Artificial Intelligence

Listen to the latest news, expert opinions and analyses on the ever-expanding world of artificial intelligence, data science and machine learning, broadcast on leading talk radio shows and premium podcasts.

A highlight from AI Today Podcast: Interview with Albert King, Chief Data Officer of the Scottish Government

AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

08:34 min | 6 d ago

A highlight from AI Today Podcast: Interview with Albert King, Chief Data Officer of the Scottish Government

"The AI today podcast produced by cognac cuts through the hype and noise to identify what is really happening now in the world of artificial intelligence. Learn about emerging AI trends, technologies, and use cases from cognitive analysts and guest experts. Hello, and welcome to the AI today podcast. I'm your host, Kathleen mulch. And I'm your host Ronald mills are in. Thank you for joining us again on our AI today podcast. We are now in our 5th season. That is we've been recording for over four years. We celebrated our fourth anniversary earlier this month in September of 2021 if you're listening to us later on. And well over 200 plus episodes. And we've been focusing on the challenges and the issues of making AI work within enterprises and organizations of all types. And you've heard us interview organizations from large banks and financial service institutions, pharmaceutical companies, especially in the past year as they've been going through the pandemic and really reworking their organizations from top to bottom. And as well as, as well as government organizations in the United States federal state and local level internationally Protestant review, lord Tim Clement Jones, the UK House of lords and Australian folks in Hungarian folks and companies in countries from all around the world. It's very interesting that we're all kind of moving this shipped forward. It's actually kind of nice how international artificial intelligence truly is. And that there really isn't such a centricity of the technology as maybe there has been what social media and other things Silicon Valley centric or computing and that's where stuff. New AI is really properly very international. And it is on that sort of note that I am really thrilled to introduce our guest presenter a guest host sorry guest. Featured guest for our AI today podcast. And that is Albert King, who is the chief data officer of the Scottish government. We are sort of thrilled totally thrilled to have him here. So thank you so much Albert for joining us on AI today podcast. Well, I am delighted to be here on apparently in such exalted company as well, what a treat. So excited to be here. Yeah, we're excited to have you today and for this interview. So we'd like to start by having you introduce yourselves and tell us a little bit about your background and also about your current role at as CDO of the Scottish government. Our listeners know that we have produced country level strategy reports on how different countries their AI adoption and their AI strategy. And so we're really excited to have you here to share what Scotland is doing as well. So yeah, so please introduce yourself to our listeners. Yeah, thanks very much. So I'm essentially my job is to make real Scottish government's vision for data that you systematically to save time money and lives and that's all about contributing to making Scotland wealthier fairer and greener. So no small job then. And so is the center of excellence with data in government, my teams are committed to helping us realize that vision. And that's really through, I guess, three things through the platforms that we've provide to support the sort of end to end data journey for analysts across government and public bodies in Scotland. Secondly, really about providing specialist expertise and analytical skills, particularly in areas where those difficult to acquire or particular kind of deep expertise. And thirdly, really around providing the policies and governance that we've put in place to deliver all of that. So I would look at peace. So yeah, as I say, we're making that vision real through the services we provide in the policy and strategies we create and deliver. Ultimately, then to help us to achieve the outcome set out in our national performance framework. So full disclosure on the date professional to trade. So I would tinker with all this technology just for fun if people like me, but it turns out it's really about the ultimately the social economic and environmental value that's creates. And the national performance framework. So our purpose as a government to create a more successful country in Scotland and interestingly actually picking up on some of what you were saying earlier. I definitely think this is an agenda where we can connect apart from the other things I said about wealthier fair and greener to that international agenda because this is really and one of the things that is really clear is that there are challenges around this technology that really do have that international dimension to that opportunity to demonstrate our international contribution and that we're not we're looking at nation language very important as well. So linking all of that back then to AI that is a big feature of our AI strategy that NPF at national performance framework woven through it connects those outcomes and it really is the AI strategy really is about putting people and society at its heart and achieving those outcomes through the adoption and use of that particular sort of data driven technology. So yeah, it's fantastic. And that's really very interesting. I think our listeners might be really very interested in the NPF the national performance framework because people are looking for frameworks in general to help guide whether they're multinational organizations and agencies or their country level agencies or organizations or even businesses of all types and sizes. We're all trying to figure out how to really make data really work in a way that's beneficial to the organization and to our customers and to our citizens and to everyone in the society. And I think that's an interesting place now. People are paying much more attention to data. I think nowadays, did the average person that we might have in the past. And as meant as Kathleen mentioned, we have covered AI strategies from Afghanistan Zimbabwe and every country in between, you know, Botswana and Cape Verde and Colombia and Hungary and it's interesting. Really, most nations are thinking about their countries positioning with regards to AI at the production or the consumption level. So in March of 2021, earlier this year, we know that Scotland published its own AI strategy and we talked a little bit about that just now. But maybe you could tell our listeners a little bit about sort of what led up to the creation. And a little bit about the strategy as well as to kind of how Scotland feels it's participating in playing within this global ecosystem of data forward organizations. Yeah, thanks. Yeah, so the background AI strategy, I guess. And the genesis of it, if you like, it was a recognition of first of all some strengths, maybe that Scotland has an AI, so we've got excellent research institutions here in Scotland. University in Scotland do some of the often recognized in top rankings for research. There was also a sort of piece around the innovative companies that we've got in Scotland. So really transforming, I guess the potential that that research creates into the economic value and also actually innovative public services that are looking to adopt this technology. And as you know, as you were saying, turn the technology into real, in this case, social value. But there was also a recognition of the risks and indeed fears that sometimes are associated with this technologies. And so that came together really, I guess, as the impetus that drove us to act and ministers are cabinet secretary for finance access to take forward the development of an AI strategy. I mean, really asking us to put citizens at the center of that to maximize the value for AI and really I suppose with the driving thinking that we would only see that value realized if it's adopted widely and underpinned by confidence and trust in the technology. So

Scotland Scottish Government Kathleen Mulch Ronald Mills Lord Tim Clement Jones House Of Lords And Australian Albert King Albert United States UK NPF Cape Verde Botswana Kathleen Zimbabwe Colombia Hungary Afghanistan Cabinet
A highlight from Forecasting in Supply Chain

Data Skeptic

06:47 min | Last week

A highlight from Forecasting in Supply Chain

"It's a fascinating and important area where time series is applied, and in this episode, I interview Mahdi abuse me about how these systems are deployed in practice and how to deliver the forecast that power them. My name is Madi Abel kasami and currently I'm a lecturer at monash university. And I recently finished my postdoc in the department of datasets and I want a university. Well, congratulations on that. And tell me a little bit about your PhD thesis in the area you focus on for research. Yeah, my PhD research was on forecasting in supply chain, specifically, I was focused on forecasting demand during promotions and in the presence of volatility in supply chain. So promotional forecasting is investigated widely in supply chain forecasting by researchers and practitioners. But it's a still open research problem. There's no Unix solution for forecasting during promotional periods. So during my PhD was looking at this promotion and specifically it was interested to see this promotion. It makes the demand volatile not just during promotion time, but of course, few periods before and after promotion, and that volatility actually caused a lot of problems in the supply chain context for your inventory planning. And of course, for your demand for casting, first of all, so my PhD focus was on understanding the demand volatility in supply chain and finding a way out to model this volatility in the supply chain context. My current research is yes, I'm continuing to work on supply chain forecasting. A little bit more focused on hierarchical forecasting, a specifically hierarchical forecast and you know it naturally sits down in the supply chain context. So recently, the last few years, I've been focused on developing algorithms specifically with machine learning algorithms to forecast hierarchical supply chains. What happens if a large supply chain neglects the efforts of forecasting? What's the worst case scenario? How does it all go wrong? Yeah, the forecasting is the basics of a lot of managerial decision in supply chain, whether it is transportation planning, production planning, inventory planning, everything it starts with the forecasting demand. So if you need click that well, you will encounter a lot of problems. You're going to spend a lot of money probably for your inventory control or if you have access inventory or you may run out of products and a lot of customer satisfaction will be damaged over there. So that's just one aspect of it. But of course, it can cause a lot of problems in your transportation planning in your production planning. You may not be able to commit to your plans. And so it is fundamental. It's the first step in every supply chain you need to really forecast your demand and know about it in advance. So promotions, at least if they're effective, they should have an impact on sales. We should see them in the data. What do they look like in the data? Yes, we can understand the promotion in the data, not always, but often it depends in the context as well. If you're looking at datasets and you can see large sparks very, very largest points. So you may say, okay, this might be a promotion, or that might be an output, or you don't know. But often we have when we are looking at the data, if we know the promotion in advance that we do have promotion in these days, if there is an spark that corresponds with a promotion date, so definitely that can tell us that that's a promotion. But it's not often, it turns out that it's not often that easy to understand that promotion is there. If you don't have the promotion date in your data, sometimes the uplifting sales is not as big. So you may think that okay, it's just a natural volatility in your time series and yourselves. But it may be promotion. So if you have data about it, you know, it okay, this is promotion time. You can easily detect them, but if not, you really need to come up with your own solution to find that which one is promotion and which one isn't. When a promotion period starts, it seems as though you've almost had a state phase transition. You're in a new mode previously. There was no promotion. Now there is one, presumably it will end. How do you look at it from a modeling perspective? When we're looking at a modeling promotion, there are quite a lot of different ways to really model them. One very simple way that I've seen practitioners use and other researchers also have used is I'm looking at simply price. We often know the promotion pricing advance. We are the decision makers. We know when we want to put our products on promotion. If something is going to be on half a price, so yes, we know that is going to impact self. So you can simply include that in a recreation model to reflect the promotion. The other way that we can look at this is if you don't want to use the price as a requisite, you can use simply a dummy variable, for example, to reflect the promotion time. So that can help us to model promotion. There are many other ways that we can look at it, but just a couple of ways that we can incorporate the promotion impact and their widely used are probably regressors and also dummy variable. So if we think of a metric like maybe it's daily sales or hourly sales with a large retail company, something like that. And if we looked at the data during a period of no promotion, so sort of a control group in a way, I still envision that there's a lot of volatility there, a volatility has many sources besides promotion. What does a typical dataset look like in a control period? Yes. We always have volatility up and down in time. It's not a constant process. Yes, we can have a trend of our train or downward trends in our cells. We can have seasonality. We can have multiple seasonality that can make it even more challenging. And there are random variations in it as well. So they all contribute to demand and we need to really look into them to be able to model them properly. There are techniques and ways that we can deal with each of them. But these are the main components of typical demand times in doing non promotional time. And how do you go about measuring the volatility of the data? Yes, volatility, there are different ways to measure that. One thing that I propose to use simply using coefficient of variations. So what does that mean core vision of variation is simply standard deviation divided by mean of your time series? So that's scaling dependent metric. If you have a large volume of sales or a small volume of sales, that doesn't impact it because we are divining it by the mean. So that is a very simple way to look at a volatility of time series. But there are other techniques that other researchers have also used. But I propose to use this technique because it is really intuitive and easy to understand. And it has been shown that it can represent the volatility in time series itself time series context adequately.

Madi Abel Kasami Department Of Datasets Monash University Mahdi
A highlight from #244  Robert Crews: Afghanistan, Taliban, Bin Laden, and War in the Middle East

Lex Fridman Podcast

03:26 min | Last week

A highlight from #244 Robert Crews: Afghanistan, Taliban, Bin Laden, and War in the Middle East

"Mistake for the United States to invade Afghanistan in 2001, 20 years ago? Yes. As simple as yes, why was it a mistake? I'm an historian so I say this was some humility about what we can know. I think I'd still like to know much more about what was going on The White House, you know, in the hours days, weeks, after 9 11, but I think the George W. Bush administration acted in the state of panic. And I think they wanted to show kind of toughness. They wanted to show some kind of resolve, this was a horrific act, they played out on everyone's television screens. And I think it was really fundamentally a crisis of legitimacy within The White House, the novels. And I think they felt like they had to do something and something dramatic. I think they didn't really think through who they were fighting, who they named me was what this geography had to do with 9 11. I think lean back at it. Some of us not to say I was clairvoyant or you could see in the future, I think many of us were from that morning, skeptical about the connections that people were drawing between Afghanistan as a state, as a place. And the options of Al-Qaeda in Washington and New York and Pennsylvania. So as you watch the events of 9 11, the things that our leaders were saying in the minutes hours days, weeks that followed. Maybe you can give a little bit of a timeline in of what was being said. One was the actual invasion of Afghanistan and also what were your feelings in the minutes, weeks after 9 11? I was in D.C.. I was on the way to American university hearing on APR. What happened, and I thought of the American university logo, which is red white and blue. It's an eagle. And I thought, you know, watching his under attack and symbols of American power are under attack. And so, you know, I was quite concerned and at the time lived just a few miles from the capitol. And so I felt that it was real. So I appreciate the sense of anxiety and fear and panic and for two, three years later in D.C., we are constantly getting reports, mostly rumors and gun confirmed about all kinds of attacks with all the cities, but I definitely appreciate the sense of being undersold. But in watching television, including Russian television, 'cause I just installed a satellite thing. So I was trying to watch world news and get different points of view. And that was quite useful to have an alternative set of eyes in Russian. Yeah, in Russian. Yeah. Okay, so your Russians is good enough to understand Russian television. The news, yeah, the news and the visuals that were coming that were not shown on American television. I don't know how they had it, but they had, they were not filtering anything in the way that the major networks and cable television were doing here. So it was a very unvarnished view of the violence of the moment, New York City of people diving from the towers are being held back on that, which was quite fascinating. I think much of the world saw much more than actually the American public saw. But to your question, you know, amid that feeling of imminent doom, I watch commentators start to talk about Al-Qaeda and then talk about Afghanistan and when the experts was born at Rubin, who's NYU who's a long, very learned Afghanistan hand.

Afghanistan American University George W. Bush D.C. White House Qaeda United States AL Pennsylvania Washington New York New York City Rubin NYU
These are the Best Buy Black Friday 2021 deals I'm eyeballing right now

Data Skeptic

00:44 sec | Last week

These are the Best Buy Black Friday 2021 deals I'm eyeballing right now

"Enough. Is your question more like, why is Black Friday important to me or what's the opportunity on Black Friday? Yeah, like I've always felt Black Friday was kind of a scam. I've never seen a deal I really cared about and I stopped paying attention, but you always seem like maybe you'll get something. Oh, there's tons of deals. When we buy furniture Kyle, we bought our coffee table for like 50% off. We bought a few other things that were like 20% off because during Black Friday, they started discounting things because they want, well, first of all, they have those lost leaders. So, you know, they have a few things that they set on sale, then that brings them in and then people buy additional things on top of it, you know, for free shipping. That's always a scam. Like I have it on very good authority

Kyle
A highlight from AI Today Podcast: Data Science through a Human Lens  Interview with Felipe Flores, host of Data Futurology Podcast

AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

06:59 min | Last week

A highlight from AI Today Podcast: Data Science through a Human Lens Interview with Felipe Flores, host of Data Futurology Podcast

"The AI today podcast produced by cognac cuts through the hype and noise to identify what is really happening now in the world of artificial intelligence. Learn about emerging AI trends, technologies, and use cases from cognitive analysts and guest experts. Hello, and welcome to the AI today podcast. I'm your host, Kathleen Walsh. And I'm your host why don't you smell those? And we have another fantastic podcast for you today. We are talking, actually, we have been talking about honestly for the last couple of hours. Felipe Flores, who is the host of the data futurology podcast, which is a fantastic podcast also really tackling a lot of the same issues that we talked about here on the AI today podcast, which is really about making AI an implementation reality. You know, we think about all of the promise and the hype and the theory of what we can do with AI, but of course, for those of you in our listening audience who are actually trying to put AI together whether you're in the private sector in large companies or small companies or you're consulting firm or you're working in the public sector, you know, the government's international federal state local or a government contractor, you know you're running into some real challenges of making AI work. And so a lot of what we do here on the Internet today podcast is first of all, I'll talk to the others who have been putting AI into practice and some of their issues and their challenges. But also talk about some of the challenges in the solutions. And on that note, we are just so thrilled to have with us as I mentioned Felipe forest is the host of the data futurology podcast a fantastic podcast that I listen to and we will definitely be linking to in the show notes. So Philly based, thank you so much for joining us on AI today. Guys are you kidding me? Thank you. This is an amazing. And just before we started recording, I was saying how I feel that we're so well aligned in terms of that we want to bring content to the community that helps people get value out of AI, make more dreams and reality and to highlight different use cases, different challenges, bring different perspectives of how people have been overcoming these challenges and to help the breach the gap between the AI hype and the AI reality in a way that empowers people and helps them move AI forward in their organization and obviously on the value side as well. So the focus is definitely very well aligned and it's amazing to spend some time with you guys. Yeah, so we're also on Felipe's podcast, the data futurology podcast, and so we'll make sure to link to that in the show notes. So that our listeners can listen to our interview with you. It was a really incredible conversation and you're right. We really are aligned. So I encourage our listeners to check that out. But I'd like to start by having you introduce yourself to our listeners and tell them a little bit about your background, your podcast, and why you started your podcast. Yeah, perfect. So I'll give you the quick version. So I'm originally from South America. I grew up in Chile in the north part of Chile in the dry desert in the world. So over there, it rains once every 7 years, and we get less than an inch of rain. And so I grew up in a small mining town 10,000 people and it came to Australia when I was about 20. And I started work and I did kind of like od jobs then got into data and loved it. And did about 7 years of working consulting, small and large businesses, and then I started my own analytics consulting company, data for about 5 years, grow it to about 50 people, and then we were doing data driven products online. We had 200 or three different products, and we did also consulting, which those consulting profits helped us hire people to do the development of the products. I sold my part. This is over ten years ago now. Yeah. No, maybe a bit less. But yeah, I'm getting old. And then I spent some time in finance, and now I'm head of head of that assigns in one of the biggest banks in Australia. And that was a great opportunity to build a team and develop the culture and think of really interesting applications on how to use finance data, particularly in the B2B space where people had people that I care about and I care about me. They had kind of like warned me not to go into the B2B space because there's very little amount of transactions that happen every year and there's many, many less customers, like orders of magnitude less customers than what you have on the retail side. But what we ended up finding is that you can leverage the retail data to do really interesting B2B work and that that can feed back into the retail side. So that was a really nice marriage. And then and then after finance, I moved into healthcare. So now I'm head of data and technology for a healthcare AI startup where we studied at the start of last year. So in Jan 2020, we got some singed funding for two companies went into a one in Australia. And then one in the U.S., the U.S. is like a huge global brand. It's called Cigna. So they're now a health insurer. And yeah, both companies come together, put some money at some speed investment into what's now. And we started this business. We've got essentially three, three pillars, where one, we have health management programs. So we help people recover after they have surgery or we help them with mental health issues. And we do that with programs that are telephonic and digital. So that's been really interesting. That's first pillar, second pillar is the contracting side and this is part of what I love to see and what I've done. I've tried to do a lot with my career, is to see kind of like how the hidden part of the world works. And that's why I was in finance. I was attracted to the B2B side because you get to see entire supply chains of businesses that serve each other and businesses that you never heard about. Like, who makes who makes the McDonald's hamburger McDonald's hamburger cases like the cardboard cases? We worked with those people, so you get to see the hidden side of the world in healthcare, the contracting piece is how payers or insurers agree how much they're going to pay the providers. So the hospitals, especially, the family doctors, and there's been historically a lot of information asymmetry in that space with large hospitals, negotiating with small insurers or vice versa, large insurers, negotiating with small hospitals, and that information asymmetry has led to really combative relationships and what we're doing is opening up the data, having a great metrics and everyone can see the same. So that's our second

Kathleen Walsh Felipe Flores Chile The Dry Desert Australia Felipe South America Government U.S. Cigna Mcdonald
A highlight from AI Today Podcast: AI Failure Series  Iteration Time & Proof of Concept vs. Pilots

AI Today Podcast: Artificial Intelligence Insights, Experts, and Opinion

05:05 min | Last week

A highlight from AI Today Podcast: AI Failure Series Iteration Time & Proof of Concept vs. Pilots

"You know, you may have heard from others that many AI products are not succeeding. You know, the failure rate and AI projects and many data analysis big data products as a high failure rate. And as we've talked about in our other episodes in this series, this is episode to think 5 or so in this particular series, there are actually many reasons for these fears. It's not because, oh, well, I guess AI is terrible, or I guess the products are really crappy. It's like, no, actually, good products. Hey, some of the best researchers in the world are working for some of these technology companies. So you can not accuse these technology companies of having all smoke and mirrors. Now, they really do. You can build some fantastic machine learning models with all these tools. So you can't blame the tools. You can't blame the technology. You can't blame the researchers. You can't blame the concept. So who can you blame when the AI projects fail and oftentimes it's yourselves? And the reason for this failure has a lot of, well, this is now the 5th episode, so we got like, you know, fire actually 5 out of ten reasons why they failed and past few we've talked about data quality issues and data quantity issues using application development methodologies for data projects. We've talked about issues around just trying to understand the data and all that sort of stuff. But in this, we're going to talk about a different reason for this time this time around. The only reason why I bring all that is a highlight is that if this is the first episode you're listening to on this topic, we encourage you to go back and listen to some of our others and this topic. We've well over four seasons and now into our 5th season of AI today and we've talked to a lot of people about their AI successes. So if you want to hear about AI successes, you're not going to hear that in this podcast, but you will hear that in other episodes in our series. So I really do encourage you to go back and listen to them and subscribe so you can continue to hear both the successes and failures of AI in our AI today podcast. Exactly, and we thought that it was important to have a series on why we are some common reasons why we're seeing AI projects fail. Because it doesn't have to be this way. So if you can see why other projects are failing, then you can learn from that and don't make the same mistakes so that you can set your projects up to succeed. So we decided to have an AI podcast series dedicated to AI project failures. And in today's episode, we're going to talk about why we're seeing projects fail one of the reasons being that the time between pilot to full production is just way too long. You know, we always say, what happened to iteration? Why are you taking, you know, why is there 18, 24 months in between the time that you have a pilot to iteration? Also, we've seen a lot of organizations talk about proof of concepts. And put out a proof of concept of POC. And we just say, why are you doing that? Because you want to make sure that you're setting your project up for success. And if you are not using real world data in a real world environment, then you are going to really suffer when this actually is out in the real world. So let's start with that. Let's start with the proof of concept. So if you're not familiar with some of this terminology that sort of like the lingo of project management, you may not have heard the terms of you may kind of generally know what a proof of concept is and what a prototype is and what a pilot is. But you may not be aware that they actually have very specific meanings, right? So the idea of a pilot is that well, let's actually start with the very basic idea of a proof of concept, because I'm trying to literally prove a concept like, oh, I think I can apply natural language processing to read some sort of issue in my system. Well, I could build a really small project, and I could try it. It sort of like a test, you know, or a trial. The thing, though, is that doesn't really actually prove anything. It only proves that the technology works, and okay, and that very specific, limited circumstance you can get something to work. But of course, the whole thing can fall apart when you actually try to use that same idea in the real world data. It might be like, oh, I guess the data doesn't really look like what I tested like. Maybe it works on my laptop, but it doesn't really work on the server. Maybe the environment in which I'm deploying it doesn't match the environment. There's so many reasons why that doesn't work. You might be like, okay, well, then I guess any logical organization wouldn't start with that. They do like what something is more specific like test the real world in a limited environment. And that would be the pilot. And the idea of a pilot is you're going to take something that's a real world you can use real world data. You can use a real world problem. And you can have a sort of a good environment in which you can test. We're still testing that whether or not it's a safe environment, there's a good way to think about it. So we can test that idea in a way that, oh yeah, if this works, then I can scale this thing up.

A highlight from #243  Kevin Systrom: Instagram

Lex Fridman Podcast

05:54 min | Last week

A highlight from #243 Kevin Systrom: Instagram

"Doom two. Worked at a vinyl record store, then you went to Stanford, turned down mister Mark Zuckerberg and Facebook went to Florence to study photography, those are just some random, beautiful, impossibly brief glimpses into a life. So let me ask again. Can you take me through the origin story of Instagram? Given that context set it up. All right, so we have a fair amount of time. So I'll go into some detail, but basically what I'll say is Instagram started out of a company actually called bourbon. And it was spelled BU BN. And a couple of things were happening at the time. So if we zoom back to 2010, not a lot of people remember what was happening in the dot com world then but checking apps were all the rage. So let's check it out. Goala four square hot potato so I'm at a place I'm gonna tell the world that I'm at this place. That's right. What's the idea behind this kind of app by the way? You know what? I'm going to answer that, but through what Instagram became and why I believe Instagram replaced them. So the whole idea was to share with the world what you were doing, specifically with your Friends, right? But there were all the rage and Foursquare was getting all the press and I remember sitting around saying, hey, I want to build something, but I don't know what I want to build. What if I built a better version of Foursquare? And I asked myself why don't I like Foursquare or how could it be improved? And basically, I sat down and I said, I think that if you have a few extra features, it might be enough. One of which happened to be posting a photo of where you were. There were some others. It turns out that wasn't enough. My cofounder joined, we were going to attack Foursquare and the likes and try to build something interesting. And no one used it, no one cared, because it wasn't enough. It wasn't different enough, right? So one day we were sitting down and we asked ourselves, okay, let's come to Jesus moment. Are we going to do this startup? And if we're going to, we can't do what we're currently doing. We have to switch it up. So what do people love the most? So we sat down and we wrote out three things that we thought people uniquely loved about our product that weren't in other products. Photos happened to be the top one. So sharing a photo of what you were doing where you were at the moment was not something products let you do, really. Facebook was like post an album of your vacation from two weeks ago. Twitter allowed you to post a photo, but their feed was primarily texted and they didn't show the photo in line. Or at least I don't think they did at the time. So even though it seems totally stupid and obvious to us now, at the moment, then posting a photo of what you were doing at the moment was like not a thing. So we decided to go after that because we'd notice that people who used our service, the one thing they happened to like the most was posting a photo. So that was the beginning of Instagram and yes, like we went through when we added filters and there's a bunch of stories around that. But the origin of this was that we were trying to be a check and app realized that no one wanted another checking app. It became a photo sharing app, but one that was much more about what you're doing and where you are. And that's why when I say I think we replaced checking apps, it became a check in via a photo rather than saying your location and then optionally adding a photo. When you were thinking about what people like from where did you get a sense that this is what people like. You said you sat down. We wrote some stuff down on paper, where is that intuition? That seems fundamental to the success of an app like Instagram. Where does that idea? Where does that list of three things come from? Exactly. Only after having studied machine learning now for a couple of years. I have a you have understood yourself. I've started to make connections like we can go into this later, but obviously the connections between machine learning and the human brain, I think are stretched sometimes, right? At the same time, being able to back prop and being able to look at the world try something, figure out how you're wrong, how wrong you are. And then nudge your company and the right direction based on how wrong you are. It's like a fascinating concept, right? And I don't, we didn't know we were doing it at the time, but that's basically what we were doing, right? We put it out to call it a hundred people and you would look at their data, you would say, what are they sharing? What resonates? What doesn't resonate? We think they're going to resonate with X but it turns out they resonate with why. Okay, shift the company towards why. And it turns out if you do that enough quickly enough, you can get to a solution that has product market fit. Most companies fail because they sit there and they don't either they're learning rates too slow, they sit there and they just they're adamant that they're right, even though the data is telling them they're not right. Or they're learning rates too high, and they wildly chase different ideas. And they never actually said on one where they don't groove, right? And I think when we sat down and we wrote out those three ideas, what we were saying is, what are the three possible whether they're local or global maxima in our world, right? That users are telling us they like because they're using the product that way. It was clear people like the photos because that was the thing they were doing. And we just said, okay, what if we just cut out most of the other stuff and focus on that thing? And then it happened to be a multi-billion dollar business and it's that easy, by the way. Yeah. I guess so.

Instagram Mark Zuckerberg Facebook Stanford Florence Twitter
Interview With Daniel Kornev Chief Product Officer at DeepPavlov

The Voicebot Podcast

02:07 min | 2 months ago

Interview With Daniel Kornev Chief Product Officer at DeepPavlov

"Daniel gornja. Welcome to the voice. Podcast much brackets and big for me to turn today today. It's my pleasure to have you. This is a long time in the making. We've been i guess chatting on slack for maybe year and a half something. Yeah i think so. I started to read your westport. Insider was fascinated by opportunity to look into your think to on hand Why not took. Yeah that that's that's how it happened. Well the is really perfect. Because we're going to talk about a few things today. Obviously d. Pavlov is a project i've been interested in for at least a year. I don't remember when i first came across it but it might have been might have been. You introduced it to me. Or maybe shortly before that i found out about it but definitely answered that project and then obviously you've been involved recently with the elec surprise social competition. We've had another conversation about that about this. What a perfect time to go a little deeper on that because it is a different way to build bots and so really looking forward to this conversation today. But i'll let you get started. So why don't you tee it up for the The audience right now first and let them know what d- pavlov is before we get deep sure depot is like lab at moscow's physics and technology. That is focused on conversational And neural efforts Officially cool to neural networks in Terrain but Wednesday were standard like full. Five years. ago it's also got to down moniker Because follow fossil famous russian scientists who discover it reflects us in all those things that encouraged scientists researchers to understand how human brace books and we still have a lot of things that we have to uncover. But that's was formed as the name.

Daniel Gornja Pavlov Elec Moscow
Google Develop AI for Detecting Abnormal Chest X-Rays Using Deep Learning

Daily Tech Headlines

02:09 min | 2 months ago

Google Develop AI for Detecting Abnormal Chest X-Rays Using Deep Learning

"On friday we talked about a nature publication by google. Ai scientists that showed how a deep learning system could detect abnormal chest xrays rays with an accuracy. Rivaling that of professional radiologists. The system only detects whether a chess scan is normal or not and is not trained to detect specific conditions. The goal here is to increase productivity and efficiency of radiologists clinical process. Let's examine some a i x ray. Science first of all how to rays work xrays are a type of radiation energy. Wave that can go through. Relatively thick objects without being absorbed or scattered very much. X rays have shorter wavelengths than visible light which makes them invisible to the human eye for medical applications of vacuum x. Ray tube accelerates electrons to collide with a metal and owed and creates rays these rays are then directed towards the intended target like a broken arm for example and then picked up by digital detectors called image plates on the other side differ body tissues absorb x rays differently so the high amount of calcium in bones for example makes them especially efficient at x ray. Absorption and this highly visible on the image detector soft tissues like lungs are slightly lighter but also visible making x ray and efficient method to diagnose pneumonia or pleural a fusion Which is fluid in the lungs. For example according to this latest nature publication approximately eight hundred and thirty seven million chest. Xrays are obtained yearly worldwide. That is a lot of pictures for radiologists to look at and can lead to longer wait times and diagnosis delays. And of course. This is why there's interest in developing ai. Tools to streamline the process many algorithms have already been developed but are rather aimed at detecting specific problems on an x ray. The google ai. Scientists however developed a deep learning system capable of sorting chest xrays into either normal or abnormal data intending. To lighten the case load on radiologists

Chess Google Pneumonia
Leveraging Mastercards DNA Onto Blockchain

Insureblocks

01:59 min | 2 months ago

Leveraging Mastercards DNA Onto Blockchain

"Leandro well do blogs. Could you our listeners. A quick introduction on yourself sure. Hi how are you in. Thanks for having me here. Who finally finding a good time to to get this done. Apologize up front i. It was completed my fault. I earned busy. Yeah so yes oh land newness here. I'm vice president of developing innovation and mastercard Much more everything related to my career. Talk a little bit about that. But it's own innovation in a mastercard with Tackle some use case for blockchain. I'm a i'm a little portfolio for this solution is that are on top of blockchain. Call the master providence solution. And i'm glad to be here in. Let's let's check more about other excellent excellent so as it is customary here at intra blocks. Could you please explain to our listeners. What is blockchain. And how does it work. Do you want to have a go at it. Oh goodness so. Yeah so jimmy. Blockchain is a distributed alleged technology that uses a in a sort of a consensus methodology Record blocks in sequence in alleger. It's technology that is driven by data governance and In by in a establishing that works to tackle the use case so it basically helps integrated participants in also the legacy systems of this participants in in Decentralized environment that they can share the data when they do that that provides a gives the visibility that you know would majorly increase the trust between the participants.

Mastercard Leandro Blockchain Jimmy
How to Achieve Extraordinary Results When Building Your Brand

Digital Conversations with Billy Bateman

02:13 min | 2 months ago

How to Achieve Extraordinary Results When Building Your Brand

"Welcome to digital conversations. I am your host billy bateman and today i'm joined by a man that needs no introduction. Cmo gong. Would you let booting you doing today. I'm great to be here. Yeah yeah man really excited to have you and In have this conversation so before we get into it you know for those. That don't know you mind. Just introducing yourself and tell them a little bit about gong. They don't know what gong is sure. So i'm rudy as said on the chief officer chief marketing officer at this is my fifth time leading marketing team for a tech company figures companies of our public and Contact this is my third time working with my currency yo ben-dov so we work together across three. Different companies was all by south. Salesforce was Influences and gone is probably not going to be bought by anyone. Because we're waiting for that Gone is revenue intelligence platform that unlocks reality to help people in companies achieve their full Whether it sales teams marketing teams product teams has were success teams week. We help. everyone has anything to do with revenue. Yeah you guys do a great job. The first time i heard about gone my brother's a sales rep at sas company. And he is asking me what what tools are you using. He's like well they just got this. This gong thing like a month ago and i. I was a little skeptical. But he's a really good. It's helped my calls quite a bit. So at a typical response we get some companies reps initially a little Don't know my boss uses against new york. Listen everywhere. I say but it's really meant for them. It's what have to enter notes that their calls so they listen other reps calls and learn from them so they can go back and see what the customer or shared their calls with the customer asks for feedback and get help companies who are really allow helping the rep succeed and turn some of the struggling ones into fantastic once gonna get a ton of value out of

Billy Bateman Cmo Gong Sas Company Rudy Salesforce New York
Generating SQL [Database Queries] From Natural Language With Yanshuai Cao

The TWIML AI Podcast

01:58 min | 2 months ago

Generating SQL [Database Queries] From Natural Language With Yanshuai Cao

"So tell us a little. Bit about touring and the motivation for it. How did the project get started right. So is this natural. Language database interface is a demo of anguish database interface built. And it's really just putting a lot of our word on some parsing space together. In this academic demo so netra language database interface the from application perspective the pin uses to a law a nontechnical users to interact with structured data. Set is there's lots of inside endure and You know who want to give out change for nontechnical users to to get those insights and from a research perspective. It's a very challenging natural english Problem because the underlying problem is you have to parse pasta in english or had our next languish than convert to see cole. And we all know. Natural language is ambiguous machine languages on bigger after resolve all amputate. He yard a too harsh correctly. Furthermore was different from compared to on other program. Language is the mapping. From adams. To see cole is under specified. If you don't know the schema really depend on what is the structure of schema and so he still model has to really learn how to reason using it. And in order to resolve all that may retail and correctly predicted the sequel and lastly this printer model some. You don't want to just work on this domain one. To work on demand is on databases. You're never seen before. So without st cross domain across database part of it and dodgers very challenging. Guess it's completely different. Distribution wants moved to different dimensions even

Cole Adams Dodgers
Seth Dobrin Talks About Trustworthy AI

Eye On A.I.

01:41 min | 2 months ago

Seth Dobrin Talks About Trustworthy AI

"We're gonna talk about trustworthy a i. It's something that is increasingly in the news and concerns a lot of people. Ibm has a product called fact sheets. Three sixty that i understand is going to be integrated into products. Can you tell us what fact sheets three sixty is. And then we'll get into the science behind. Yes so let me start by laying out what we see is the critical components Trustworthy a at a high level Three things there's a ethics there's govern dated ai and then there's an open and diverse ecosystem an ai ethics is fully aligned with with our ethical principles that we've published with arbin dr ceo co leading the initiative out of the world economic forum. And i'm adviser for essentially open sourcing our perspective on a ethics from a govern data in ai perspective. It falls into five buckets. So i is. Transparency second is explain ability third is robustness. Fourth is privacy and fifth is fairness and so the goal of fact sheets is to span multiple of these components and to provide a level of explain ability. That is needed to drive adoption and ultimately for regulatory compliance. And you think of it as a nutritional label for ai where nutritional labels are designed to help us as consumers of prepackaged foods to understand what are the nutritional components of him. What's healthy for us. What's not healthy for us. Factually is designed to provide a similar level capability for a.

Arbin Dr Ceo Co IBM
Everyone Will Be Able to Clone Their Voice in the Future

The Vergecast

01:49 min | 3 months ago

Everyone Will Be Able to Clone Their Voice in the Future

"World today often feels like it's full of digital voices with a assistant siri amazon alexa and google reading your messages announcing the weather in answering trivia. Here's what i found on the web but if you think things are chatting now just you wait. The voices of these a assistant used to be based unreal recordings. Voice actor spent hours talking in a studio and these clips would-be cut up and rearranged to create synthetic speech but increasingly. These voices are being created using artificial intelligence. This means we can not only create more realistic computer. Voices clone the voices of real people much more quickly creating endless artificial speech at the touch of a button for example it was surprisingly easy to make a synthetic version of my own voice. In case you missed that. That was not me talking. That was all made digitally by typing into a computer. So why would some want to do this. Besides the obvious novelty of it. You might have guessed a reason to make some money. I listen to this was going on. Kevin hart here. I wanna talk to you about why. We have to have mac and cheese every night. Think about it. That's why. I recommend thousands of new shows and this is a promo from baritone one accompany. That's working on an ai product to create synthetic voices and make them something. The media industry wants to us. So we've created a platform. Ai which at the end of the day turns unstructured data into structured data. That's shaun king executive vice president. Ed veritas one. So if you're thinking about audio thinking about video things that are typically unstructured and we make that searchable discoverable author a host of different a cognitive engines that are there from transcription beaker detection speaker separation. And then we provide those tools to you know many different industries that are eating

Amazon Kevin Hart Google Shaun King Ed Veritas
Interview With Patrick Bangert of Samsung SDS

AI in Business

02:01 min | 3 months ago

Interview With Patrick Bangert of Samsung SDS

"So patrick i'm glad to be able to have you with us on the program here today and we're gonna be talking. Ai at the edge particularly in the world of medical devices. Which is i know where a lot of your focus is here. We're gonna get into some of the unique challenges of leveraging data and ai at the edge in the medical space. But i want to talk first. About what kinds of products. We're talking about people think medical devices. Okay well medtronic is tracking my blood sugar on the side of my arm and you know. Then i've got a big cat scan machine kicking around over here. What kind of devices does your work involve with. And and his edge relevant From your experience. Thank you for having me on the show pleasure to be here. We are dealing with medical imaging devices. So if you have a smart watch on your wrist. That's not what we deal with. Even though those are very useful of course to measure your exercise and sleep patterns we're dealing with technologies like an ultrasound and mri is not an x ray. And what's called digital pathology which is where a biopsy is removed and put on a microscopic slide. Those kinds of technologies produce images that are relevant to telling you whether you're sick at all hopefully not or if you are what kind of disease it is. And so the job of computer vision in this case is to detect whether is a disease diagnose what it is to find out where it is to find out how big it is advanced in if cancer stage one. Three how advanced it is. And all of these outputs can of course be created. Virtually instantaneously by executing artificial intelligence models at the edge and the edge in this case is the device itself. Yeah okay so. Some devices are huge. Mri scanners take up a whole room. As some devices are quite small ultrasound. Machines view could transport it in your suitcase and so there's obviously also price difference here but nonetheless. All of these technologies do produce an image that that is then analyzed by

Medtronic Patrick Cancer
Social Commonsense Reasoning With Yejin Choi

The TWIML AI Podcast

02:07 min | 3 months ago

Social Commonsense Reasoning With Yejin Choi

"All right everyone. I am on the line with jin. Choi eugen is a professor at the university of washington. Yajun welcome to the air podcast and excited to be here. Thanks for having me. I'm really looking forward to digging into our conversation. I'd love to have you start by sharing a little bit about your background and how you came to work in the field of ai. Right so i primarily work in the area of natural language processing but like any other feels of ai. now the boundaries become looser losers and. I'm excited to work on the boundaries between language and vision language and perception and also thinking a lot about the connection between a i and human intelligence and what are the fundamental differences in that in terms of knowledge and reasoning And so let's go a little bit deeper into that. Talk us through like some of the ways that you take on those topics in your research portfolio. What are some of the main projects. You're working on the things that you're exploring right so currently i'm the most excited about the notion of commonsense knowledge and reasoning. This was in fact the only dream of a field. The in seventy eight as people love to think about it and tried to develop formalism for it. It turns out it's really trivial for humans but really difficult even for the smartest people to really think about how to define it formally so that machines can execute it as a program so for a long time. Scientists assumed that it's Doomed the direction. Because it's just too hard so i didn't really thought about commonsense for for a long time and then it's only in recent years. Some of us got excited to think about it again. Which is in part powered by the recent advancements of neural modell's that is able to understand large amount of data.

Choi Eugen Yajun JIN University Of Washington
Ultra Long Time Series

Data Skeptic

02:21 min | 3 months ago

Ultra Long Time Series

"My name is a foley counter. I work with essentially neurosis. They'll finance and economics in beijing. China background statistic computing and nowadays we focus on forecasting ways a lot of skill of data on distributed systems. So i haven't yet had the chance to interview anyone specifically about distributed time-series. It seems like that would be some extra challenges because the data sequential what happened before relates to what happens next. How can you spread that across. Many machines disputed hampshire is is just time is that alcohol. it can't be billions of observations. Historically we build up statistical models based on assumptions the narrative and other assumptions those assumptions do not work on distributed system and the industry like apache spark actually defacto standard for data processing and the the street people star a huge amount of data on distributed systems. We how to make a model that really works on sack disputed system and we have to work on their language to make our forecast more robust bus on his outer series. Yes spark is naturally a good choice to us because it's such a good reputation and a lot of reasons to look at it for big data solutions. But it's not obvious to me that it's necessarily the right choice for time series because it's not really baked in right. They've moved more and more towards like a sequel style and data sets. Are there any technical challenges to implementing time series via spark. And if you consider all single time that's fine but if you think about what we are doing we are streaming tate. Data is like times commun- out like water like re-re coming up now. You're really need nonstop system to process in the whole system. They simultaneously and without much delay that demand for temps is forecast in and out to claim that i think a lot of people agree with me nowadays arteta pam because we collect data. We always have the time stamp. So that's a windy. Temperatures for distributed systems. And there's a new challenge. I think emmanuel areas like atmosphere electricity and adi and other domains

Beijing Hampshire China Tate Arteta Pam Emmanuel
Are Amazon's Algorithm Bosses Coming to Your Workplace Next?

WSJ Tech News Briefing

01:37 min | 3 months ago

Are Amazon's Algorithm Bosses Coming to Your Workplace Next?

"You think your boss is watching you at work. Monitoring your every move well if your manager is a series of cameras sensors and algorithms. Then you're not wrong. And that may also mean you work at an amazon fulfillment center. The company known for the detailed tracking of packages and user. Information is also tracking the movements of workers at its warehouses looking to precisely measure efficiency and increase productivity wall street. Journal's tech columnist christopher mims calls this bazo schism named after amazon founder jeff bezos. He's got an upcoming book about it called arriving today from factory to front door why everything is changed about how and what we buy any joins me now. Hi christopher hay zoey. Thanks for having me so christopher you coined the term bazo schism. What exactly does that mean so. Basis ism is the combination of sensors and software to measure. How well somebody is doing their job. And then use software which has of course logic or an algorithm in it which was defined by an engineer. Somewhere to then tell that worker okay. You're doing a good job or you're not doing a good job or you need to be doing this differently. And so bazo. Schism or management by algorithm or management by software. At the end of the day it is just about creating a set of rules and then handing it to a machine to enforce those roles so the person's boss is

Christopher Mims Amazon Christopher Hay Zoey Jeff Bezos Journal Christopher
Andy Mauro CEO of Automat on Conversational Commerce

The Voicebot Podcast

02:08 min | 3 months ago

Andy Mauro CEO of Automat on Conversational Commerce

"Anymore. Oh welcome to the voice by podcast brett. This is a long time coming now. We've done a couple of clubhouse sessions. But we've i guess known each other at least through social media for several years. Now he's sort of back and forth have always appreciated your comments and our exchanges there and it's really nice that we have this one on one time to really talk about you've been doing because you have a long history in the industry you've seen a lot of different parts of it. You're doing some of the more interesting things. I think right now from a conversational standpoint in the market right now with automatic but it. We should start with where you started. So how'd you get into the industry. What did that look like early on. And what were you to the to. The tech. Sure and likewise. I've been looking forward to this for a long time. So excited to see where this goes i history. Hopefully it doesn't take too long. I've been working in what. I like to just say computers. You talk to for now over twenty years so i guess dedicated all but a couple years of my career to this space. I really love it. I mean i feel like it's a privilege to work in this right. I mean i think flying cars and talking computers. This is the stuff of childhood sci-fi dreams and you know. I really honestly feel lucky to get to work in the space for as long as i have and so it goes back to my days. My only job before. This space was at the canadian at the time unicorn nortel which is sort of a competitor. Cisco's in this back in the late nineties and Back when everything was just internet infrastructure was the big business. And i had a job. They're working on crazy low level. Ip over atm stuff. I was a programmer. I love that stuff. And i all. My friends started quitting one day and they were going across the street. And i said where you're going like this they said this cool startup nuance and i was like. Oh that sounds fun and just like you do in your early twenties. Just quit my job like literally the next day and over and got a job at this other place where all my friends were and man. That was lucky. That was just one of those life. Changing things i didn't know about conversation. We didn't even call it conversationally. I'm back that rain speech recognition.

Brett Nortel Cisco
Deep Reinforcement Learning for Game Testing at EA With Konrad Tollmar

The TWIML AI Podcast

01:48 min | 3 months ago

Deep Reinforcement Learning for Game Testing at EA With Konrad Tollmar

"Conrad woke him to the tuomo. Podcasts thanks sam. Thanks for inviting us to be here. I'm really looking forward to digging into our conversation. We'll be talking about As the audience might imagine the intersection of and games before we do. I'd love to have you share a little bit about your background. I mentioned what is k t h. Okay teaches royal institute of technology in stockholm. It's a technical university where i did my undergraduate as well as might be hd. So i i think my interest for a i started longtime ago starting with computer vision. I always been passionate about photography. And i saw them. There was an opportunity to combine my kind of interest for photography than webs kind of my academic. And the so. That's kind of my starting point here. Nice and tell us a little bit about the kind of research that interests you in your professorship and on your graduate studies so my phd more symbolic media spaces and we build different kinds of interactive in viramontes to connect places with vdi streams but also being able to use sensors to convey other kinds of information. If you're close or if you're in the proximity of a space for that led me and benchley to explore that further or after my ideas and i spent some time working smart and interactive environments some over this work for play and some were for more like everyday use and i think some of us could remember recall. The kind of demos sue sorted out the mit's media on the late nineties.

Royal Institute Of Technology Conrad SAM Stockholm Benchley MIT
Ethics Panels Reject and Delay Biometrics, AI Projects for Google, IBM, Microsoft

Daily Tech Headlines

00:29 sec | 3 months ago

Ethics Panels Reject and Delay Biometrics, AI Projects for Google, IBM, Microsoft

"With ai. Ethics chiefs at google microsoft and ibm published by reuters looks at what. Ai projects these companies reject in september. Twenty twenty google's cloud unit rejected a financial firm looking to us to determine credit worthiness and earlier this year. Google blocked features from analyzing emotions. Ibm turned down a client request for a more advanced facial recognition system microsoft place limits on software mimicking voices. All three companies said they welcomed clear regulation for

Google IBM AI Reuters Microsoft
A Powerful Intersection of AI and Robotic Process Automation With Merve Unuvar

AI in Business

01:54 min | 3 months ago

A Powerful Intersection of AI and Robotic Process Automation With Merve Unuvar

"So marvan. I'm glad to have you here with us on the show and i know we're diving into the topic of rpa intersection with a i. I think given the coverted era is a lot of thinking about gaining efficiencies about finding opportunities for automation when you're working with big enterprises obviously. Ibm works as many of the largest firms in the world. How do you walk people through finding those pockets where automation could make a difference. What does it look like spot opportunities in workflows yet. Thank you that. This is a very interesting area. Especially as he emphasized during this pandemic a company has realized that some of the workflows could be rethought through given most of their workforce with two remote working right so before we discussed this topic further. I'd like to open up the definitions of key concepts here for the audience. So what is it business workflow. It's basically an execution of business processes that contain tasks information and paperwork related to all of these right and then they're passed from one person to another to achieve a business school better. It could be alone operable for a bank or could be a claim submissions furnishings company. So this usually moms one or more people and a hub can best leverage automation in these workflows needs to be thought through in a few dimensions so the first one is from overall process and the workflow performance point of view so in order to analyze the performance. Right manner i. We need to understand the end goal of workflow if he thinks through the same mortgage scenario is the goal to sell more loans or is it to process loans faster or it can be combination of these metrics but we need to really define the key performance indicator or the goal of these workflows and then start monitoring the performance towards these goals and one of the very obvious waste of flying. The pockets of automation is then to find the bottleneck tasks in these workflows that will impact the

Marvan IBM
Jaron Lanier on the Future of Humans and AI

Lex Fridman Podcast

02:18 min | 3 months ago

Jaron Lanier on the Future of Humans and AI

"You're considered the founding father of virtual reality. Do you think we will one day. Spend most or all of our lives in virtual reality worlds. I have always found the very most valuable moment in virtual reality to be the moment. When you take off the headset and your senses are refreshed and you perceive physicality afresh. You know as if you were newborn baby. But with a little more experienced he can really notice just how incredibly strange in delicate and julia impossible. The real world is Sue the magic is and perhaps forever will be in the physical world. Well that's my take on it. That's just me. I mean. I think i don't get to tell everybody else how to think or how to experience retreat. At this point there have been multiple generations of younger people who've come along and liberated me from having to worry about these things But i should say also even in a what. I called it mixed reality back in the day in these days. It's called augmented reality But with something like a hall and even then like one of my favorite things to augment a forest. Not because i think the forest needs augmentation but when you look at the augmentation next to a real tree the real tree just pops out as being astounding you know it's it's interactive. It's changing slightly all the time if you pay attention and it's hard to pay attention to that but when you compare to reality all of a sudden you do and even in practical applications My my favorite early application of retrea audi which we prototype going back to the eighties. When i was working with dr joe rosa and at stanford med near near where we are now. We made the first surgical simulator and to go from the fake anatomy of the simulation which is incredibly valuable for many things for designing procedures for training things then to go to the real person. Boy it's really something like Surgeons really get woken up by the transition. It's very cool. So i think the transition is actually more valuable than the simulation

Julia Dr Joe Rosa Audi Stanford
Exploring AI With Kai-Fu Lee

The TWIML AI Podcast

02:31 min | 3 months ago

Exploring AI With Kai-Fu Lee

"All right everyone. I am here with kaifu. Lee chi food is chairman and ceo of innovation ventures the former president of google china and author of the new york times bestseller superpowers. And we're here to talk about his new book which will be released next week. A twenty forty one kaifu. Welcome to the tuomo. Ai podcast thank you thank them. It is great to have an opportunity to speak with you. I'm looking forward to digging in and talking more about the book before we do though i'd love to have you share a little bit about your background and how you came to work in the field of ai. Sure i started With my excitement in back in nineteen seventy nine. When i started my undergraduate at columbia i worked on language and vision at columbia and then i went to carnegie mellon for my team at which develops the first speaker independent speech. Recognition system based on machine learning actually Some the earlier thesis in machine learning in nineteen aba. I also developed a computer program that the world's fellow champion is all in the eighties. Very early years after mike graduation from Cmu i talked there for two years than i joined apple and led a a lot of apples. Ai speech natural language and media efforts later joined sgi and then microsoft where i started microsoft research asia in beijing in nineteen ninety eight which kind of became one of the best. Tom research labs in asia. Later i joined google and ran google china for four years between two thousand and five in two thousand nine. We did do a little bit for how they i mostly was Really developing google's presence in china in two thousand nine. I left google and started my venture capital firm assign ovation ventures and at san ovation ventures we invest in the bow for the ai companies. We were about the earliest and probably invested in the most companies we invested in about seven unicorns in ai alone and with a few more Yet to come so they're excited to be in the era i it's Was not so hot during much of my career. But glad scooby with the catch. The recent wave and participate in it.

Kaifu Lee Chi Innovation Ventures CMU Google Columbia China New York Times Tom Research Labs Microsoft Asia SGI San Ovation Ventures Mike Beijing Apple
Is the AI Market Saturated?

Eye On A.I.

02:08 min | 3 months ago

Is the AI Market Saturated?

"My first question is is the market saturated and without picking winners. What products us rising to the top. Good that you're asking this question right now in general timing of the world because here we are for those who are listening to podcast. August the twenty twenty. One people might be listening to this year from now. So this'll all seem really quaint. To those in the future but the markets actually in the midst of consolidation. Saying we're actually starting to see a lot of acquisition activity and we do track over one hundred vendors and machine learning platform space about seventy two of which meet the minimum threshold of viability. There's lots of startups in the space. We love startups. We have an affinity for companies of all sizes but when we're looking at companies who are buying products and services we tend to look at those companies that have either at least ten customers or have at least ten million in revenue or at least ten million dollars in venture capital they if they have like to customers and no venture funding raised in a little bit of revenue than. We're like just grow little bit more a little bit more. So this is about seventy two companies. At least that are in that that john rao of course all the cloud vendors are in that space. The major cloud vendors microsoft. Ibm google amazon And a few others. So those were recalled the cloud sas machine learning as a service vendors basically and then there's a whole other category of pure play machine learning platform vendor so you may be familiar data robot or did i do in that space a bunch of others that are kind of trying to pull together all the components of what's required to put machine learning and advanced analytics solutions into play and increasingly. What they're doing is they're growing through Both building out their product suites and through acquisitions so she did robots but on a tear lately did i. Two as well as been been really growing raise very significant round recently but the answer is that this market is actually starting to

John Rao IBM Amazon Microsoft Google