it's an awesome tool for making gifs from any video on the internet. put it on your screen -- you can buy riffy but you google.uy it's super cool and quite the headline but when you have another search engine that is so the ful, they don't have same database as you guys do or the same tagging system, et google but you know that wants this space. >> yeah, well, i think the gif you guys t to give some like reference on it, it's still small. doesn't know what a gif is. most of the world doesn't know what a gif is. there's room enough for to be part of the gif space. what we're trying to do is make the entire space popular. in gif e who is working technology, we want to work with you, like everyone. tried to do our best to work with everyone that's out there. technology, like gif grabber, and it's an amazing giphy capture.

pretty much anyone who's out there. gifs, it's working in a huge industry that's going to happen because we're basically just making paragraphs move. and so if you think about the photography landscape of how many companies are part of that industry, there's room for all of us. so we'd like to work with everyone. competition, you know, in terms of gif search engine, we're the largest gif engine. everyone has a particular -- that's what we do. other people are keyboard companies. people are headlines on the internet and social networks. search engine. >> absolutely. let's talk about giphy cam. hand if you know what giphy is. that's what i thought. up.e's five or six hands giphy cam i thought was like a genius idea and it still is, but i don't hear people talking about it. expected it to really take off. can you give us kind of a giphy cam? ort on i don't hear a lot about it anymore. >> yeah, giphy cam was our first

gif camera so it's an app for your iphone. itwas awarded -- apple named one of the best apps of 2016, 25.of the top it's won a lot of awards. filters ake a gif, put in snapchat and other things and send out a gif to your friends social. it's very popular with a small set of audience. as soon as people know it exists, more people will start using it. boomerang, snapchat filters. pretty much everyone is coming into the space which is good popularizing the idea of these short moving images across social networks and across messaging. so yeah, i think giphy cam was one of the first to kind of really start the trend but now ton of other apps. and as soon as more and more eople are exposed to these applications that you can actually take a gif on your camera and send it, live photos version of another a giphy cam. they just call it something else.

it something s else. but essentially they're just short silent videos just like out into the t world. >> and that goes into the whole idea of like gif has kind of as this term that isn't -- it's not always a gif right? pe, >> it's kind of like the kleenex offacial tissue or the xerox paper. when it was invented in '87, it format ecific specifically used for putting, sites on construction your web site, that when you were constructing that you were never going to build ever and some web ill see sites. i think i've seen a few that still have that thing on it. ut now it's just kind of used as a catch-all phrase for any 3-5-second video that loops and, you know, we have 20 ifferent renditions of every gif that comes onto the site, photos, so we support them. gif really is about the time

versus and the concept the actually format itself now. just like photographs aren't like silver plates anymore. they can be prints, digital, anything. >> swinging back towards giphy one of the interesting things about it was you guys that built a feed into app. in fact, there's no giphy that has that social feed aspect to it and i when my question is, you're taking something that is inherently fun, right, gifs are expressing fun and yourself. it's not a utility. it's not like uber is to transportation, right. we don't like need it. we want it. and so when you don't build that i want to into it, understand that decision making process, right. there's not a whole lot of giphy. ss to >> yeah, stickiness is a new utility. o you have two options on the internet. you have the option of creating

a social network and then trying out how to get people to stay there. so giving them utilities. or you have the option of utility and having people come use your utility every day. is theogle, for example, most popular site on the internet but isn't a social network. they tried it but it hasn't worked out. other social networks have tried myspace, had a lot of people and tried to build the utility and that got boring over time. has never been about building a social network. if you're going to go to a facebook, ork, go to go to tumblr. that is the community of people talking to gs and each other. we want to be a utility. we have the plumbing that takes all of that content into other places of the world. pinterest says they're not a social network. they are a utility. fundamentally s first a utility so people can express themselves finding gifs internetng them on the if that makes sense. >> it does make sense.

social network. >> if you ever started -- i had a social network before this. hardest thing to do. shout out to everyone who's ever made a successful social network. that is a struggle. facebook is kind of like it is social network. it's very hard to compete with facebook. utility, can build a that is like -- and people use it every day, that is the something to build meaningful on the internet. > can we talk about like for a little while. when you talk about giphying and he night of the oscars and you see lady gaga walking by leon d dicaprio. you had that up in a matter of seconds. how does that work. entire team of editors on the west coast and east coast that sit and make moments officially for our partners as they're coming out.

there ahead of everyone because we actually have the those the time making gifs. a example, oscars 2016 has officially y so we make all of those for them. we are the curators and creators so we're hoping all the partners that are having events that want gif versions of those create the contents. right. what would you say is the biggest vertical in terms of giffing. is it sports? entertainment? >> right now, it's just what you popular would be in culture. right now, it's a lot of award oscars, the vmas, the emmy emmys. golden globes. peris the nk that cas number 1 searched gif on the world. of all the content scene is from tvs, movies and celebrities.

there is money changing hands. we're doing a lot of work to bring gifts. a lot of people do not know what it is . we are to get to a point where our deals, where we will cut all and let them out in the world for publishers to come in right articles about, they are your contents and they allow us to do this.

it is a trade of work. you have raised a ton of money. we are about half a million? i think we have about 75 million. jordan: you are not bringing in any money. there is not revenue deals. what is the revenue plan? start of all of the people, we are still in exponential hockey and growth. it is not slowing down. market?is the gif all of them are into accelerating the growth. growthin the pre-revenue phase. it is hard to. we have grown 2x in the past two months.

if so, we will need another round of funding. jordan: there are a number of ways to do that. i think we should zero in on that. alex: ultimately we're going to make money. jordan: obviously. since they went it had a revenue plan in place. we have been thinking about this since day one. .ne is getting election deals there are a bunch of different revenue models. the major ones are mobile messaging. get ad an into mobile messaging? people are sending hundreds of millions of messages a day. have you donated advertising into mobile? gifs are great for that. 8% are from tv, celebrities we have deals with.

mobilerld where format, itnd and ad is a perfect medium. if we are serving a billion plus some advertising can be there and we can charge for getting the content into the. we do a lot editorial work to bring it all in and have those to all of the network. basically what google is doing of hyperlinks, if they were a moving image, they are making billions of dollars. we are doing it with highly more engaging images. right. i went to get back to the mobile messaging bit. mobile messaging in bringing a native ad is very interesting to me because the idea that i could flag one of my friends friday

bean and it shows to whiskey, it almost like under the radar. of that do not do any right now. we will work with partners in the future. we see a world when people are testing gifs through any medium, expression,king up hungry, send to a friend. adt is an entire kind of ecosystem. the is trying to figure out how we start making revenue on mobile messaging and the entire messaging platform. gifs are the perfect medium. models have to be explored and we will be exploring the incident. jordan: i agree with that.

my question is have muhtar kent that? how do you know what gifs they search for? when you think about launching an advertisement network within messaging which would be the themselves, how do you target at that point? what information do you use? alex: we use a life in doubt on where people are viewing gifs. opa we know that tender is looking for a bunch ofque. flirty gifs. any of thate information. that is to get the search better so we can serve better content. in the theater, that can be used to serve better advertising.

we also have our own tools that can directly send them as well. we also have a website that has a lot traffic. information on users coming in and there's the island. ir search patterns. jordan: the keyboard is a good point. if you have to messaging have made you know a lot. alex: we have a lot of info on our website. we have a lot of info from our partners coming in. jordan: i want to talk a little bit about giphy keys. took so long? i waiting for this for so long. friende did find a version since day one that we define different versions and day one. they were never white right. -- quite right.

we have invested in a couple of keyboard company. slash is an amazing keyboard. there have been some good keyboards. view,e found our point of out our view. we wanted to be minimal. it is not like you have to switch. once we discovered that, it was time to put out our own keyboard. the keyboard market is kind of small. anywhere that you can distribute a gifs we want to work with other partners to build apps. we how are a lot of other keyboards out there. we do not believe that one

keyboard will live the entire space. we think everyone will have a rd theyr a life -- keyboa like him we want to bring all the content to them. a lot of partners are using the api to distribute content. one thing we talked about it the are partnering with flash to launchn the fall an entire branded keyboard platform. platform wherel you can create a couple of buttons and create a branded keep her. we are offering all of our partners three rented keyboards keyboards.nded that is a good thing for the gif ecosystem. you can make customized

experiences for people. jordan: we are pretty much of time. oldast question is a age question. is it gif or jif? alex: we officially support both but -- jordan: that the most that minecraft i heard in my life -- diplomatic craft i heard in my life. jif is the old saying that people way back in the 80's talk about it as a nostalgic thing. we are trying to reclaim that term as gif. it means a totally different leakingan the 80's under construction sign. it is a new generation of

content that are being sent out. if you are sending second messaging, you are probably sending gifs. jordan: yes, say gifs. you heard it here first or the hundred time. fre toe go. a big round of applause for alex. we are going to shift gears in a violent way. tore going to go from gifs privacy and security that makes sense. be the government was protecting us from big corporations. it seems like they've corporations are trying to protect us from the government. panelis thatwo are going to help to take us through this. a big round of applause.

[applause] all right. welcome. thank you. us the ceo of hacker1. for discovering vulnerabilities. withwe also have nate that with the electronic frontier foundation focuses on cryptography and brief the issues. bitnt to start off a little broad. thick question i have is why do you think we are seeing an of the privacy of people information? when we built the internet 20 had to bow step.

now let connected to the world. is hitting the web system. we must protect them. it is a huge shift. nate: every photo we take are all documented online. we are still really bad security. we barely understand how to secure devices like all of you have in front of you. startedarely getting with this. the fact that companies like apple are starting to your it out is causing a challenge. speaking of fact, we have apple and the fbi. a facebook situation in brazil. was toulatory agencies

act as a consumer advocate to corporations. it seems like many companies are now marketing themselves as to being that asked kenyans of privacy. robuste do not have a legal model. -- european got out and runs it in front. in the out of the regulatory framework, we need something to protect our data. that is the pressure that apple is reacting to. it deals with the department of defense on one end. what do you think are the most common vulnerable systems?

marten: we have the department of defense. we have mover. general motors. all of them know there is no organization strong enough so it does not meet the health of hackers. the open of your program and they will tell you. everybody has vulnerabilities. every single system vulnerable. i you have been have to really do not know it yet. had been half or you do not know it yet. you think they are more willing to be open? about it.lk you gave your plan to the public.

we are seeing a reversal of that. it provides an open chart of systems. our companies were willing to the open? i was there witnessing the open source revolution. to be properly secure is to be open about your vulnerabilities and to invite you do not know to help you. i agree 98%. 2%, one is oracle. if you have not read the blog do so.lease they deleted in morning after she posted it. she is the chief security

officer. she said do not mess with our mef or reasonable, soo or ue you.me soo the other 800 pound gorilla that nobody was to talk about our embedded systems, medical .evice, voting machines they have never have to worry about security because they did not have network or a radio. internet of things -- matthew: you get the internet of shit. shit.nternet of interconnected cars and now it could be the apocalypse. they are not wrong.

those with no security staff do not know what to do with able mobility report when i am counseling a hacker or companyer, the software are nearly always seamless. i will always knows what to do with able mobility report. even oracle knows what to do. medical device companies do not have a clue. marten: i agree with that. in the old purity paradigm, people felt human beings were tech is theand solution. now we're finding it is human techs are the problem and is the solution. way yourthere is no

staff can do that. if the department of defense decide they are not secure enough without external help, you can do it? there are some that seem reluctant. aboveo acknowledge when -- bu are acknowledgedgs. lotare companies a resistant philosophically? i was told by apple engineer that these evil positives and negatives that choose not to fear what do you see the most common refrain? they are already taking that. they are on a journey. they will be good. no were no idea

that their system is vulnerable. breached, itas came through the air conditioning system because it was remotely controlled. no one had an idea. tell people who are completely ignorant. apple does have a robust security team. they know what to do with the vulnerability report. steve jobs apple was much more along the oracle line. that version of apple did not produce security white papers. in their updates the good not say what was being fixed or who reported it. it was a black box. apple is great. they are three quarters of the way there. leslie to inscription.

-- let's go to inscription. having down from single messaging to be corporations. people are really clueless about this. what do you think the chances are that something so aggressive gets passed by congress? how clueless is congress? in this congress the chances of anything getting past is slim to none. read theknow if people bill or stories about it. taken literally, the back door l would ban general-purpose computers which could not possibly have been their intense. they are. how naive

it is the next draft we have to we are here it is going to be introduced the week after next. matthew: one of the common things people talk about with encryption is the right we have to private communication. that is obviously one of the challenges in communicating with the government. you need to pursue wrongdoers. one and the thing i find interesting that is not get talked about as much is class problem we have of the poor not having access to sit is that our secure -- to systems that are as secure. do you think there is a solution for the problem or a way to move forward to getting everybody the

s?me right marten: we should do it if it is possible. nate: we have a privacy divide. as mostikes me of the apple/at the fbi's stances they should have access. any sophisticated actor will be even if an a signal item message goes back to non-encrypted. the fbi has only worried about the masses. that is going to further deep in digital privacy divide. it makes me deeply uncomfortable. matthew: what are the common

refrain of the default government agencies are people that are oppositional? i don't care about government hacking because i have nothing to hide. what do you have to say to people that take line? nate: my response is it is not about you. that is a deeply narcissistic position to take. i do not have anything to say but i still benefit from free speech because everyone around and theirerse opinion right to speak those opinions enriches our lives. even if i do not have anything to hide, it is not about me. in democracy and social change from the civil rights to the gay rights have to have on the ability

privacy away from government surveillance. you cannot have a gay rights movement anywhere in a theocratic state without privacy and security. is not about. it is about everybody else. i do not hear you think you have nothing to hide. you are not that special. it is not about you. matthew: when you are advising these companies and seeing here is why you need to open your code, are you seeing more companies think of the government heart of their rent model? you need to protect yourself against any malicious activity and all of the different groups whether , political

organizations, terrorists. theoes not matter from operator of a web system. you just need to find the vulnerabilities and remove them before they are being maliciously use. for us it is important to know where the attack comes from. for you as building the defense, it does not matter. to need to take every step protect yourself. you talked to companies when you are negotiating with .hem aren't you seeing more willingness to view the government and oppositional force? sometimes, yes. messaging with companies.

if you collect the data, they will come. that included attack risk, , intelligencee agencies. if the data is there, if you will have to protect it. one way is to not collected in the first in which the company put to great use. wh does nota have access. a keep it secure. tsapp matthew: it seems as though that become more prevalent. nate: that is what apple development line looks like. that raises a whole host of issues. --is hackerone parading preparing for situation? marten: our role is not directly to the side.

we are a marketplace of the allest community of hackers around the world a companies that need the surface. sometimes we are like the uber of security. we connect hackers with companies. nate: we have a white paper for online service writers best practice that is woefully out of date. have a cheatalso sheet for messaging system. nate: we actually just did it down. that to get down. we have a scorecard which he took down last week because they are updating it. hide the ball.to when exactly it

will go up. one last thing. the issue has arisen of info -- vs fingerprints. personal view of pincodes?versus this i use touch id on device. it is more robust than a pin. i turned it off through customs. it does notoot

work. there was an article in 2013 when it was first released discussing the implementation of touch id. amendment is the right to be free from compelled self-incrimination. i cannot be compelled to turn over my password except in certain circumstances. i can be compelled to give a breathalyzer test or left test will provide my fingerprint for a comparison. in that same light cases, the argument is pretty good that you can be compelled to put your finger on the device. think about whether you want different finger. something to keep in mind. matthew: got it. thank you both very much. [applause]

thank you very much. [captioning performed by the national captioning institute, which is responsible for its caption content and accuracy. visit ncicap.org] jordan: i cannot is called "building an excessive audience." we will get into it. it takes time for us to warm. wease welcome danielle moderator katie roof.

us.ie: thank you for joining h s is geared toward young women. it is a cliff notes of the news of the day. producers.be tv news what inspired you to quit your job? multi-form i. our mission is to make it easier to be smarter. they had a personal first-hand experience. geeks. up news we wanted to work there our entire life we met in college. we had the latter. we would go home and our

friends who are all smart and educated have great job and would ask us what happened in the world? we were being paid to know what was going on. they were being paid to know what was going on in their own industries. there was a gap of what you should know in the world and have an intelligent conversation. a niche we knew the missing link was to fit into their daily routine. we started our company with e-mail. the alarm goes off and you check your mail from friends and family first. katie: you have over 3.5 million subscribers. more than the new york times.

that is pretty impressive. program?u spread the we have ambassadors. the way we started were very organic. solving a problem. we ended up quitting our job. we were roommates living in a tiny apartment downtown. the first day we literally corralled all of our gmail and and we enoughcts about 5500 people and said we just quit our jobs. will you sign up for this? date well we got about 700 people to sign up. a forward.ted paying we were lucky that the president did of right away.

-- the press he did right away. mentor and ask a them to share. every time someone wrote it, we would say thing you, please share this with your friends. everyone with a sure. it was amazing. we started developing relationships. that our friends are calling you a brand ambassador. he became the thing. we highest someone that became a reader. we are a lot of stuff again the law, figured out what did not work and now we have 13,000 skimmbassadors.

has brought in rather 19% of our overall user growth. >> we are amazed if we look at the growth. we started from our account. only the fastest growing e-mail newsletter on the market, we are the third rate .orning tv show it was never about getting the audience to feel. who's making sure we engaged in the hunt is on the way. there is always going to be a drop-off. from the beginning you can't grow at how can you make the still love it?

one, voice. voice thatwe had a resonated. secondly, when people read the skimm they feel they are part of a community. they feel invested in our g rowth. launched anecently from a beyonce out for them to a job report. it is two dollars 99 a month. how may people have signed up? we are blown away by the response. we have never seen positive engagement. day one we had over 1205 star reviews. we are doing very well. when we think about this new product, as a company we make it

easier to be smarter. there were so many directions we could go in. we have become really good at saying no to the. it was exciting to say yes to an idea. we note this audience check their e-mail. -- we knowg that you this audience is checking your e-mail. we know they are doing that. we all live on our calendars. i have no idea where i am going to let my calendar tells me. that is that in truth. -- that is sad and true. our friends who are supersmart and have good jobs literally tell meit is great you the state of the union is on tonight but i have plans. i know you're going to recap it, and i would have watched it.

what it that way that we can fill? e-mail makes it easier. ahead is being let's be smarter about the future. it is leading to integrate into the iphone. we are staying here are all the you.s that we are telling about. it is really just about being canter and how skimm

introduce you to that. danielle: the primary revenue stream is sponsorship base. do in a way that is organic to our audience. have the subscription business. audience all works from the newsletter being the anchor. it was never lets create a newsletter company, it was less create this engaged audience. we can create different products. katie: with the native advertising, do you always let your readers know that it is

danielle: we do. it is clearly sponsored. it never touches the actual news. it has been our book section or wine section but it does not touch the news of the day. it was something we learned early on. we used to tell our audience that you are going to see an ad tomorrow. i felt like that we made a decision to be transparent. it is a free product. we are a young company. we need to make money. are open, people know there needs to be advertising a needs to be done well.

the weekend thank you notes from our audience. we do not work with brands. >> it has been really exciting to see the growth that has happened. danielle: it's exciting to see the revenue. uswas something that enables to not have to raise revenue. we used to say we are a pre-revenue company. it was a gift the way to say we did not have money. who is really calling the shots? danielle: we never thought about it in terms of that. it is a partnership.

we are in it together. if we are ever at a place where we have a different opinion, we come up with a plan. we are trying something out and get a run. there are not a lot companies with co-ceo. we looked at a lot of different models. we had never raised money before. we had a really strong partnership from day one. we were friends, roommate, business partners. it is unique. we get asked that a lot. when we checked in money, it was moneyure out who to take from. recently it was great.

with an earlymeet investor. they interrupted the meeting in front all of the partners and that i just have to say that you should be proud of the business but the two of you have such a unique partnership. you should be out in his doing so well. katie: it is good that you're getting along. you have a very small team. three people? carly: our team is 20 people which seems huge. to go from two people on a couch , leaving they are so great. sometimes we feel like the duggar model. katie: 20 people that work there total? carly: three people are on our

editorial team. are on our analytic team. mean. lean and littleusually there is a witty or start the comments .efore each news item i what to say the right into your back or have you determined that? we created a character. it has a very specific voice. background, what she likes to do on a friday afternoon, what she likes to do brunch. .hat health our writers it helps our business team think

about who she would work with. when we started in my friends would say i had no idea she was so funny and her friends say that about me. i think that is funny. we get asked where is that snarky park come from -- park come from? everybody has that no bs friend. gir represent that. skimm girl represents that. and how to integrate normal conversation into that. these helpfule

guides. n.w to use snapchat, bitcoi do you plan on expanding for the newsletters? the app, there are clearly a lot more products we would like to create we will get to in time. that came from our audience. writing in and said we know you are writing about the background of syria and i still have questions. a lot of this is really complicated. it is hard to sum it up in a paragraph every day. for those particular topic, we wanted to dedicate more time. from there, it is really

becoming a great thing that people are looking forward to. they know they can get it through the newsletter. it has been amazing to see. are actually our own content. you have a wide range of topics, pop culture, tech, finance. he cannot be experts on everything. have to come up with this? we never said that we were. we say we are the best in the business. they have political experts but translators. how can you translated in a way that people understand it?

people reach out that the i am an expert on this cannot work for me? experts. want people this to make it easier to be smarter. is there a voice that makes it easy to understand? when we were fund-raising people with a are you a media orbotech company -- or tech company? if we have to pick, we are going to be an audience company. that is our buzz word. what does our audience want to be ? everyone wants to feel they have an opinion is something to contribute to a conversation. you used to the tv

producers. is there going to be a video component? >> it would be a shame if we do not use our video background somehow. want to takee every time you look at your phone and i see so many of you , we wantt your own every time you are looking at your phone being the definition of meantime. -- me time. we are doing that in your morning routine. we're doing that in e-mail. now we are doing in your calendar. it is grazing to women entrepreneurs getting funding and doing well. do you have advice?

danielle: you have to really believe in what you are doing. we heard again and again "no, no, no." we were told e-mail is dead. there are two different paths you can take. one is made is not a good idea. the other is the our jobs, we believe in this, our friends like it. we want to fix this. we just did it. the easiest thing you can do is give up. kno your audience.

w there are very few facebook. that has been our true north. thank you. we are out of time but it was so conferencear in may. it took place in brooklyn, new york. good morning. can you hear me ok? all right. great. you. just out of curiosity, how many people know what the word viv mean? life we are going to use it to breathe. inanimate objects of our life through conversation. that is where we are headed with

this. roughly every 10 years a new paradigm, along that changes the way we interact with computers. think it is a secret what the next paradigm is going to be. we are not alone. pretty much every major technology company is investing billions of dollars in the intelligent system space. this is a race to the single interface for the user. we think a lot of the noise is a reallyarea important question. how do we take today's basic technologies and transform them experientially into something that is a true paradigm shift?

it is an essential aspect of what this is going to take to go to the net level. -- next level. you really just want one assistant. you don't want to have to look at think about which service providers you are going to be using to as what questions and what commands they require. you just want to ask. wait until you are looking at a hands-free device that has a hundred, it does not scale. you wanted to be personalized to you. to learnyour assistant from you, learn your preferences, learn whether you like irc or windows the plan, your food preferences for

fighting restaurant. you want it to know all of these things regardless of what device you are talking to. this needs to transfer. you do not want to have to start over every time you have a different device. the most important part about this is no one company in the has the resources to plug in every one of these are you might want to use. that is what viv has spent a lot time and money on. we think this will be the crucial thing that takes us from versions that each do 20 or 30 different things to doing tens of thousands of things. do, lets meet viv.

a couple of notes. are talking to pretty much all of the major companies around the world to fit viv different devices. putting this on the wrong today todayy.hone built soplatform is developers can come in and make things quite easily and create experiences that are straightforward. it also has the capability of allowing developers to create a much more complex experience. whetherng to use a view -- weather examples.

.tarting on the basic what is the weather like at home today? pretty straightforward. pretty straightforward type of query. was it raining in seattle three thursdays ago? see, viv has a much stronger understanding when you can teach and develop the natural language aspect of it. going to take another bigger delete -- leap. will be warmer than 70 degrees new the golden gate bridge after

5:00 p.m. the day after tomorrow? ok. this is a pretty sophisticated query. so what we're looking at here is this developer center. this is the place where and develop ll go new apps and teach new things. is load 'm going to do one of the queries that i did earlier. run that.to and then we are going to open the curtains a little bit and a look at what's going on inside. looks fairly straightforward here but there's actually something pretty extraordinary going on. the first thing that we do is

we have our friends from nuance e're using fur our speech recognition today turn the then you can rds see we have sophisticated natural language understanding that's something called an intense. here's where the magic comes in. so we've got a new technology that we've been working on it's a computer science breakthrough called dynamic program generation. it understood the intent it generated this program. so this is software that's writing itself. a really important aspect of scaling the ssistance, because every other platform like it has a program manager that says we're going to we're going to do something else. we're going to lay it out. nd they program exactly what happens when you say some query related to some domain.

hard-coated. but that doesn't scale. program that mic in 10 milliseconds, writes an lf, that creates execution program that goes out and ties the pieces of the dialogue, generates the lay out, does everything that happens after intent. let me show you one more example. here's another query that i did earlier. now, you remember this one a minute ago. sophisticated. milliseconds, it wrote a 44-step program that details out all of the around the context of the faculty that the golden gate ridge is a point of interest and when the day after tomorrow is, connecting all of the ervices in mind, it's pretty incredible technology. just a little bit of what goes

the scenes. a little eady to have fun? audience: yeah. all right. what i'm going to do now is show you where we put our emphasis on development of this. this is something we call conversational commerce. here is how easy can you make it to get things done talking to things, right? o let me give you a few examples. bucks for the drinks last night. up, nds from mail comes knows who adam is, knows what it's about. send going to go ahead and that. that's it. it's done. adam has got his money, one and it's done.

let me continue. send my mom some flowers for her birthday. from pro flowers come up with some beautiful arrangements here, but well she tulip lover. let's try that. tulips? t reat, so some beautiful arrangements here. let me go ahead and do that. where my mother lives, seamlessly puts all that together, and we'll go ahead and buy that. and that's it. way, owers are on the right. let's keep going. room in palm springs for labor day weekend. ur friends at hotels.com come up with some cool options. i've stayed at the andreas

before. it was pretty nice. i'm going to go ahead and get that. i think a deluxe room would be good. let's book that. has anyone ever seen a hotel booking that's that simple before? you guys. is this exciting stuff or what? . applause] all right. let's do one more. i need a ride for six people garden.ison square so our friends from mover will help us out here. they know a car that holds six a suv or xl.es request a ride.

looking.t, now, they're well, i'm sorry, robert. i'm going to have to cancel this hopefully we get a chance in the future. okay. see what we've got here. i'm getting a phone call. think? you guys this is all 100% live. [applause] we just did four transactions in two minutes by talking. so we're very excited about this stuff. showing you here -- ringing] e i don't know who's trying to get in touch with me, but i'm going them back.call

what i just showed you is just a the slice of where we see world headed. imagine when you've got hundreds developers of plugging in new services, and you're able to make the fficiencies of using conversational commerce like this. so this is just the beginning. in and look just this y at the inside of brain. what i'm showing you now is a walk-through of the actual capabilities on the today.of this so these are the actual models that the developers are building into. i'll call this the universe of capabilities that are in there today. and this is just with a few building our office these. so like we can take a little walk-through here. say, like -- we can go

look at what the weather person built. a sense for how to model something in here. models consist of. let's take a look at another area. i didn't really talk much about certainly but i'm going to partake in this tonight. what's in a sense for brains that we call it. and this is going to be -- you can get a sense of this universe what the developers will be adding to over time and you can imagine that with housands of people entering, the power that this is going to gain. summarize. consumers, this is going

to be the intelligent interface to everything. you're going to be talking to things erent kinds of that will be doing all sorts of different things for you and for developers, this is going to be next great marketplace, right. you've got app stores today. after thing that comes app stores is this new type of marketplace. for arketplace that works all the different kinds of devices that the internet of things will, and use cases that generate and the marketplace that will become the next big area. look ahead, let's say in the next five years or so, we hink there's going to be a new icon that's going to be very recognizable, as recognizable as wifi and that's going to be this one. ecause when you see this, you're going to be walking around in more and more devices this u're going to see icon and that means you can talk we think ing and that

will be the end result of this power over time. thank you very much you guys for watching. applause] >> thank you very much. that was very impressive. chat about it for a minute. what did you guys think? stuff? ting [applause] so we saw the demonstration app that you had there. very first question is sort of two-fold. ou know, when is interesting launching that we could use and then when is something launching that developers can start on? ding dag: so we're going to start a olling launch probably towards the end of theiary, and we'll be show casing various types of things i ike the various ways,

apps and phone apps and other things we're working on with partners and then we'll slowly tart opening up the developer world, and we'll start working with some select partners towards the end of the year and the ally open it toward full crowd. matthew: so the chat bot era is of y upon us now and most them pretty much suck, and then people are seeking kind of like behaviors that those are good for. so why is this going to be the chat system that makes our lives better instead of just talk to things we don't want to talk to? dag: well, the whole idea is go in and pers can build any kind of experience that they want and, of course, they're going to have a to -interested motivation make it work really well. so whenever you're looking at or the ls examples weather examples or any kind of commerce or any kind of app, the that are going to be building them are going to be super incentived to make it work

very well and the system is do t very easily to make it that. so it's teaching and training all the time. you can imagine when that goes things that ferent most assistants do today to 100 hings to 5,000 things, this becomes a much bigger part of your life and more, you know, i grow up kids will asking, how did you ever get along without your assistant? when i was in college, my kids are incredulous computer.n't have a so generally speaking, the these tion of all of developers building these incredible experiences together a very powerful medium. matthew: so right now you have a system that you have that have a desire to serve people directly, like amazon, for instance. to sell stuff directly out of their app, and you have the consumer on the other end of wants to buy the same. in the middle, we have this sort

interface that is made up of search. google is a big example. is this something that's going kind of eliminate that middle part and just bypass google entirely? dag: well, i don't think -- search isn't going anywhere, but i think you will ind out that the more capable these intelligent assistants comes, the more primary source as the gs they become user. when there's a thousand things i can do with it. gets.w how seamless this you don't really want to go back to the old way, right? let's think about what we take granted today and the experience of the internet. if you want to do something, you a url or you download an app. first of all, you have to have app, then you download the app. you have to sign up for the app. then you have to learn how to app.the and then you have to place it on your home screen or wherever else it is. these are a lot of steps involved in that.

it's easier to have this assistant with you. you state your intent and what it is you're trying to do and thatexperience, i think as becomes more powerful and developers band together and incredibly rich experiences, users will migrate world.s that new so no, i don't think search is going to disappear but i think assistant is e inevitable. matthew: and it sort of means goes from even installing an app that is notifications h or via like a text message or something like that to not even app at all lled the but still access its benefits. dag: yeah, so think about this of the internet of things, for example. you're not going to be downloading a piece of software refrigerator or to your mirror in the morning or even necessarily to your car. to live in the cloud and it all has to be dynamic. really nk that model is great for mobile, but there's

going to be this next world got this ve cloud-based art official intelligence agents that you'll talking to and they'll follow you around on all the devices that you have. something, uying consumer electronic device, taking it out of the box, wall, using into the it as some i.d. like you do with and having that thing say hey, dag, nice to meet you. set your preferences in this device. do you want me to tell you about it, right? kind of a new thing we're talking about and it's not that far away. so if you have your way, viv will be imbedded in download available for and enabled inside other apps. so it's not about launching a app.le it's about launching a system people can imbed in their their hardware and software. dag: we'll start out with a similar experience i showed you

today to show case the power of and invite the developers to have at it and use it for whatever self interest they have. watch the market evolve. it's hard to say where it will exciting to l be see. matthew: you've used the term cam brian explosion in the past terms of how fast a.i. will grow as a sector. hat do you think are the benefits a.i. will bring normal people? dag: it will just make things mundane and rally task-heavy much easier to do. and also this personalization you t that i talk about, don't really -- every time you're asking for something, you don't have to explain every bit of detail. starts to know you like a real assistant might. an efficiency game where it can help you. there's all sorts of interesting cases on how to apply that and one of the most interesting things is going to be to see viv when ens with people start applying all the creativity. think about, you know, apple launched the iphone in 2007 with

or nine apps t that they made. but what an incredible creative they ion happened when opened the app store, right. happening same thing in a.i. and we're kind of at the same stage in a.i. begin to imagine all the different ways people will start excited to t we're watch the development. matthew: so when developers build for viv, they're going to add sort of instructions to tell viv how to do new things. but you mentioned part of the things that the set it apart from other a.i. amazon like siri or echo, aside from the extensible platform is the dynamic generation. you explain a little bit more about when a developer creates a module that says hey, ow you know how to do this or that, what added benefits does that domain program generation give? dag: so you're honing in on a really important part of this new system. kind of a break-through.

the way g to change programmers work with computers, because they're no longer the red to actually teach computer step by step coding every single line. program we're talking synthesis here, right? dag: yeah, it's a variation of that, right. so instead of having to write xactly every code as instructed, you're really describing what it is you want it to do and modelling what you and the computer does the rest. so there's a lot of benefits to that. faster tot easier and actually program. box 's a bit of a black there so people who have bad intentions can't deliberately program -- they don't have creating wreaking the havoc other systems might have if they don't have a system like that. that's a super crucial aspect to this. important computer science break through that our science department came up with, excited about

that. matthew: you mentioned wreaking havoc. hawking said artificial intelligence could spell the end of the human race. says he's concerned about it. el akin to a aid it's demon. do you believe it might enslave s all or do you have an outlook? dag: i was waiting for this question. atthew: got to give the people what they want. i actually am glad these questions are coming up because i do think this is going to be road.ue down the we were joking a little bit last if there is some a.i. scanning the internet looking for something, one of the places they'll go is viv to over that network of capability. but no, i'm not worried about that. don't think this is the beginning of the end quite yet. o, i'm not too concerned with that issue today.

matthew: so this is the second you've in this vain that started. what is it about the conversati u.i. that nal convinces you it's the future? dag: it's just easier. it's just a natural way for us to interact. i had a guy i was talking to when i was about 65 and he was asking me about what i was doing these days. him and he it to said well, it's so techie. i don't really understand all that stuff. actually technology is coming to a place where it gets even easier for you to use to hold e you know how a conversation. it's just a natural way for humans to interact. they've been doing it thousands years. if that became a de facto interface for pretty much that you do or at least many things where it's appropriate, i think that's very owerful and radically simplifies the world, especially with tech. atthew: so you've -- and you cofounded siri, sold that to apple, and there have been many start-ups in the

a.i. stage that were acquired to companies as the a.i. system. as you mentioned, everybody sort a department working on this. if mark zuckerburg took you for walk in the facebook world and offered you the kingdom of the world, would you do that, or are see this ined to through the natural inclusion. to well, we're in talks acquire techcrunch right now, so not sure how that would work. kidding. our goal for this is ubiquity, going to follow the path that gets us to ubiquity. that's very important to us. had acquisition offers previously that we've not gone with. matthew: anybody i would know? dag: yeah, you would probably know them. read about it probably. we're friends with all these that nd there's a lot of interest and stuff going on in this area but we are going to whatever we think

is the right way to get to ubiquity. we're not going to predetermine what path that is but we're determined to finish the job for sure. matthew: and we touched on this your le bit but does vision for viv include hardware, systems? ded dag: we're not having any plans to build any hardware ourselves, are talking to many hardware providers to as i was talking about, into many types of existence. n fact, there's some interesting approaches we've had that i can't tell you about today, but i wouldn't have even it before.t of it's really fun to see the reativity of what may happen here. matthew: well, if i promise not to call your uber driver and got you what happened, i the number off the screen, but you'll tell me back stage. much, dag.very appreciate it. thank you. [applause]

[applause]. yeah, i guess you guys need that at some point. else notice he spent like $1100 on stage? a lot of money. we're going to switch gears again because that's how this going s going, and we're to focus on fixing tech's culture problem. some of these people we've had on our stage before and excited again.e them welcome eric o'baker from slack, derrick brown from intel, romero from paradigm and vicky.erator, megan rose >> so as we all know or should know, the diversity in tech conversation is nothing new. evolved in the last

year or so. erica, can you tell me how that evolved? on has >> previously, people didn't really talk about diversity in at all.h if anything, they would speak how we need more women in take maybe and now people are all aspects ook at of diversity in tech from the gender problems, to race problems, exclusion of people disabilities, exclusion of young.who aren't so the conversation has been a multifaceted. >> before we came on stage, you you weren't the biggest fan of the title of this panel. right. >> fixing tech's culture problem. tech has so much of a culture problem. i think it has an exclusion problem. the default is to exclude people patterns match the that people expect, so i think

once we fix the exclusion on some of focus the issues that underlie that, racism, sexism, i think we can a lot more progress than just looking at it as like a culture problem. danielle, speaking of there on, so at intel, are 3.5% african-american in 2014. that year, it was 3.4%. so there's nothing much of an improvement. we seeing much growth there? >> yeah, so what i would say learned on our own journey is that this has to be a tech erm commitment that companies make. this is not overnight change. it's difficult to move numbers, a company like people.at's 107,000 you have to be in it for the long haul. i do think you've got to set goals, which we've certainly had some success with in setting goals around hiring retention.

but you do have to really be ommitted to driving the change over time, and it never happens fast enough for our liking. that. readily admit >> and what's retention like at intel? >> so for about the past decade we saw our diverse employees leaving at greater umbers than their non-diverse counterparts. so our underrepresented minorities and women were intel at greater numbers. last year, we were able to stop hat trend and even out our retention rates where women and underrepresented minorities were leaving at comparable rates to non-diverse counterparts. >> what changed there? hy are women and underrepresented minorities not rate? g at as high of a >> yeah, i think we've really tried to invest in making inclusion a very significant initiative that's top-down.the ur ceo brian kirzanich made some pretty big bold

announcements about the importance of diversity and inclusion in our workforce. measurable goals. we tied every employee's pay to chieving those goals and we started releasing our data and i think our diverse employees opefully sort of realize that, hey, this was more than lip service. this was something that we again, cared about and, we were committed to for the long haul. we've invested a ton this year training, in culture training for all our employees, and i think it's starting to the needle, megan. we're still not where we want to be but we're on the right track. said you tie employee's pay to diversity numbers. everyone? company, yes.n the so we have a company-wide bonus program. every year, we set a number of corporate goals. for the past two years, we've hiringoal around diverse and diverse retention and every single employee at intel gets hit those numbers. a really powerful way, i think, o say you're serious and

committed. >> chris, do you a lot of work paradigm at companies like pinterest. b & b, based on your experience with those companies, what advice do huge ve for intel, this company that's trying to tackle diversity and inclusion? > i think one thing that intel is doing really well is taking a data-driven approach to addressing g and barriers to diversity and inclusion, and that's really the way that we work with our clients. i think the way that kind of diversity and inclusion initiatives have traditionally noticing ached is by that there's a problem, looking around and seeing what are other ompanies doing and kind of relying on so-called best practices that aren't really and what we arch want companies to do is look at your uniqueee where barriers are and then based on strategies to n address those particular barriers and don't wait one to years to see if the numbers

change. look at the proximal outcomes you're can see if what doing is working sooner. >> what are some of the most glaring culture problems you've in your work? >> yeah, something i've been thinking about a lot is this ulture of genius or culture of brilliance that we have in tech. so really similar to something graduate school called fixed mind-set. lot of research is showing that people's beliefs about the nature of talents and abilities, whether they're fixed, stuff color, which is referred to as a fixed mind-set or whether you believe that talents can be developed and more of a growth mind-set. have these duals beliefs? individuals who have a fixed focus more on o improving themselves, and they give up more easily. individuals with the growth mind-set are interesting in improving and remain resilient in the face of challenge. it turns out companies and even can have these

mind-sets too. and when companies or fields ave this more fixed mind-set view and when there's a culture of brilliance, people that don't kind of fit our stereo type who is brilliant in our society are not only less attracted to those companies in fields, but once they're here, there's more pressure where you have to feel like you have to prove yourself. which is about group ability. do we get rid of this stereo type? >> in terms of growth and fixed mind-sets, there are a lot of things companies can do to growth mind-set and i know something intel is this.g on are you focused on innate traits? are you saying you're so x, or are you helping people focus on the process, understand what they effort they put in, the strategies they use. or openly f success

talking about mistakes and making it okay to take a risk, even if that doesn't necessarily lead to success. those are the things that we now from the research to help develop more of this growth mind-set culture. some of the t are culture problems you've seen at slack around diversity and inclusion? slack is a fast-growing company and we hire a lot from other companies in tech. know, people come with their biases that they've formed and so other companies it's a challenge to try to, you get them to think in ways that are different. but i feel like we're doing a of that.cent job i think that it's important for talking about how slack, to matters to continue working on inclusivity and i feel like if we keep doing that, you know, we'll have a pretty decent culture. i think it's important that we're tarting now while

still a small start-up. i've said frequently that it's right. turn a tanker, so you know, a big 100,000-person company like big 50-000 orw, a more person company like google, to change t's hard the culture of those companies. but if you start small, start at start-up stage when there's 25, 100, 500 people, it's easier and ke in diversity inclusion at that point so that when your company grows your already where it needs to be. is anielle, of course intel huge, and you've been in your role for a little less than a year now. do you wish intel had brought ago? n 10 years >> yeah, you know, i talk to a lot of my peers at start-ups and a lot about is when you're small, now is the time to set those aggressive bold goals. is the time to start examining your data and really go about building your culture

deliberately because at intel we've had, you know, 40 culture decisions and building, and this is very hard to drive change when you've had very long sort of tradition rooted. so i think for start-ups, the is really now to seize the opportunity and do things differently. i think there's so much small companies have to really get this right. you recently launched project include. in ou teamed up with women tech such as ellen powell, rita just kind of this dream team of diversity in tech. and so with project include, the idea is to provide concrete start-up ation for found ers and for tech. feedback been like vc the start-up and community. >> we've gotten great feedback

vc m the start-up in community. last i checked, we were at 800 our interest page. that was a few days ago. is that interest saying to you? 00 people saying we want to learn more about what you're doing. >> definitely. there are a large portion of bcs e who are partners at who want to sign up for our ceos at program and start-ups. there's one start-up i won't particular but like five people from their company signed up on the page and they reached tracy and reached out to everybody they knew at project involved. try to get >> that company has a serious problem. >> or they may want to get things right. but may not have a problem they really care about making sure they do things right and i like we're going to continue to see that sort of positive feedback. so project include is a bit of a side project for

you. mentioned back stage that they powell at pinterest, actually gave her a time to work on it but a slack that time was not offered even though about 20% of your job is supposed to be inclusion. >> so i don't know that they wouldn't have given the time. i didn't ask for it. i should have asked for it. to tracy did give the time pinterest to work on it. i think it's important. it is very important for us to change the industry and biases in the industry right now. ecause, you know, we're having this boom of start-ups right now and to make sure that those diverse and inclusive, when they grow to be big companies, you know,

set the tones will for the next one so it's now. ant to set it in >> last thing on project whatde, i think, we'll see happens. >> all right. >> so it's a pretty hefty resource. i admittedly have not gone of it and i am just being know, honest -- i am someone who really cares about diversity and inclusion in the industry. how do you get people who may about as passionate diversity and inclusion to really look into all the resources on project include. we get them to look at statement? >> yeah. >> we've broken it up. not everything will be as to everybody. so we've broken it up into different sections and people can go and look at what they that moment and i would hope they would look at the whole thing but if there's something they want to focus on, different sections for

them and also there are little guidelines on the pages that, ou know, this applies to companies that when they're like 25 or more employees, so if a seven-person startups up, you don't need to worry about that. pretty hefty s a piece of work so we try to make intosier for people to get it. >> are you tracking which parts of project include people are looking at? analytics we have happening somewhere but i'm not looking at it. we had a long discussion about that, yeah. that ielle, you mentioned it's really never too early to start working on diversity. advice do you have for the startups in the audience right now? >> sure. big thing that we have learned on our journey is start with the data, right. i think we are by and large an engineering culture and what measuring where you a

and setting concrete goals, really being very clear about how you intend to report on them. what we're finding is, again, data, understanding what the real issues are. we're now reporting our data which, again, keeps us accountable and setting real holding employees accountable for them is truly needle.ves the i tell people don't be afraid of the data. it's probably not going to be where you want it to be. ours certainly isn't. if you don't know what you're dealing with, it's really hard change. this change ined the area of diversion and inclusion in the company, brian at reverend jesse jackson's there's erence, that actually been back lash inside of the company. could you tell me a bit more

about that? yeah, certainly. so i think anytime you embark on -- culture e change of this magnitude. clear, diversity and cultural change is big in the tech industry. that's okay. it should turn you off. it's big culture change. anytime you embark on that, you who wholeheartedly embrace it and are really excited about it. y the way, i think that's the majority for intel. we've had so many employees tell us how much they appreciate and our commitment. at the same time you've got a lot of people that this is challenging the status quo. threatening. and people worry with what it means for them. i think that's human nature. continuum of where people are at on their journey and i think our job is to bring conversation forward. bring them to the other end of say that hey, nd true inclusion means you include everybody. it's not about excluding the favor of the minority. it's creating a place where everybody is heard and valued.

that change doesn't happen overnight but it is worth it. brian also mentioned it had something to do with white men threatened. do you agree that has something to do with it. megan is we haven't done enough job of bringing our asian male population into the conversation. i don't think you can do without engaging that majority population and when i these people d is want to be part of the conversation but it's sort of frought with land mines and sure about how to go about engaging. and i think we can do a better engaging our peers in the dialogue, talking about issues like white privilege and things talk aboutst hard to and creating a safe space where you can get the real issues out instead of sort of keeping them under the surface. i think that's the next step we really need to take as an industry. >> definitely. >> erica. >> i found that the key to ringing people into the

conversation is making it okay for them to be uncomfortable, right. saying up front like this is going to be an uncomfortable conversation. that's okay. e're all going to have some discomfort around this. own it. nd then you can move forward with the conversation. >> i see what you're saying people in to be part of the solution, i think it's really important and we see in is what research in large metaanalysis looking at what makes diversity one of the main moderators that the research finds is that the managers feel like they're a part of the not just a part of the problem. that's going to be more likely to make whatever kinds of more s you're doing successful. >> right. brings me to what are the costs of implementing these either y policies, financially or culturally and e've kind of touched on culturall

culturally. it takes time. like one thing that -- and i honest u have to be about that. this is not something that happens overnight. that you need ng multiple people to be involved in. longer to t takes bring people in that are different because those people aren't traditionally exposed to industry, and so you need to be honest that that's going communicate r and the value of that to people. eople understand like this is worth our investment. this is work taking more time to make this happen. take's definitely going to a lot more time, and it's going to take some different sort of training. do the t continue to same things over and over again that you've been doing and results.fferent so, you know, the time it's going to be is much more and it's going to take a lot more training about how to audiences, rent different groups and i feel like in tech a lot of companies are

sed to like the influx of people coming to them, especially for hiring. it's like oh, we don't have to find about going out to people but, you know, it's a different way of thinking to go people who aren't coming to you in that. >> you know, i do think you've this as an investment just like any other investment you do in your business. that nly intel announced we were investing $300 million theolving this challenge by year 2020. that's big money, right. but i think it's also because the problem ng really holistically, looking at chain, our hiring practices, our culture and retention and the way we train invest in supply ers and startup. it isn't cheap but big important issues are rarely inexpensive to tackle and i really think again, problem to attack the holistically and with a lot of solutions, not just money, not just training. things. >> and to be clear, so intel

layoffs.announced some do those layoffs have any effect and tel's diversity inclusion efforts? >> yeah, so no, they don't. again, i think our commitment to diversity and inclusion is because we truly believe it's to intel's inclusion as a company and that's something we continue to be committed to, our folds absolutely have not though some of the business emperatives we're undertaking right now. it's still important and we've to be committed to it in the long term. >> so all three of you had a revious relationship with each other, right, or you knew who v were.rs -- so for ng companies who maybe don't have ahead of ces to hire diversity and inclusion or hire a consulting company like like gm or have someone you who's just willing to step

away from engineering from time -- to to work on these can on this issue, what startups do right now, or what are the solutions for them? >> project include. i mean, that's what we've right. it for, we know that companies don't have the money to spend on can't forward to bring on ahead of diversity and inclusion right now because they more engineers to actually build their products so e created project include to give them this road map, this guide book and recommendations worry don't have to about spending money and they also don't have an excuse not to do it. value ink it's a lot of of having a place where all of research is synthesized for people so people aren't relying to spend best practices but are works and out what how to design the culture from excited ning so really about the work you guys are

doing. >> thank you. i would say two things. i think the first thing is reach out. so whenever i'm asked to speak to another company about diversity, however big, however small, i always make the time to do it because i believe if there's one thing i'm doing that might help another company, there are no secrets. it elevates the experience for everybody in the industry. i think the second part and you certainly don't need a chief diversity and inclusion officer this, just gather the data. see where you are today. data tells you and set one or two goals you'll take on this year that moves the in a meaningful way. >> speaking of the data. tech company diversity eports, we rarely see stats on lg lgbtq status, eligibility status. slack did include lgbt status common? sn't that why aren't we seeing that very much? >> i think one reason is don't collect that

data so we need to do a better job asking people to self because if we're not measuring these things it can be hard to track what is the people in like for these different groups. >> and i also think companies might be scared to track certain ask somebody , to lgbtq.heir -- if they're it seems a little scary, like traditional h.r. manuals say you talk to your employees about that, right. so it's a tricky place and i think we can grow there for sure. stats there any other that you think are missing from these diversity reports? >> definitely. data.'t see retention so the stats that we see are like percentages of employees, retention.t see we don't see how many harassment settled out of court. we don't see how many people are taking the parental leave that's offered. there's a lot of data that just isn't in those reports that speak to like what the culture is like at a

company. though you're starting to see companies on that data continuum. our year we started showing retention data. we started showing our pay data. we're t thing i think going to start looking at is diversibility , and veterans data. i think you've got to start somewhere. analysis you get the paralysis, if i don't show everything, i'm not showing anything. everybody has to start somewhere and for better or for worse, people have started with women nd broaden it to underrepresented minorities. i think where the future is eaded is showing data about everything. i think you've got to realize showing your data is a good thing. afraid of hing to be and the flood gates open. >> last thing and we touched on but earlier, institutionalized racism and discrimination. those are hard topics to discuss people.e how can we get everyone in this about eling comfortable talking about it?

>> it's tricky. it's an uncomfortable people, ion for many right. it's hard for people even outside of the tech industry to alk about institutionalized racism, but i feel like it's very important. should people in tech look at themselves as being isolated from the rest of society. like we're in this bubble. different from everybody else. but the reality is, you know, and e come from everywhere there's institutionalized racism sexism and bias against people of all different sorts come to terms with the fact that we aren't different from the rest of work on nd we have to and address these issues also, progress. some good >> that's all the time we have. thank you so much. investments.round of this is about 25 minutes. >> i'm very excited to do this

panel. it, use, i mean, let's face valuations are done, the ipo market for tech company seems all but shut. there is a lot of nervousness out there and yet somehow vcs together more and capital commitments in the first quarter than they have since 2006. if you are wondering what is going on, join the club. we're hoping these three savvy help us out today. guys, thank you for joining us. hris, you are an investor in josh's fund. biggest question is, it seems like vcs are returning to their investors much faster than historically. i think maybe over the last decade it's been standard to or four und three years. that might be conservative. xl, light seeing speed, founders fund coming back. . at's going on >> as an lp, i'm kind of the money behind the money. people in my seat fund people

like josh and andy and i used to endowment and 's i work in the fund's group. so we're kind of a sun light in venture world.e all of the energy in a sense comes from us. f the lps stop showing up the trees will kind of wither. that sound grandiose? maybe. but one of the challenges that lp capital isn't infinite like some people think is. in fact, dollars going into venture capital compete at the institutions where these dollars come from against other asset classes in terms of what's the attractive risk of return. so one of the challenges venture aces is that it's further side of the money, longest dated option that most institutions buy. so as a result, there's a lot of pressure on venture. and one of the things that's challenging is you kind of discussed over the last few periodicity of

fundraising has shortened. two-year cycle. he challenge is investing for nirvana for an lp, to the point to e you are mature enough recycle distributions. hopefully each dollar comes back in three of its friends relatively short order. shorter cycles were four or five years and liquidity cycles were longer. gone on now as fundraising cycles have shortened, liquidity cycles have gotten longer so i describe it sphincter. it >> he did just say what he thought you said, by the way. >> so we're pushing this capital out and kind of feeding the money, nd lps give gps gps give to startups, startups get liquid and georgia back to lps and it starts over again. when you have the snake getting meler and fuller, it reminds we're in brooklyn so not exactly

but there's a song method man where one of the methods of torture they talk say i'm going to sew your rectum shut and feed you. what it's like to be an lp today because the dynamic he escribed is really an unsettling long term dynamic for long term ecosystem health. is it happening? i know on the one hand it seemed startups were raising unding every six months or so but critics very recently said these guys are racing back to investors before their paper gains disappear. -- ou think that's so >> so it's a really important point because right now, a lot of venture funds are looking paper, right. and in fact, one thing that's a pet peeve of mine is i get day that say our fund is market whatever and somebody said to me the other just had a bunch of

mark-ups so our fund is sitting on cash. no it's paper on cash. market stuff before business school and we always talked about the pit you get in turns omach when a trade into an investment. similarly i fear as an industry we're headed for this moment unrealized -- because everybody has unrealized things in the portfolio. there's a moment when the unrealizable omes and that's the moment of nausea. bill is right. last fundts to be the in the market when the music stops. it's going to be a game of lp ical chairs and from the perspective, everybody is tapped out both in terms of capital. people blow through their companies come and we're all d so exhausted. there's this kind of psychic lps.ustion amongst everybody has been running too hard and kind of reacting rather than being proactive. andy, you're from union

square ventures. 166-ish million dollar fund. >> we don't really talk about it. >> okay. filed a form with the s.e.c. suggesting as much and you had in 2012 so a fund you're sort of sticking to your traditional trajectory it sounds like. josh, what are you. you raised your last fund in 2014, $175 million. you in the market this year? >> we typically raise -- target raising every two and-a-half years. that would put us on pace to raise at the end of this year. okay, great. do you feel s, like -- you can say no. you obviously would never say no good reason. when i spoke to bill he made this point that it's very hard say no.because you you can lose your seat at the table possibly for the duration career as an lp. do you feel that's true? >> that's absolutely true and