(upbeat music) >> Live, from Washington DC. It's theCUBE. Covering AWS Public Sector Summit. Brought to you by Amazon Web Services. >> Welcome back everyone to theCUBE's live coverage of the AWS Public Sector Summit here in Washington DC. I'm your host Rebecca Knight, co-hosting alongside of John Furrier. We are excited to welcome to the program, General Keith Alexander former NSA Director, the first Commander to lead the US Cyber Command, Four-star General with a 40 year career. Thank you so much for coming theCUBE, we are honored, we are honored to have you. >> It is an honor to be here. Thank you. >> So let's talk about cyber threats. Let's start there and have you just give us your observations, your thoughts on what are the most pressing cyber threats that keep you up at night? >> Well, so, when you think about threats, you think about Nation States, so you can go to Iran, Russia, China, North Korea. And then you think about criminal threats, well all the things like ransomware. Some of the Nation State actors are also criminals at night so they can use Nation State tools. And my concern about all the evolution of cyber-threats, is that the attacks are getting more destructive, the malware has more legs with worms and the impact on our commercial sector and our nation, increasingly bigger. So you have all those from cyber. And then I think the biggest impact to our country is the theft of intellectual property, right. That's our future. So you look out on this floor here, think about all the technical talent. Now imagine that every idea that we have, somebody else is stealing, making a product out of it, competing with us, and beating us. That's kind of what Huawei did, taking CISCO code to make Huawei, and now they're racing down that road. So we have a couple of big issues here to solve, protect our future, that intellectual property, stop the theft of money and other ideas, and protect our nation. So when you think about cyber, that's what I think about going to. Often times I'll talk about the Nation State threat. The most prevalent threats is this criminal threat and the most, I think, right now, important for us strategically is the theft of intellectual property. >> So why don't we just have a digital force to counter all this? Why doesn't, you know, we take the same approach we did when we, you know, we celebrated the 75th anniversary D-day, okay, World War II, okay, that was just recently in the news. That's a physical war, okay. We have a digital war happening whether you call it or not. I think it is, personally my opinion. I think it is. You're seeing the misinformation campaigns, financial institutions leaving England, like it's nobody's business. I mean it crippled the entire UK, that like a big hack. Who knows? But its happening digitally. Where's the forces? Is that Cyber Command? What do you do? >> So that's Cyber Command. You bring out an important issue. And protecting the nation, the reason we set up Cyber Command not just to get me promoted, but that was a good outcome. (laughing) But it was actually how do we defend the country? How do we defend ourselves in cyber? So you need a force to do it. So you're right, you need a force. That force is Cyber Command. There's an issue though. Cyber Command cannot see today, attacks on our country. So they're left to try to go after the offense, but all the offense has to do is hit over here. They're looking at these sets of targets. They don't see the attacks. So they wouldn't have seen the attack on Sony. They don't see these devastating attacks. They don't see the thefts. So the real solution to what you bring up is make it visible, make it so our nation can defend itself from cyber by seeing the attacks that are hitting us. That should help us protect companies in sectors and help us share that information. It has to be at speed. So we talk about sharing, but it's senseless for me to send you for air traffic control, a letter, that a plane is located overhead. You get it in the mail seven days later, you think, well-- >> Too late. >> That's too late. >> Or fighting blindfolded. >> That's right. >> I mean-- >> So you can't do either. And so what it gets you to, is we have to create the new norm for visibility in cyber space. This does a whole host of things and you were good to bring out, it's also fake news. It's also deception. It's all these other things that are going on. We have to make that visible. >> How do you do that, though? >> What do you do? I do that. (laughing) So the way you do it, I think, is start at the beginning. What's happening to the network? So, on building a defensible framework, you've got to be able to see the attacks. Not what you expect, but all the attacks. So that's anomaly detection. So that's one of the things we have to do. And then you have to share that at network speed. And then you have to have a machine-learning expert system AI to help you go at the speeds the attacker's going to go at. On fake-news, this is a big problem. >> Yeah. >> You know. This has, been throughout time. Somebody pointed out about, you know, George Washington, right, seven fake letters, written to say, "Oh no, I think the King's good." He never wrote that. And the reason that countries do it, like Russia, in the elections, is to change something to more beneficial for them. Or at least what they believe is more beneficial. It is interesting, MIT has done some studies, so I've heard, on this. And that people are 70% more like to re-Tweet, re-Tweet fake news than they are the facts. So. >> Because it's more sensational, because it's-- >> That's food. It's good for you, in a way. But it's tasty. >> Look at this. It's kind of something that you want to talk about. "Can you believe what these guys are doing? "That's outrageous, retweet." >> Not true. >> Not true. Oh, yeah, but it makes me mad just thinking about it. >> Right, right. >> And so, you get people going, and you think, You know, it's like going into a bar and you know, you go to him, "He thinks you're ugly." and you go to me, and you go, "He thinks you're ugly." (laughs) And so we get going and you started it and we didn't even talk. >> Right, right. >> And so that's what Russia does. >> At scale too. >> At scale. >> At the scale point. >> So part of the solution to that is understanding where information is coming from, being able to see the see the environment like you do the physical environment at speed. I think step one, if I were to pick out the logical sequence of what'll happen, we'll get to a defensible architecture over the next year or two. We're already starting to see that with other sectors, so I think we can get there. As soon as you do that, now you're into, how do I know that this news is real. It's kind of like a block-chain for facts. How do we now do that in this way. We've got to figure that out. >> We're doing our part there. But I want to get back to this topic of infrastructure, because digital, okay, there's roads, there's digital roads, there's packets moving round. You mentioned Huawei ripping off CISCO, which takes their R and D and puts it in their pockets. They have to get that. But we let fake news and other things, you've got payload, content or payload, and then you've got infrastructure distribution. Right, so, we're getting at here as that there are literally roads and bridges and digital construction apparatus, infrastructure, that needs to be understood, addressed, monitored, or reset, because you've had email that's been around for awhile. But these are new kinds of infrastructure, but the payload, malware, fake news, whatever it is. There's an interaction between payload and infrastructure. Your thoughts and reaction to that as a Commander, thinking about how to combat all this? >> I, my gut reaction, is that you're going to have to change, we will have to change, how we think about that. It's not any more roads and avenues in. It's all the environment. You know, it's like this whole thing. Now the whole world is opened up. It's like the Matrix. You open it up and there it is. It's everything. So what we have to do is think about is if it's everything, how do we now operate in a world where you have both truths and fiction? That's the harder problem. So that's where I say, if we solve the first problem, we're so far along in establishing perhaps the level so it raises us up to a level where we're now securing it, where we can begin to see now the ideas for the pedigree of information I think will come out. If you think about the amount of unique information created every year, there are digital videos that claim it's doubling every year or more. If that's true, that half of, 75% of it is fiction, we've got a big road to go. And you know there is a lot of fiction out there, so we've got to fix it. And the unfortunate part is both sides of that, both the fiction and the finding the fiction, has consequences because somebody says that "A wasn't true, "That person, you know, they're saying, he was a rapist, "he was a robber, he was a drugger," and then they find out it was all fake, but he still has that stigma. And then the person over here says, "See, they accused me of that. "They're out to get me in other areas. "They can exclaim what they want." >> But sometimes the person saying that is also a person who has a lot of power in our government, who is saying that it's fake news, when it's not fake news, or, you know what, I-- >> So that's part of the issue. >> It's a very different climate >> Some of it is fake. Some of it's not. And that's what makes it so difficult for the public. So you could say, "That piece was fake, "maybe not the other six." But the reality is, and I think this is where the media can really help. This is where you can help. How do we set up the facts? And I think that's the hardest part. >> It's the truth. >> Yeah, yeah. >> It's a data problem. And you know, we've talked about this off camera in the past. Data is critical for the systems to work. The visibility of the data. Having contextual data, the behavioral data. This gets a lot of the consequences. There's real consequences to this one. Theft, IP, freedom, lives. My son was video-gaming the other day and I could hear his friends all talking, "What's your ping start word? "What's your ping time? "I got lag, I'm dead." And this is a video game. Military, lagging, is not a game. People are losing their lives, potentially if they don't have the right tactical edge, access to technology. I know this is near and dear to your heart. I want to get your reaction. The Department of Defense is deploying strategies to make our military in the field, which represents 85% infantry, I believe, some statistic around that number, is relying on equipment. Technology can help, you know, that. Your thoughts on, the same direction. >> Going to the Cloud. Their effort to go to the Cloud is a great step forward, because it addresses just what you're saying. You know, everybody used to have their own data centers. But a data center has a fixed amount of computational capability. Once you reach it, you have to get another data center, or you just live with what you've got. In the Cloud if the problem's bigger, elasticity. Just add more corridors. And you can do things now that we could never do before. Perhaps even more importantly, you can make the Clouds global. And you can see around the world. Now you're talking about encrypted data. You're talking about ensuring that you have a level of encryption that you need, accesses and stuff. For mobile forces, that's the future. You don't carry a data center around with an infantry battalion. So you want that elasticity and you need the connectivity and you need the training to go with it. And the training gets you to what we were just talking about. When somebody serves up something wrong, and this happened to me in combat, in Desert Storm. We were launched on, everybody was getting ready to launch on something, and I said, "This doesn't sound right." And I told the Division Commander, "I don't agree. "I think this is crazy. "The Iraqis are not attacking us down this line. "I think it's old news. "I think somebody's taken an old report that we had "and re-read it and said oh my God, they're coming." And when we found out that was a JSTARS, remember how the JSTARS MTI thing would off of a wire, would look like a convoy. And that's what it was. So you have to have both. >> So you were on the cusp of an attack, deploying troops. >> That's right. >> On fake information, or misinformation, not accurate-- >> Old information. >> Old information. >> Old information. >> Old, fake, it's all not relevant. >> Well what happens is somebody interprets that to be true. So it gets back to you, how do you interpret the information? So there's training. It's a healthy dose of skepticism, you know. There are aliens in this room. Well, maybe not. (laughing) >> As far as we know. >> That's what everybody. >> But what a fascinating anecdote that you just told, about being in Desert Storm and having this report come and you saying, "Guys, this doesn't sound right." I mean, how often do you harken back to your experience in the military and when you were actually in combat, versus what you are doing today in terms of thinking about these threats? >> A lot. Because in the military, when you have troops in danger your first thought is how can I do more, how can I do better, what can I do to get them the intelligence they need? And you can innovate, and pressure is great innovator. (crunching sound) And it was amazing. And our Division Commander, General Griffith, was all into that. He said, "I trust you. "Do whatever you want." And we, it was amazing. So, I think that's a good thing. Note that when you go back and look at military campaigns, there's always this thing, the victor writes the history. (laughing) So you know, hopefully, the victor will write the truthful history. But that's not always the case. Sometimes history is re-written to be more like what they would like it to be. So, this fake news isn't new. This is something where I think journalists, historians, and others, can come together and say, "You know, that don't make sense. "Let's get the facts." >> But there's so much pressure on journalists today in this 24-hour news cycle, where you're not only expected to write the story, but you're expected to be Tweeting about it, or do a podcast about it later, to get that first draft of history right. >> So it may be part of that is as the reporter is saying it, step back and say, "Here's what we've been told." You know, we used to call those a certain type of sandwich, not a good-- (laughing) If memory serves it's a sandwich. One of these sandwiches. You're getting fed that, you're thinking, "You know, this doesn't make sense. "This time and day that this would occur." "So while we've heard this report. "It's sensational. "We need to go with the facts." And that's one of the areas that I think we really got to work. >> Journalism's changing too. I can tell you, from we've talked, data drives us. We've no advertising. Completely different model. In-depth interviews. The truth is out there. The key is how do you get the truth in context to real-time information for those right opportunities. Well, I want to get before we go, and thanks for coming on, and spending the time, General, I really appreciate it. Your company that you've formed, IronNet, okay, you're applying a lot of your discipline and knowledge in military cyber and cutting-edge tech. Tell us about your company. >> So one of the things that you, we brought up, and discussed here. When I had Cyber Command, one of the frustrations that I discussed with both Secretary Gates and Secretary Panetta, we can't see attacks on our country. And that's the commercial sector needs to help go fix that. The government can't fix that. So my thought was now that I'm in the commercial sector, I'll help fix the ability to see attacks on the commercial sector so we can share it with the government. What that entails is creating a behavioral analytic system that creates events, anomalies, an expert system with machine-learning and AI, that helps you understand what's going on and the ability to correlate and then give that to the government, so they can see that picture, so they have a chance of defending our country. So step one is doing that. Now, truth and lending, it's a lot harder than I thought it would be. (laughing) You know, I had this great saying, "Nothing is too hard "for those of us who don't have to do it." "How hard can this be?" Those were two of my favorite sayings. Now that I have to do it, I can say that it's hard, but it's doable. We can do this. And it's going to take some time. We are getting traction. The energy sector has been great to work with in this area. I think within a year, what we deploy with the companies, and what we push up to the Cloud and the ability to now start sharing that with government will change the way we think about cyber security. I think it's a disruptor. And we have to do that because that's the way they're going to attack us, with AI. We have to have a fast system to defend. >> I know you got to go, tight schedule here, but I want to get one quick question in. I know you're not a policy, you know, wonk, as they say, or expert. Well, you probably are an expert on policy, but if we can get a re-do on reshaping policy to enable these hard problems to be solved by entrepreneurs like yourself expertise that are coming into the space, quickly, with ideas to solve these big problems, whether it's fake news or understanding attacks. What do the policy makers need to do? Is it get out of the way? Do they rip up everything? Do they reshape it? What's your vision on this? What's your opinion? >> I think and I think the acting Secretary of Defense is taking this on and others. We've got to have a way of quickly going, this technology changes every two years or better. Our acquisition cycle is in many years. Continue to streamline the acquisition process. Break through that. Trust that the military and civilian leaders will do the right thing. Hold 'em accountable. You know, making the mistake, Amazon, Jeff Bezos, says a great thing, "Go quickly to failure so we can get "to success." And we in the military say, "If you fail, you're a dummy." No, no, try it. If it doesn't work, go on to success. So don't crush somebody because they failed, because they're going to succeed at some point. Try and try again. Persevere. The, so, I think a couple of things, ensure we fix the acquisition process. Streamline it. And allow Commanders and thought leaders the flexibility and agility to bring in the technology and ideas we need to make this a better military, a better intelligence community, and a better country. We can do this. >> All right. All right, I'm thinking Rosie the Riveter. We can do this. (laughing) >> We can do it. Just did it. >> General Alexander, thank you so much for coming on the show. >> Thank you. >> I'm Rebecca Knight for John Furrier. Stay tuned for more of theCUBE. (electronic music)

>> Announcer: Live from Orlando, Florida. It's the Cube. Covering, Grace Hopper Celebration of Women in Computing brought to you by Silicon Angle Media. >> Welcome back to the Cube's coverage of the Grace Hopper Conference here in Orlando, Florida. I'm your host Rebecca Knight. We're joined by Jarvis Sam, he is the manager of global diversity issues at Snap Inc. Welcome. >> Thank you so much for having me. I'm really happy to be here. >> So, I've gotta--first of all, you're wearing a Rosie the Riveter shirt, we've got these tchotchkes here, can you explain to our viewers a little bit about them? We got to, we got to talk about these first. >> Of course, so, the shirt was actually inspired by our Lady Chilla, that's our local women employee resource group at Snap. The idea was take the ghost, a representative mascot of Snap Inc. and parlay that with the idea of Rosie the Riveter, of course powerful in her own right. >> Rebecca: Alright, I love it, and then these spectacles are...? >> Yeah, so spectacles are Snap Inc.'s first ever hardware product released earlier this year. They allow for you to take an in-the-moment Snap, to be featured on your phone, using Bluetooth technology for iPhones and then WiFi technology for Android. They allow individual users to record Snaps on their phone, while of course not distorting the experience of being able to use their hands in the moment. >> Rebecca: So, I love it, these are the recruiting tactics: your own products. >> Exactly >> Want to play with these toys? Come work for us? >> Yes! >> So, tell us a little bit about what you do, Jarvis. Before you were at Snap, you were at Google. You were interested in really engaging in these diversity issues. So what do you at Snap? >> Yeah, so, at Snap, I manage our global diversity effort. What that includes is analyzing the diversity framework across three key verticals; first on the pipeline layer. So, what are we doing by way of K-12 education to ensure communities of color as well as women-- >> Rebecca: K-12? Wow. >> Exactly. >> Have specific opportunities in the space to be impactful. We often create this framework or archetype for what we think is ineffective software engineer for example or account manager. Reframing that by providing access and opportunity is showcase to people that the image that we have is not always the image that we want to portray, is critical. Next then we focus heavily on the idea of the candidate, so candidate experience. Deep diving into understanding key talent acquisition measures as well as key HR practices that will allow for us to create the best experience, moves us forward in that regard. But then finally, and this is where we get to the whole global perspective. Is the idea of the employee. Creating a nurturing community where the idea of psychological safety is not only bolstered but ensuring that your community feels empowered to the idea of inclusion. Making sure inclusion is not just a seat at the table but rather a voice in the conversation that can be actioned upon. >> So I want to dig into that a little bit, this voice in the conversation. Before the cameras were rolling you were talking about these very difficult candid conversations that employees at same have. Tell our viewers a little bit more about that. >> Yeah, so I think one of the greatest challenges across the tech industry and at Snap as well is the idea of referral networks. The tech industry on its own right has grown so greatly out of referral networks. People that you have worked with perviously, people that have the same academic or pedagogical experience as you. The problem with that is, the traditional network analysis would seem to let us know that you often refer people who look like you, or come from a similar internal dimension background as yourself. In a community that's largely rooted in a dominated discourse by white or Asian males. That means that we're continuing to perpetuate that exact same type of rhetoric. >> Rebecca: That's who you're recruiting. >> Exactly. And so then idea of getting more women or communities of color involved in that space can often be distorted. So that remains a challenge that we as a company as well as the tech industry need to overcome is understanding; one, how do we encourage more diverse referrals over time. But then two, creating an ecosystem where this seems natural and not like an artificial standard. >> Okay, so how do you do it? I mean that we've pinpointed the problem and it absolutely is a problem, but what are the kinds of things that Snap is doing to improve the referral process? >> So it's the idea of being innovative by design. One thing that's unique about Snap in particular is that we are an LA-based company. >> So based out of Venice Beach and Santa Monica, California. We don't face a lot of the core challenges that we see in Silicon Valley. And as a result have the opportunity to be more innovative in our approach. As a result when we look to referral networks in particular. One thing that Snap has focused on is the idea of diversity recruiting as a core pillar or tenant of all of our employee research groups. Not only do they join us to attend conferences like Grace Hopper, like the National Society of Black Engineers. But we actually do sourcing jams. Where we sit down with them and mine their networks. Either on LinkedIn-- >> Rebecca: Sourcing jams? >> Yes >> Rebecca: I love it. >> Yes Either on LinkedIn or GitHub or any of the various professional networking sites that they work on. Or technical networking sites to find out who are great talents that they've worked with before. >> Who do you know? Who can join us? >> Exactly. And what's more significant than that, is creating a sense of empowerment where we actually having them reach out to their network as opposed to a recruiter. This creates more of a warm and welcoming environment for the candidate. Where the idea of being a simple passive candidate is further explored by activating them to showcase how your experience has been great. >> And how are you also ensuring that the experience at Snap is great, particularly for women and people of color? >> Yes, so one area is our employee resource group. So we have a couple, so Lady Chilla is of course what I am wearing today. But Snap Noir for the black community. Snap Pride for the LGTBQ plus community and Low Snaps for the Latin X community. >> Rebecca: How big is Snap, we should just-- >> Yeah, about 3,000 people globally. >> Okay, 3,000. Okay, wow. >> And so one of the exciting things that we do is ERG that. So it's where we bring all of our employee resource groups together and they hold massive events every single quarter. To encourage other communities that are either allies or individuals of the sociological out group to understand what they do. But this deploys in so many different ways. In June, for Pride for example, we held drag bingo. Where our LGTBQ plus community participated. In March, we did a whole series of events celebrating women in engineering, women in sales, and women in media that resulted in a large expanse of events allowing for people to come in and learn about, not only the female experience more broadly, but particularly at Snap and some of the great endeavors that they're working on. >> And I know you are also working with other organizations like Girls Who Code, Women Who Code, Made with Code. Can you tell the viewers a little bit more about Snap's involvement. >> 100% Made with Code is one of the most exciting projects that I've had the opportunity to work on. It was for me personally this great combination of working with my previous employer Google, and Snap. So Google's Made with Code project is an idea that started to empower teen girls to code, ages 13 to 18 primarily. What they found is was that's exactly the same demographic that primarily uses our product. And so about three months ago, we decided to come together to launch an imitative where we'd have teen girls make geofilters, one of Snap's core products. The project actually launched one week ago, and teen girls are using Blocky technology to actually go about creating their own geofilters. And then writing a 100 word personal statement defining what their vision for the future of technology is. I'm personally exciting to say after checking the numbers this morning, more than 22,000 girls have already submitted responses to participate. And they will culminate in an event, November 1 through 3. Where we will take the top five finalists to TED Women in New Orleans. To not only showcase women who have done incredible things in the past and present. But also showcase their work at participating in this competition, as the women of technology for the future. >> Rebecca: And the next generation. >> Exactly. >> So we're running out of time here, but I want to just talk finally about the headlines. It's very depressing, you know the Google Manifesto, the sexism that we've seen against women. The racism in the industry. These are are-- we don't want to talk about it at this celebration of computing because we want to focus on the positives. And yet, where do you feel, particularly because you have worked at large tech companies, on these issues for a while now? >> Not facing challenges head on is going to be the greatest threat to the tech industry. The idea of avoiding conversation and avoiding sheer communication of these challenging issues will continue to raise-- >> Rebecca: And ignoring the bad behavior. >> Exactly, and it results in negative rhetoric that inherently put these communities out of wanting to work in this specific industry. But arguably given that technology not only represents the face of the future but how every single product and entity is made for the future, we have to include individuals. Everyone often wants to highlight the McKinsey study from Diversity Matters. Highlighting all of these great ways of diversity impacting business, but we need to look at it in addition from an ethic standpoint. The idea that technology represents how we are building our future. Leaving entire communities out of that primarily focusing on people of color and women, will result in a space where these communities will never have access, opportunity and thus employment to exist in this space. Being able to attack these issues head on, address the bad behavior, highlight what the potential implication is step one. Step two though is being proactive in everything that we're doing, to attempt to ameliorate that from the beginning. You'll notice one thing that's very different about Snap's diversity strategy is we seek to build infrastructure first, then focus on talent acquisition. Once we can ensure that communities of color and women are entering a space that is psychologically safe, open, and inviting. Then we can focus on how we're bringing in talent effectively so that the idea of retention and advancement is not an afterthought but rather top of mind. >> Right, because you can't recruit them if they haven't had the opportunities to begin with. >> Exactly, and that's what Snap often upholds the value of the idea that diversity is our determination, while inclusion is our imperative. >> Jarvis, I love it. >> Thank you so much. >> This has been really fun talking to you. >> Thank you. >> We will have more from Orlando, Florida at the Grace Hopper Celebration of Women in Computing just after this. (upbeat music)