create generational wealth for my children. I want to teach them about the value of working hard as well as the value of family and that you can do both. these things every single day, I end up like seeing opportunities to make those intentions come true or be real. And so like if I see an opportunity with my children.
It might be just a moment where they're like, mom, can you color with me? And I'm like, this is the time I can spend. I'm going to take 30 minutes, color with you. I will feel fulfilled. And it's a flywheel, right? So the more fulfilled I feel with my family, even if it's just these little moments, the better I am at my work and the more I want to share of myself. I created another organization called Love Fluencers. And it's basically like being a social.
influencer, but around like empathy and kindness. And one of those things is like, I want to take all this positive energy that I'm, you know, building in my life, in my professional life and personal life. Then I just want to share it with the world.
And I think so many of us, if we had more opportunities to share the good news, you know, of our lives, of our stories, we have plenty of sad stories to tell. But it's great when we can do both, right? When we can share our challenges, but also share our overcoming. So that's kind of how I do it is I'm constantly energized by that.
Honestly, that's so motivational. I have to start doing this as well. And I've been talking about doing this all the time, but I don't know why. Yeah, it's like, I don't have time. I know, but that's why I do one sentence because it's so much. Literally, I'm like, okay, I have five minutes.
But I also like the idea of envisioning the financial part in the long run, because ultimately, we as women, we're still struggling with a big... gap in that sense yes like just looking at the releases of the um world economic forum it's like 26 percent of women in ai so that's that's so little so little yes that doesn't even talk about what they get paid compared to their male peers, yes. Why do you think we're still so far away from an equally represented number?
For some reason, and this has been true my entire career, we are not transparent about pay. And there's no reason really why we wouldn't be. Yeah, but it's an old mentality. And I even wonder, you know, sometimes I have a sticker on my laptop that's like...
Rosie the Riveter from World War II, you know, the girl and she's like, we could do it. It's interesting because I wonder, you know, in those moments where we came to, you know, save the day in the World War II, what perceptions and kind of impacted that.
moment have on the rest. We're willing to do this. We're willing to sacrifice to go in. What did the pay even look like in those scenarios, right? Back at the very beginning, they were doing the same work as the men would have done. Did they get paid the same debt? I doubt it. I wouldn't think so. But now we don't even talk about it, which is what's weird. We know it's a problem. But even me asking my female peers how much they make, there's like this weird sense of I don't know.
secrecy. So I feel like one of the things we can do is like transparency, right? Like just talk about it, say it out loud. What's the big deal? Either you make more or less, but either way, knowing is better, right? Ignorance is not bliss in this scenario. Especially when we know there's a known inequity. And so same with like, I feel like pay inequity, injustice in the workplace, like bad behavior. It only exists because we're afraid to say anything and call it out.
take, you know, make people, hold people accountable. So I feel like, yeah, if we turn the light on and start talking about it, but I've had this dream for many years and it is systemic challenges are much harder.
to uh to change but but it just we've seen this it could just start with one person yeah but it's also i think of western country thing i lived in myanmar where really everyone was just saying yeah i earned this and i earned this and it was super open otherwise the company would get caught.
So, but it's also a thing in Western countries. I think that us women, maybe it's a cliche, but we often say, oh no, I'm not good enough. Or I'm like, I'm not technical enough. I should rather not do tech because I don't know coding. And yeah, then it leads to fear or to turning away from building the career in AI. So have you also faced situations where you had to cope with... Others discriminating somehow. Yes. Well, I will tell you when Alexa was just starting and I was.
thinking about joining the team as an early member, like employee 10. It wasn't led by a VP. There weren't thousands of people, like it was a very small team. And I wanted to join that team. And my manager at the time was like, oh.
I'm not sure that's something you could do. I mean, you've never done anything like that before. It seems like it's too technical for you. And it's interesting to me now because it's those types of voices and some of them are in our own head, right? We have that own voice in our head that says, Oh, you've never done that before. Who are you to go do that thing? But luckily for me, I've learned to.
you know, acknowledge those voices, whether they're external or internal and say, thank you for, you know, your input. I'm going to go do this anyway. What's the worst that'll happen, right? Yes. Many times I've had both internally and externally, people will say to me, Something that's...
ends up being untrue. I don't remember where I got it. It was an anonymous quote, I think. But I started to say to myself, you know, work hard in silence and let success be your noise. You know, I always, when I get a new job or a promotion or I go to do a new really...
cool thing with a company. I always think to myself, this is what I want people to look at. I don't want them to be like, oh, she got her revenge. I want people just to be focused on like, here's the good I can bring to the world.
whatever false perception someone may have had of me. But I think this is a big struggle for women because it's not his fault that he thought that. Again, a lot of it's systemic. He didn't think I was even capable because I was a woman, because I was a minority, because I was. was younger than him. It doesn't just have to be about color and gender and race. It can often be about how old I am. Really, it's anything that's different.
then who's been successful in the past? We all, all three of us represent different. than who's been successful in the past. But yeah, it doesn't indicate your success. What it does do though, it's harder for us to even try because these voices exist, right? There are some people who are completely supported no matter what they do. They are allowed to go, try, fail, and they're comforted in that. While there's others of us...
that it's presumed we could never do that thing. It's presumed that we don't have that capability. So we're, without even starting, already going against this current of resistance. And that is what it is, I think, to be a woman in tech. why I think it's so important to have
you know, shows like this and events where we get together, you're going to have that current no matter what. So we need to have a bigger current of support of, you know, us coming together as women and, you know, being like, you're not alone. We got this. Honestly, I have those same thoughts. as well very often that I've never worked in tech I know coding now for my master's but still I'm not not like an engineer I'm not competitive and yes yes do you have some words of encouragement
young women or for women in general interested in joining AI? Yes. So one of the things that gave me confidence, and remember when I got recruited into Alexa, I had no AI experience. I did not. Like I've never done anything. I've never even really built an application, like a app in a store or anything like that. All things that I would be about to do.
And what I ended up doing in order to gain confidence as I just started building, I built as much code, deployed as many repositories as I could on GitHub. And they weren't all mine, right? That's the beauty of GitHub is I would... There's this really cool idea, it's called ultra learning or super learning. Anyway, there's certain people who have been able to acquire, you know, there's people in the world that have learned 12 languages or who, you know, learned how to do something really fast.
And one of the techniques for artists to learn artistic technique faster is to use this concept of tracing paper. I don't know if you've heard of it where you put tracing paper on top of a more elaborate. design and then you trace over it and it gives your brain this like mental memory and for those of us who are like futurists like me it's like putting on a vr headset
Right. And let's say you've never been in front of a group of a thousand people. You go in a VR headset and you stand in front of a thousand people. The benefit is the same. Your brain thinks that you've done this before. It doesn't recognize that you're chasing someone else's work or that you're.
VR headset is giving you this cognitive rehearsal. But what it does is it gives us this internal confidence that we've done this before. And so that's why that's the only way really I feel like I was successful at Alexa in the way that I was because I built over a applications in my first year. And I just built and deployed, built and deployed.
Any opportunity. I combined people's skills. I used open source. You know, it was just, it was like this constant learning and building activity. And that way, when somebody asked me, I actually had more build experience. than most of my peers, right? It wasn't on production projects, but I had written more code than almost all of them. And so I feel like learning by doing whatever you're interested in, just dive in and start building. We have now low code.
and no-code solutions, where you can really dive into the tech without having to know the lines of code or the programming language or the syntax. You could still build really technical solutions without all that. Yeah. And especially if you practice more, it brings so much benefit, not only to yourself that you're more confident with it, but also to the enterprises because it only starts on the developer side. But if all these technologies are built by...
then it will affect the user's side as well. Yes. If we look at all the AI technologies, what are some long-term consequences of us females being underrepresented? Yeah, exactly. I mean, we're starting to see that even now. When I first joined the Alexa team, we really were thinking about it as like a device.
in people's kitchen. So it was like meant for like the 1% of the 1%, right? Like the whole world wouldn't use Alexa. Like it's for very special people who are going to buy a voice enabled device for their kitchen. Not everybody's going to be in that world, but that changed dramatically. when the accessibility of that technology...
was starting to be realized, right? So I, as you mentioned, I have four kids. One of them has Down syndrome. The fact that he could now use his voice to navigate things, right? To ask questions, to get information, to find out the day, the week, the month. And then my dad lives with
me the exact same benefits for elder care, right? Being able to find out the days and the weeks and the months, set appointments, get reminders, pill distribution, like all these really cool things. You know, our team, I don't think really thought through. I recently had said like, We don't necessarily intend to be biased when we build something.
vision is like, oh, we're building this product for this person. And when we do that, unless we have a diverse team, unless we have diversity of thought in that team, and I don't just mean women, of course, that's hugely important, but also different geographic locations, different ages.
different personalities, introverts, extroverts, right? All those people help us broaden the intention of our software that we're building so that when it does release, it meets more people's needs. Because almost always, especially those who built it,
They are disappointed when they realize they missed a demographic. It's almost painful, right, for engineers who are like, I didn't know I would disenfranchise someone. Like, nobody wants to do that. Most of these engineers are like, they're like you and me. We're in a big cog. of a system like Alexa, right? There's 2000 of us now, but we all have an intention to do good and build good things. And that's why.
Our voices are so important. It's funny when you are in a group of people that all look the same, talk the same, agree. They don't even know the questions to ask to provoke, right, diversity in the outcome. There's no malintent, right? Like they're like, do you think we should add anything? I don't think we should add anything. It looks good to me. Everyone's in agreement. And so there's this happy path that they go on. And it's only after that they're like.
I never thought to ask someone who was hard of hearing. I never thought to ask someone who was 80 or five, right? And when we have these, that's why it's nice to have moms and dads and like all these different ideas come to the table when we're creating software and specifically artificial intelligence, because it helps us increase the intention of that software. AI is never.
just for the one thing we're building it for. It is always ever evolving past that original intent. So the bigger we make that intention from the beginning, the better we can serve the world. about alexa which was supposed to be a small device in the kitchen helping a little bit and you were actually part of building it from scratch basically yes um could you maybe give us an answer why alexa became female and or why is it female because this is such a cliche yeah
It is. It is. Well, and if you remember, so Alexa is actually not, was not the first to this party of building voice assistants. Siri was first. And I actually met and spoke, shared the stage with the original voice of Siri. So back then they took it. human and chopped up all the syllables that that person would say and crafted Ciri's voice. There also was Cortana. Cortana was also around before Alexa, both female voices. And the reason, so there's a couple.
couple things. One, I think it's okay that the voice was female because there's research, lots of research around the tonality of a female or feminine voice. However, there is... Huge debate around why we would choose a female name. We could have easily chosen a gender neutral name. Yeah, like Echo. Like Echo, exactly. So there was a lot of healthy debate. When the device launched, it launched with three.
names Echo, Amazon, and Alexa. And what the organization realized was that they wanted to personify or anthropomorphize this device to make it your best friend. And in the... time where no one had one and no one knew what it was. If you could think of it as like your friend in the kitchen, it was an easier, easier marketing sell. But again, no one on the team was someone who maybe had a daughter named Alexa or maybe had, right, someone who would be.
impacted by that choice. So we all looked around. I wasn't in this conversation. This was an executive decision, but I'm presuming they looked around and everyone's like, Seems okay. As a matter of fact, it was a bit noble. The biggest library in the world is the Library of Alexandria. And it was kind of based on that, right? Like, it was this soft pull towards like, this is a library of information.
Your friend, a lot of early documentation called it a her, even though it's an it. But yeah, we just never thought about the impact of these. And I think that goes to your earlier question as well. If we have more diversity at the table, someone will mention that, right? Someone will bring it up and be like, who else is impacted by this decision? Who of our customers? Because now there's an entire sea of Alexas out there, people named Alexa that are like.
My life is different now because you built technology, right? Like I can't, you know, go into the same room or people have to mute devices when I, like it's weird for them. And how do we create more empathy? for everyone who's impacted by the technology choices. Do you think the government should do more in that term or do you think it's just a company level decision? There's lots of companies that were doing what GDPR has now enforced.
before. And even in the US, unless you're global, you don't have to do what GDPR requires. However, there should be companies that have a mission that... lead with this sense of integrity around the work that they're doing in AI and around data. And there are companies that do this. There are even consortiums, like I'm part of an organization in Seattle called Mira, and they represent what are we doing to give more...
agency to users around their data rights within a system, right? Within like an AI solution. I'll tell you a quick story. I asked a whole sea of people in a conference, if you could trade off privacy in order to get a better experience? Let's say traffic. Would you be willing to trade off surveillance of you and your vehicle and possibly your face in order to get
better traffic systems in and out of your city and remove rush hour. And 90% of the audience raised their hand and said, absolutely take all my data. The problem is, is today we use that data to solve rush hour. We have no idea.
who or what that data will be used for tomorrow or 10 years from now or 100 years from now, right? That data stays forever. You know, data can tell the story we ask it to tell. And so there's a lot of... integrity and ethical decisions that have to be made and I think we're very as consumers we're very quick to say oh sure for this use case okay and what we don't realize is that in AI
It doesn't really work that way. It's not one-time usage. That's why literacy is so important. We have to be much more careful about when we give this information away, how we give it away, and the license around it. I do feel like companies will only do that if they're regulated.
and forced to do it because it's expensive. So yeah, I think to your point, and not just diversity, but creating products that serve more people, creating products that protect user rights around data. There's some companies that will do it because they want to. but more companies will do it when they have to. So I think government regulation is a huge part of that. Yeah, so staying with your topic of data privacy, we're going to ask the classic question.
Is Alexa always listening? Yes. It's a complicated question. It's always listening, kind of. If you have a device, it has a push button mute. And you can tell when you press it, there's actually this tactical response. And I remember working with some government organizations who were like, it can't be a software mute. I literally need to know that the hardware is muted, like the hardware is muted.
And so when you press that button, you'll feel an actual toggle switch turn on or off. And in those cases, of course, when that toggles off, then yes, it's not listening. However, when it's in on mode, it's definitely listening. way that it listens is a little bit different. For example, it's the same exact listening that this phone is doing, right? I often...
tell people that sure, Alexa's listening, but your phone's been listening to you way longer and, you know, collects way more data because it's not just listening to what you say. It's listening to. What you say, where you are, are you moving or not? All of our devices have the ability to collect data about us. Alexa does, of course, listen. It uses the concept of a wake word, and the wake word has to listen all the time.
and it's looking for a specific signature. However, many times that wake word analysis happens on device. So even though it's all listening, it's not like it's listening in the cloud. You know, it's not like that. That analysis is happening in someone's data center somewhere. It's happening right in your office or in your kitchen. But these nuances for technologists, we get it. But for the rest of the world, if it's listening, I don't want any part of it. So about 30%.
of the world today is too scared or worried about this data privacy to invest in a voice-enabled device, 30%. And that number is actually growing. But I have found, again, from an accessibility perspective, extends, you know, my dad and my son. ability, but now also my four-year-old and my six-year-old can do way more than they ever would have done at this age 10 years ago because they can just ask for it, right? They can ask for mathematical equations. They can ask for the definition of words.
In a way that, you know, us parents, we've heard these questions many times, but sometimes you're like, ah, not now, honey. Especially homework. Like if somebody has a question or homework, like just ask Alexa. So yes, it is always listening, but not necessarily in the way that I think people are concerned about. So Noelle, thank you so much for making the time today. We really appreciate you sharing your experience and your thoughts with us here at AI Literacy.
so nice to have you and yeah now to wrap it up can you maybe tell our listeners where they can find you and be inspired by you every day Yes. Well, like you all, I met these wonderful ladies through LinkedIn. So that's a great way to connect with me. I connect with everyone who connects with me. I try to respond. Obviously, I also have a full-time job.
time family, but I try to be as engaged as I can. I also, every, so connect with me on LinkedIn and every, I think Thursday at 3 p.m. Eastern time, I'm doing This Week in AI, which is a live broadcast of...
kind of the cool, fun things I'm working on. Last week, I did video analytics and talked about, you know, computer vision and all these fun things. And it's meant for everybody. So if you're just getting started, you can come join me live. It's always published, of course, to YouTube and things like that.
And then if you want to just see the work I'm doing, I always publish. My blog is on Noelle.ai. So my first name, .ai. And you can check me out there too. But thank you so much. This was so much fun. If you've enjoyed listening to this episode and want to learn more about AI, Make sure to subscribe to AI Literacy on Spotify, SoundCloud, or your podcast platform of choice. Thanks for being with us today. We can't wait to share other insights on AI with you and help you become an AI literate.