Skip to main content

tv   Cambridge Analytica Data Privacy  CSPAN  June 5, 2018 3:34am-5:51am EDT

3:34 am
he was joined by a political science professor from tufts university and the director of telecommunications studies at the university of florida. this is two hours and 15 minutes.
3:35 am
3:36 am
thank you to everyone for coming. the facebook matter involving alexander cogan and cambridge analytic, shed a bright light on the data practices of some of our largest technology companies. although advertisers and political campaigns have collected and used to data for years, the public seemed generally unaware. this story has forced both the public and lawmakers to confront serious issues that need to be addressed, including the role congress should play in promoting trans-parity -- transparency regarding collection and use while ensuring a well-functioning marketplace for our data
3:37 am
dependent technology to drive further innovation. we started that conversation with mr. zuckerberg last month. i hope at today we can continue a productive and meaningful debate about these serious allegations. unfortunately, events like these more often than not, tend to get muddled by artisan ship and efforts to score a quick soundbite. the facebook story first broke december 25 -- december 2015. somebody had allegedly transferred data to cambridge analytica and breached policies. according to cambridge press releases, and a recent internal report, facebook requested
3:38 am
cambridge and its affiliates to remove any data received from the doctor. cambridge said that they removed the data and filed legal certification to facebook saying as much. i had requested that cambridge analytica appear at this hearing to explain these facts and tell their side of the story. cambridge recently commenced proceedings and therefore determined they could not participate in this hearing. the underlying story has not changed much since 2015. except for two different events. first, cambridge began doing work for the the trump campaign and secondly, the trump -- president trump won the election. these two that sounded an alarm that revived the cambridge
3:39 am
story. this does not diminish the importance of this discussion. it only highlights the extreme partisanship in play and more importantly, that this conversationd y taken place 20 this conversation could have taken place much earlier. advertising agencies and political campaigns have utilized tools for many years. campaigns including those of presidential candidates in every election year since at least the 1990s have used data to micro target. during the past presidential elections of these strategies have expanded to social media platforms, specifically facebook. president obama's campaign developed an app utilizing the same facebook feature that
3:40 am
cambridge used to capture the information of not just the app users but also millions of their friends. president obama's app potentially pulled even more information than cambridge's app. a former obama campaign official, carol davidson, recently wrote, ceok was surprised that we were able to suck out the whole social graft. in the 2012 election. we could also be talking about more recent events like buzzfeed's partnering with multiple democratic and trump super pacs in 2017. buzzfeed's advice president said that one of the problems buzzfeed was working with other partners to solve was, how are we going to get women who do
3:41 am
not like hillary clinton to vote for her? that type of voter outreach is not surprising to many. that is because it happens all the time. similarly it shouldn't be surprising that president trump's campaign used consultants to help reach voters as well. regardless of these events, and whether such tactics are actually effective, it is clear that the use of data across the political spectrum is only increasing. instead of creating this as a partisan issue to score political points, the important policy discussion should really have is whether tech consumer and congress, where we should go from here. our tech companies have access to some of our most sensitive data. are these companies doing enough to properly disclose
3:42 am
their data policies and protect their user -- their users? many of the services offered by such tech companies provide huge benefits to consumers at little to no cost. our consumers blissfully unaware or are they making informed choices with respect to how their data is collected and used? in 2015, the consumer technology sector provided many jobs and generated more than a lot of output and labor income and tax payment. how do we ensure the proper amount of regulation to protect consumers without damaging an industry that has been vital to
3:43 am
the economy? these are the questions we should be asking. i hope today's hearing will allow us to continue that discussion. >> thank you very much. thank you for holding this hearing. in march of this year, a seri of articles and videos were published online regarding cambridge analytica and its efforts to use personal facebook data of millions of americans to influence united states elections. to date, numerous governments have launched formal investigations into the company , including the united kingdom, australia, canada, nigeria, kenya, and india. there is much we do not know about cambridge analytica. but there are significant facts already in the public record. we know that cambridge analytica was established by robert and rebecca mercer in 2013 at the urging of former
3:44 am
white house chief strategist steve bannon as an american subsidiary of a london-based firm. has been reported that the intent of creating an american -- was to give the appearance of compliance with the united states election law that prohibits foreigners from working on united states elections. according to ceo alexander, cambridge analytic worked for candidates in 44 united states ections in 2014. during the 2016 election cycle, he stated that cambridge analytica, did all of the research, all of the data, all of the analytics, all of the targeting. we ran all of the digital campaign, the television campaign, and our data, rmed al
3:45 am
for the trump campaign. in addition, cambridge analytica obtained detailed personal information on approximately 87 million people from facebook without their knowledge. the massive da set, which reportedly included approximately 4000 data points on each individual, was used by cambridge analytica to develop a comprehensive voter targeting and online behavioral influence tool called project ripon. this was a software program that used sophisticated algorithms to allow campaigns to segment tersinto groups based on psychological characteristics such as, neurotic or introverted. once
3:46 am
individuals were identified in groupthe platform then provided preselected and focused group tested images and keywords that were most likely to alter the behavior of those individuals. examples of the messages loand used by cambridge analytica included keywords such as drain the swamp and deep state as well as images of border walls. in an undercover video, cambridge analytica managing director mark koelbel explained that cambridge analytica also branded defeat crooked hillary. they also created hundreds of online advertisements for that brand, including online videos that were viewed 30 million times. through project ripon, selected images were then sent to the
3:47 am
relevant indivials roug online advertisinservic like google and facebook. these websites provided feedback on an individual's reactions to those advertisements which were then set automatically -- sent automatically back to the targeting program. this is what we have learned in the past several months, however significant questions remain. there is still much we do not know about cambridge analytica. we do not know the extent to which it worked with hackers to illegally obtain damaging information on candidates, including the united states. it was reported in 2015 that cambridge analytica has parent company facilitated the hacking and theft of sensitive medical records from a nigerian
3:48 am
presidential candidate and published them online. we do not know whether cambridge alytica used these tactics in the united states. this pattern of activity was certainly used by russian intelligence during the 2016 election. we do not know the teof cambridge analytica connections to wikileaks or other russian interests. it has been reported that alexander nix contracted wikileaks in june 2016. he said this was his only contact with wikileaks. however, s former partners have suggested that cambridge analytica is the first contact with julian assange was between 12-18 months prior. it has also been reported that other employees atambridge analytica had direct
3:49 am
connections to julian assange including through his former attorney. in 2016, alexander nix also provided white papers and briefings to utives from russia's second largest oil firm about cambridge analytica's political activities in the united states. the oil company is currently under united states sanctions related to the russian activities in the ukraine. in march 2017, they revealed its former information sharing partnership with the russian federal security service, the successor to the kgb. finally, we still do not know whether the data obtained by cambridge analytica was ever shared with or obtained by a third-party. the data was
3:50 am
originally obtained through a facebook application developed by a russian born professor named alexander cogan. professor cogan maintains a teaching position at st. russia, state-funded university and has traveled frequently back and forth to russia. these are concerning questions. not only for the united states, but for all democracies around the world. based on what we have already learned, there is no question that the future of data privacy will have a significant impact on every aspect of our lives, including our basic constitutional rights. today, we are going to hear stimonfrom christopher wylie who served as research director at cambridge analytica from june 2013 until november
3:51 am
2014. i understand he will be able to share insight into some of these significant questions based on his first-hand experience. i look forward to hearing from him and i thank you again mr. chairman for holding this hearing. >> i am not going to introduce our witnesses and after i do that i would ask you to stand and be sworn in. also it was our intention that senator feinstein and i come down and shake hands with you and welcome you. i am sorry that everything was caught here and we didn't want to mess everything up but both of us want to thank you for participating and particularly you mr. wiley, coming as far as you have to help us with proper testimony. our first witness is dr. aiden hirsch. he is an associ
3:52 am
professor from tufts university focusing on american politics. he spent 7 years as an assistant professor at yale university. he is a nationally published author. he has a masters degree in political science from harvard university and a bachelor's degree from tufts university. next, besides thinking you we welcome you to this hearing and hopefully you feel comfortable. i know you have testified elsewhere. ristophewylie is a former director of research at cambridge analytica. he worked with president obama's campaign for the canadian liberal party and for the united kingdom's liberal democratic party from 2013 to 2014. he also was studying for a phd.
3:53 am
he has an undergraduate degree in law from the london school of economics. lastly, dr. mark jamison is a visiting scholar at americ emprise institute. he is also the director and professor of public utility research center, the university of florida's college of business . dr. jamison earned a phd in economics from dulles college of business in florida. would you three please stand? do you affirm that the testimony you are about to give before the committee will be the whole truth and nothing but the truth so help you god? all have affirmed so we will start with professor hirsch and then mr. wiley and then dr.
3:54 am
jamison. >> thank you. in queue for inviting me here to london today. >> your microphone. see if it is still the same. is it okay? >> imakes it appear that you have a frog in your throat. >> is there anything we can do? here, we have another person coming with another microphone. we have had this happen before. it is not you. pull it as close to you as you can. go ahead.
3:55 am
>> thank you again for inviting me here. >> it still doesn't work. >>we have to turn it on here, is what i am told. >> turn it on and off again. then turn it on. there you go. >> good? thank you. the controversy surrounding cambridge analytica and facebook have raised some serious concerns. the personal privacy of facebook users and cambridge analytica's voter targeting strategies, my expertise is on civic engagement and targeting and i hope to be up to answer questions for you. let me summarize some points. every election brings exaggerated claims about the
3:56 am
effects of the latest technologies. after an election, there is always a demand to figure out why the winning campaign one and the latest technology used by the winning campaign gives a good storyline even if it is false. campaigns also have a special interest in offering skills and they often impose their roles to the media. hemorrhage analycal was relatively new in 2016 and promoted to strategies to persuade voters. from everything i have seen publicly disclosed about cambridge analytica, i am skeptical of the idea that it's strategies of voter persuasion were unusually effective or contributed fully to the election outcome. understanqut useful to divide a campaign strategy to mobilization and persuasion. mobilization includes finding supporters and encouraging them to vote. persuasion entails encouraging voters to support your candidate. compared to mobilization, persuasion is very hard. in facebook data or data from
3:57 am
other sources doesn't help very much. persuasion affects decay rapidly. resear hasn at an ad may change a voters mind for a fleeting moment that the effect goes away amidst all of the other ads, news, posts and stimuli that accompany an election season. second, persuasion is hard because it is actually very hard even with top-notch data to figure out which voters are persuadable. the reason for this is, no person is persuadable all the time. persuade ability is not a stable disposition. you may be persuadable now but not tomorrow. persuasion is hard because when you try to figure out who is persuadable, your prescript -- predictions of who is persuadable is often important -- is often imperfect. campaigns often want to predict the racial identity of a voter.
3:58 am
predictions of which voters are black or hispanic are wrong about 25% of the time. when a campaign sent the message targeted to these motors, one quarter them will be missed targeted. research suggests that voters penalize candidates who missed target them. they will have much more error in estimating something nuanced like a personality, how extroverted or neurotic you are based on your facebook likes. that is what cambridge analytica claimed to be doing. the suggestion that facebook could encourage voters through these ways is unlikely. despite my skepticism, there is lot we still do not know about how campaigns are using social media to target voters.
3:59 am
this controversy gives us some anxiety in part because we don't really know where the line is between ads attempting to persuade voters and adds attempting to manipulate or deceive voters. this anxiety is all the more understandable because facebook hasn't really taken seriously its solemn civic role as a facilitator of news and political communication. in my written test money, i described the initiative currently under way in which independent researchers can measure the effect of facebook ads targeting and news sharing. the success of this program depends on the serious commitment by facebook to share its data, especially in cases that would bring negative press to the company. i do not need to suggest that there aren't serious concerns on this controversy. i hope we could focus on the more serious issues thank you
4:00 am
for your attention to this subject matter. >> how is that? mr. chairman, think you for the opportunity to invite me to speak today. cambridge analytica is the canary in the coal mine. we must address the digital echo chambers that are being exploited to segregate american society. online communities should unite us and not divide us. data is the new electricity of our digital economy. online platforms turns and conditions present users with a false choice because using the internet is no longer a choice. the american revolution
4:01 am
required the constitution to ensure that citizens of the young republic are protected from the -- the industrial revolution required protection for workers against hazardous conditions in the workplace and the environment. so too with the digital revolution we must realize there is a new game being played. the right to life, liberty, and the pursuit of happiness can be defended using new technology whether they are corporations, nations, or other. i have come here today voluntary as a witness at a whistleblower. i have already reported these matters to the authorities and they should be made clear that i am not a target of these investigations. i am the director of research for cambridge analytica from 2013-2014. there is a group that was a british military director. kemper jedlicka was created to
4:02 am
allow us to work in the united states. kemper jedlicka did not have any employed staff and all of its clients were handled by us. although lawyers warned about using foreign citizens, the firm installed a non-us citizen as its ceo and embedded non-us citizens at american campaigns. cambridge analytica offered voter disengagement services in the unittates and there are internal documes thati have seen that make reference to this tactic. my understanding is that it was targeted -- targeting african- american voters. i was made aware of the firms policies. i had seen policies where the firm sought to adhere -- some of these targets of these special intelligence services are currently heads of state in various countries. i have also seen internal documents that make rerence to the use of specialized intelligence services from former members of israeli and russian state security services.
4:03 am
a further concern is cambridge analytica's employment of people closely associated with wikileaks and julian assange. in addition to making approaches to wikileaks, the higher to staff, both who were --, tallica stop to identify mental vulnerabilities in voters and work to exploit them by targeting information designed to activate some of the worst characteristics in people such as neuroticism and racial biases. to be clear, the work of cambridge analytica is not equivalent to traditional marketing. cambridge analytica specialized in disinformation, spreading rumors, and propaganda. for those who claim that profiling does not work, it is proven in scientific journals. dr. alexander cogan was selected by cambridge analytica
4:04 am
to lead the harvesting operation and over 80 million suects had their data misappropriated. this could be one of the largest breaches of facebook's data. originally, contractors had also worked on pro-russian -- in eastern europe. cambridge analytica set up focus groups and polling on american views on the putif vladimin and nsiosm in eastern europe. dr. cogan was also working on russian state funded research project. the russian project in sync perch -- st. petersburg -- cambridge analytica -- we were also in close contact with executives at one of russia's largest oil companies. they presented them with
4:05 am
information outlining their experience with american data assets. facebook knew about cambridge analytica's scheme since 2015. bre, they threatened to sue ory the gudian and sue me. it reveals the unrestrained power technology companies can exercise over ordinary citizens when a person's entire online resins can be so quickly and so thoroughly eliminated from existence. there is no check on this power. it raises a serious question for republicans and democrats. what happened to our mocracy when these companies can delete people at well when they speak out? mark zuckerberg's continual refusal to cooperate with the british inquiry reveals the towns that her countries face to hold companies like facebook to account. the british parliament -- the
4:06 am
cambridge analytica scandal has exposed that social platforms are no longer safe for users. they are a critical part of american cyberspace and are in desperate need of protection and oversight. i am still optimistic about the future of technology but we should not walk into the future blind. it is the job of lawmaks to ensure that technology is for our cizens. thank you.
4:07 am
new regulations aimed at facebook's errors are more likely to affect the competition then consumers, let me summarize my thoughts on each, the most effective were the national campaign's of president barack obama, i will focus there, unique capable at leveraging facebook, facebook cofounder chris hughes put together the strategyfor president obama's campaign, taking things to a new level. the campaigns of teamed facebook date of supporters and friends through apps, and by having supporters log into my rock obama.com, and the -- my barack obama.com. one facebook outreach program let them know if a close friend had not voted, encouraging the
4:08 am
supporters to contact the friends, the intimate knowledge of the facebook friends, and the nature of their relationships. regarding facebook's failings, the primary failure is not being clear and candid with the users, i distinguish between the users, who are the subscribers, and the customers, where those entities that buy ads and other services, users should have complete and understandable information on the nature of the services they are using, even those that have zero monetary price, as in the case of facebook. this isn't happening, it is changing how it uses the subscriber thhave more of a connector of communities, investigating people's lives now, and faith test filtering there can medications. the pivotal moment appears to be in 2007, when their growth stalled, from their beginning,
4:09 am
adding users was a primary goal, the stall had them hire a team that attempted to use psychological manipulation to increase users time on facebook, since then, it seems fair to describe the company's business is thering people into a context, where they reveal information about themselves, so that others can market products and ideas, in a sense, the users are the products. one of facebook's methods for expanding its reach is the development of newsfeed, it changed facebook yet again, making it into a discussion monitor, determining who was a lot of ways, and who here's what voices on the platform. faapbook's sar to be the case of a company allowing itself to drift, each step over the years probably made sense bites of, but taken as a whole, they constitute a change in who the company is. many businesses have gone down this road, with things go too far, competition steps and.
4:10 am
-- steps in. new regulations are likely to harm facebook users, two reasons, the probmsare business problems, that commonly occur, the regulations are unlikely to help, ey be fed onthe main issue. second reason is that new regulations would likely serve to protect facebook and other companies from competition. they -- the european union's new regulation protection is a point, driving some tech companies, especially small ones out of europe, such regulations impede freedom and stifle innovation, an i will be glad to answer any questions. >> thank you to everybody for your testimony, we will go to questions, and we will have five-minute rounds for each individual. and we will start with mr. wiley.
4:11 am
i think these will be easy questions for you, but listen while i lead up to them. you joined cambridge analytica group of companies in june 2013, and in may 2014, the group received facebook data from alexander, in early 2016 they contacted the group and yourself, and that they asked that the data be deleted, in march 2017, following an internal audit, the group certified it had purged all of mr. kogan's facebook data from its servers, the group was retained by the trump campaign, in the summer of 2016. so these three questions that i think will be easy for you to answer, so scl, cambridge, had
4:12 am
not been retained by the trp campaign during your employment? >> the trump campaign was not a client, no.>> you did not work for the trump campaign wall at sco cambridge? >> i have never worked for the trump campaign. >> you weren't there when the company made its certification to facebook? >> i was not in the company or engaged at the company when they had that dealing with facebook, no. speak -- >> now doctor hirsch, media attention around allegations that cambridge analytica help president trump improperly influence the 2016 elation by using data it received from facebook, cambridge claims that it was ineffective, was deleted upon request from facebook, and was not used in their work for the trump campaign. if cambridge, had kept and
4:13 am
utilized the facebook data, what impact do you think the organizati le cambridge analytica, and strategies which mr. wiley described, as military style information warfare,", can have on influencing the outcome of the election, given the kinds of data is received from facebook? >> as i said my testimony, it is hard to move people, it is easier to mobilize or demobilize or persuade people, but there hasn't been any evidence that cambridge analytica, data that probably does exist, that could answer the question, in other words when they run ads, they do so -- they do so with experimentation, they could know the answer to this question, so if these things were actually effective, i think cambridge analytica and facebook would know.
4:14 am
in the history of targeting about why it is hard to move people, i'm skeptical of this data and these ads moving people in a substantial way.>> doctor jamison, the media has portrayed the trump campaign use of data and firms like cambridge analytica as nefarious actions to multiply the public, my question, are these strategies and use of data something new in the political world, and how about advertising generally? >> thank you mr. chairman, no, this is not new in the political world, it's been around for a long time, it feels new because it has never gotten into the public press before, at least not this volume, it is undstandable that people fe thesays violated a norm, but it has been a norm for a long time. >> i will reserve my time and go to senator feinstein. >> thank you, mr. chairman.
4:15 am
mr. wiley, in february 2018, special counselor robert roller indicted 13 russian nationals in three companies for their pa in a well-funded, coordinated campaign of information warfare, using social media. this information warfare campaign spearheaded through the russian backed internet research agency, started as as 2014, what can you tell us about possible connections between the sco group, or cambridge analytica, and russia? >> thank you for question. so one of my concerns is the level of engagement that the cover -- the company had, cambridge analytica, from lucoil, russia's second-largest
4:16 am
oil company, they sent documents to lucoil tt made reference to its experience and disinformation, reference to its experience in rumor campaigns, tudina inoculation. alexander nix emailed me to say that he passed on the white paper, to the ceo of mccoil -- of trenton, it discussed the data assets that the company had and the strategies twas employing, the lead researcher, that they used, doctor -- doctor kogan, at the university of st. petersburg, was working on the same kind of psychological warfare, they also engage contractors, who had previously worked in eastern europe for pro-russian
4:17 am
parties, and d, the company decided to test americans views t adership style of vladimir putin, and americans views on eastern european issues, russian expansionism, the only foreign leader that was contacted was letterman putin, and they focuseon russian expansionism in 2014, -- vladimir putin. there was a lot of contact with both russian companies and then also by dr. kogan's research, and i can't say that i had a connection to the internet research agency, but a lot of noise is being made to companies and individuals who
4:18 am
are connected to the russian government, and for me, that is a substantial concern. >> do you think it is possible or even likely that the facebook data harvested by cambridge analytica ended up in russia? >> what i can say is that the lead researcher, dr. kogan, who is managing the facebook harvesting project for cambridge analytica, was at the time working on projects that related to psychological profiling in russia with a russian team as at wasgoing on, i know that he was traveling to russia, and based on conversations that i had with him at the time, he was making it known to his colleagues in russia about the project. and so i can't say definitively one way or the other if these data sets did end up in russia, but i can say that it would
4:19 am
have been very easy to facilitate that. >> he told the uk house of commons that cambridge analytica pitched the russian oil company , lucoil, in its services, and you also said that alexander nix gave them a white paper that you wrote that explained their data collection and line targeting of americans, and lukoil, is said to be connected to the russian kgb, formally so, were released taking place? >> in london,united kingdom, and the phone. >> what did cambridge analytica tell lukoil about its data on americans? did it share any of the data, or is it possible that russia acquired any of its data on americans?>> in sending the
4:20 am
white paper, and discussions that i had with alexander nix about what he was speaking to the company about, i know that the scale the data, the location of the data, was made known, and also, that dr. kogan was involved in that data collection, project, and the concern that i have is that if you were intending on acquiring the data, even if you are not intending to acquire it with the willing participation of cambridge analytica, what was made known was that this data could have easily been acquired by something as easy as a keylogger on his computer, when he was in russia. senator kennedy? >> dr. hirsch, i don't really have a question, but i will come back to you five-time, if what you're saying is that people in america are not
4:21 am
persuadable or persuaded by advertising, i think that is rubbish. i think some really smart people spend $200 billion, $206 billion last year on advertising. and i hear kids all the time while grown, saying dilly dilly , they didn't just bring that up, okay? but mr. wiley, i'm not interested in innuendo, speculation or rumor, other than facebook, list for me the sources of all of the data that you know caridge analytica used while you work for them?>> there were several different consumer data vendors that were used. >> can you name them please? i would like to be specific. i believe experian data was
4:22 am
used, i believe axiom data was used, a mature. >> who else? >> state voter rolls and registration. smaller firms which i can give you of the top of my head, who had more specialized niche data, and in terms of online data, their work spare was done on collecting other social media data, like twitter, that was modeling that was being ne at the hat i was there, was primarily using facebook data.>> did cambridge analytica -- put facebook aside for sick, the justice department and fbi will get to the bottom of that, did cambridge analytica obtain
4:23 am
any of this information unlawfully while you were there?>> i am not a lawyer and the nights discuss i couldn't comment on whether it was lawful or unlawful. >> well,d they do it properly? you have been making normative judgments quite often, don't get religion now on me.>> the facebook data, i believe, was -- >> other than facebook? >> other than facebook, the consumer data lists that were used were acquired via contracts that were siednd paid for. >> did cambridge analytica hack anybody? while you were there, that you know of? >> i have seen documents that make reference to special intelligence services, and information gathering networks.>> do you have copies of those documents? >> did cambridge analytica get
4:24 am
money or information from wikileaks? >> not while i was there, no. >> do you know if they got it after you were there? >> in terms of money? >> not money, i misspoke, data, i'm still worked up. >> i wasn't there when the request was made. >> excuse me for interrupting, i'm not trying to be rude, we have a limited amount of time, while you were at cambridge analytica, give me the names of all its clients. issue or candidate?>> at risk of misspeaking, i am happy to give you a complete list of the clients that were being used at the time. >> who are they? >> well, i know there were pacs
4:25 am
of various candidates. >> what issues and candidates? >> i believe some were primarily financed by robert mercer. >> who el >> several senatorial and congressional candidates, i believe. >> any issues? >> issue campaigns, john bolton's packan issubase campaign -- pac. >> did cambridge analytica work for russia? >> we did not have a russian client while i was there. >> i yield back my nine seconds. thank you, mr. wiley. this has been fascinating, and i am looking back, to early 2014, under the leadership of steve bannon, they reportedly
4:26 am
started testing slogans like build the wall, deep state, drain the swamp, also message testing images of policies of russian president vladimir putin. later on, build the wall, deep state, drain the swamp, in the trump campaign, but they were testing these before there was a trump campaign, is that correct?>> yes, so the company was testing, as you said, slogans, like drain the swamp, images of walls, paranoia about the deep state, in 2014, before the trump campaign was in existence. >> what did they learn from this testing? >> they learned that there were segments, of the population, that responded to messages like
4:27 am
drain the swamp. or images of the deep state, that were not necessarily always reflected in mainstream polling or political discourse that steve manning was interested in using to build his movement.>> you noted that cambridge analytica that they didn't have any employed staff, it was a front group for the sel group, and rebecca mercer, steve bannon, others on the board of directors, their client work, was handled by scl, am i correct about that?>> yes. now our laws prohibit non- americans from using -- from working on you is campaigns, to protect our elections from foreign interference. a 2014 memo from cambridge
4:28 am
analytica, outside legal counsel, rebecca mercer, steve bannon, made these restrictions very clear. -- u.s. campaigns. did cambridge analytica follow these legal requirements in its woonamerican campaigns quick >> the source that provided the media with the memo, which i saw the tail end of my engagement with cambridge analytica, to my understanding, that memo was disregarded, because alexander nix continued to be the ceo, and they continue to send people to the united states who are not american citizens, one thing that i would say is that many people who were sent to united states who were not american citizens, were not privy to the memo, and were not aware that potentially there were violations of u.s. law. >> facebook of course has been discussed as being used in election advertising for some time.
4:29 am
somebody lives in a particular state, they could target ads that reflect the interests of that state. for cambridge analytica, correct me if i'm wrong, but they obtain the unauthorized facebook data of 87 million people, and then targeted them with manipulative disinformation, is that a correct statement? >> that was not everything they did, but yes, that is something that they did do, yes. >> how does traditional online marketing differ from how they obtained individuals' information how they used to?>> so when you are looking at traditional marketing, first of all, doesn't misappropriate tens of millions of people's data, if they are performing
4:30 am
illegals judy -- legally. it should not be targeted at people's mental vulnerabilities, such as neuroticism or paranoia, or racial biases, traditional marketing does not exacerbate people's innate prejudices, and coerce them, and make them believe things that are not necessarily true.>> but you also say that they were a full- service propaganda machine, discouraging people who are more prone for voting for democratic or liberal candidates, and some of their u.s. clients prevailed in extremely close races, why did cambridge analytica american clients and investors like the robert mercer family, investing $20 million, believe these tactics might be helpful? >> steve bannon is a follower
4:31 am
of something called the breitbart doctrine, which posits that politics is downstream from culture, so if you want to have any lasting or enduring chans politi, you have to focus on the culture. and when steve bannon uses the term culture war, he uses that term pointedly, and they were seeking out companies that could build an arsenal of informational weapons to fight that war, which is why they went to a british military contractor that specialized in military operations. >> thank you, the russian state security services, i have these questions. >> the record stays open for a week, and that if you get written questions for people who are here, like senator leahy, we would appreciate if you get them back to us as soon
4:32 am
as you can.>> mr. chairman. i would like to start with you that's okay. you said that they obtained without authorization, for voter tag ef atthe same time you took that same data with you upon leaving the country -- the company, is that right?>> because there was no staff it cambridge analytica, most people were contractors, or indeed had companies, so i received a copy of the data. >> so you had that data, then you started your own company? >> no, my company was in existence before i left cambridge analytica. because i was a contractor. >> but after leaving the company, you had a meeting, or perhaps a series of meetings, with a major campaign, to discuss some micro-targeting techniques. insured, -- in short-
4:33 am
>> i did not meet with another campaign inerenceto that data set. >> it appears that you try to use some of the same market as your former company, use that data for something right? you are not just going to leave it idle?>> to be clear, the data was never used on any commercial -- >> why did you take it with you?>> is not that i took it with me, it was just in existence at the time that i left. to be clear, i didn't take any data from cambridge analytica. >> you didn't take it in the sense that it was already with you?>> yes. >> what type of work that -- did you anticipate that you would perform after you left cambridge analytica? >> so i work mostly in data analytics, looking at different kinds of social trends, so after i left cambridge
4:34 am
analytica, i continued working on independent projects, but to be clear, i did not use that data on any commercial contract. >> couldn't that data have proven useful in some of your work? >> it could have, but i did not go on and sell it. >> but you could have, and had you been successful, and had you gone on in your business be successful, could you not have been at the receiving end of some of the questions going to cambridge analytica? >> yes, but it did not happen. >> understood. in paragraph 15, you say, when i was at scl, i was made aware of their black ops capacity, which i understood to include using hackers to break into computer systems to acquire comp or matt -- compromat, how
4:35 am
did you learn about these black ops pa? >> alexander nix told me. >> and who as a practical matter was involved in these black ops? >> my understanding was that in different parts of the world, misappropriated information was used as kompromat in elections in particular, against opposition candidates. >> in paragraph 30, you say quote, the russian project undertaken had a particular focus onthe dark traits of narcissism and give hiand psychopathic, they also conducted behavior research on
4:36 am
online trolling, how did you learn all of this, and how did -- can you describe these projects in more detail?>> at the time that dr. kogan told me about the research he was doing, he told the company about some of the research that was being done by the team in russia. this email correspondence from the firm, for example, it referenced the work that he was doing for the russians, so initiay through conversations that i had through dr. kogan, and then later through the investigative reporting, being done through the year from the guardian and others, more details have emerged.>> and dr. hirsch, the use of social media to micro-target is fairly recent, and my understanding is that micro-targeting is not,
4:37 am
and the use of provided information to mobilize portions of electorate has a long history of political campaigns, so is the use of social media to micro-target different than what it has been done, and how it has been done in the past? >> we don't know about its effectiveness in some ways, but often looks the same, and responding to senator kennedy, just because campaign spent a lot of money on the particular ad kind doesn't mean it works, a lot of times they spent money on robocalls, countless experiments show that they do nothing, in an environment where there's a lot of stimuli, a lot of things going on, a lot of campaign ads don't work, i believe that nobody in this room or nobody that people don't change their mind, for someone who is a director of research for cambridge analytica, they should know, i would be shocked if they didn't know, the actual affected estimate of which campaign ads to what.
4:38 am
part of my skepticism comes from the fact that given that there has been whistleblowing, any such evidence, an randomized controlled ennments, in which someone shows, here's the effect, this ad targeting neuroticism, or whatever, has effects, it is hard to sort this out as technology changes from 2000, 2004, 2008, 2060, lots of things are new, one point we come back to, in a presidential election particularlyso much going on, the effect of one ad, one kind of ad, or robocall is usually zero.>> hello mr. wiley, you said that cambridge analytica and sco group are effectively the same thing, and that cambridge analytica was effectively the front facing operation for scl?>> yes. >> what is scl elections? >> there is a group company or was, called sco group, which had several different divisions, the largest one when
4:39 am
i joined was defense, scl elections, commercial, and they had different markets for the company, so scl elections handled political. >> what are scl canada and aggregate iq? >> subcontractor set up during the time i was there to build out a software infrastructure, they played a significant role in building the actual infrastructure, the software product. >> are they essentially the same, scl canada d iq? >> they are a franchise. >> using the facebook data, was the same software? >> yes, once you have the same algorithms any target, a set of targets, you need something to connect those targets with an online display network, that is
4:40 am
part of the that ripon played. >> you said it became a company to service cambridge analytica, correct? >> it became a company, as i understand it, so that he could sign a contract with cambridge analytica, but technically scl, yes. >> is it fair to talk to them as a coordinated network? >> yes. >> what was the role of robert mercer and funding that network? >> he was the primary funder that put in tens of millions of u.s. dollars into cambridge analytica, which then distributed the money to that network. >> did that, including scl, did they have a contracting relationship with blk cube? >> when i was there that we did
4:41 am
not have a contract. >> heavy since become any cash become aware of a connection between them in worki projects?>> i have become aware of relationships that the company had with former members of israeli security services. >> how about asi data science, cto, worked on scl projects, a subcontractor, and a revolving cast of data scientist, between the organization and cambridge analytica? >> yes. >> they are a frequent contractor? >> i believe there role was as a supplier, as the company was growing, an increased demand for more more data scientist, and they were, i believe, the contract, was to provide data scientist.>> a part of the
4:42 am
cambridge analytica network that we describe! >> if your definition of that work is anyone who has an ongoing relationship, then sure. >> you said that the company pallanteer, helping to build the models for the ripon program? is that correct? >> yes but to be clear, but the work being done by their staff was being done in a personal capacity.>> i've got 1 minute left, let's look at brexit, some of the forces were -- vote leave, be leave, and veterans for britain, all of them had contracts with aggregate iq? >> yes. >> is there anything peculiar with those contracts? >> it is highly suggestive of coordination and data sharing, but that is currently being
4:43 am
investigated by the electoral commission. >> aggregate iq would be hard to find, it did have a website?>> no. >> and vote leave actually funneled money into the other two, that what you acrid iq -- aggregate iq? >> between you kip, and eldon -- ukip, and eldon, yes? >> that side of the leave campaign engaged with cambridge analytica. >> my time has expired, if there is a second round, i would love to have one. senator cornyn, and you take over at the end of his five minutes. >> mr. wiley, let me start with you, the sort of data mining that you have been describing and targeting, of messages, has
4:44 am
multiple applications, correct? it could be a commercial application, you buy something on amazon, were a movie on netflix, they can send me information about something else i might like, it could be used to persuade people in a political campaign for or against a candidate or for or against a particular issue, and also used for covert information operations by governments, correct? >> all of that is correct, yes. data is like any kindf tool, you can use it for various means, beneficial legal, and others not.>> did scl or cambridge analytica servo covers? were you open to business for anybody who wants to purchase the services? >> that was the impression that i got, alexander dix was quite keen on -- alexander nix was
4:45 am
quite keen on selling contracts, but after he put the investment into cambridge analytica, the only restriction that we had was not to work with democrats.>> for exame, fusion gps is much of the news, they provided opposition re to the dnc, against the trip campaign, the data itself, and the means by which you analyze it and use it, is pretty much agnostic. you can use it for or against a political candidate, a product, or for an information campaign, like we said? >> yes.>> so mark zuckerberg, when he was here the other day, he kept saying, we don't sell data. and i responded to him, will you clearly rented. i don't know whether that is a fair characterization or not, how would you characterize the
4:46 am
social media platforms like facebook's use of personal data, they say they don't sell it, how would you characterize it? >> they have created a platform that encourages the use of data, so it is true that you can go to facebook and simply by facebook's data, that make it readily avlae to its customers via its network of applications, or the fact that the layouts of's profiln ok makes it very conducive to scraping data, for example. although facebook say they don't allow that, they still cr a setup which catalyzes its misuse, in my view. >> mr. zuckerberg also said that the terms of service that people agreed to with a sign on
4:47 am
to facebook, you said people probably don't read it, or if they do, they don't really understand it, is it your impression that most of the public has no idea about what they are sharing with these companies?>> when you even talk to lawyers who read to the terms and conditions, some of it is even dense for a lawyer. so i think that it is unreasonable to expect a regular ordinary person to have the burden of understanding dense legal text, and the other thing i would say is that social media is not really a choice for most people, the internet is not a choice for most people, it is difficult to be a functioning member of the workforce or society and refused to use the internet. i don't know a job that would go in and refused to use google, for example. most hiring requires a linkedin profile now, so although we use this narrative of choice, because someone is pushing a
4:48 am
button, even if they had read the terms and conditions and understood it, they substantially dohave a choice, because the modern workforce, you have to use the internet. >> they have other options? >> i don't know a job that would hireeone whrefused to use the internet. >> to think it is too much to get the users and their informed consent? if you're going to consent to a surgical procedure, by your doctor, that your consent must be informed, in other words you have to ask what your green to. do think that it is too much? >> -- what you are agreeing to.>> people should have informed consent, but the analogy is not equivalent, when you go see a doctor and you need surgery, you need something, you consenting to that is proportionate to the
4:49 am
benefit you're getting caught when someone consents, quote unquote, to something online, even if they understand that they are consenting to their data harvested, if that is the only way that you get a job, it is not really genuinely a fair situation. the point that would make to you, is that we should take a step back from this narrative of consent, anstart to look at the fact that people don't have a choice, they have to use these platforms to foot -- to be functional. >> i can use twitter, i can use -- >> but all these platforms to the same thing, as soon as you sign up to twitter, as soon as you some of the google, as soon as you sign up to anything, they will all be collecting a hugely disproportionate amount of data compared to the utility that they provide to you. and third, the other thing i would say is that when you're looking at technology, there is a question of not just informed consent in the present, but reasonable expectations the future, when i first signed up
4:50 am
to facebook, it did have facial recognition, i posted my photos, then they have algothms that can search the internet and find other things that i'm doing. it is not a question of just informed consent, it is about isoportionate to the benefit that the consumer is getting, is it reasonably expected in the future, with technology, that the consumer did actually consent to that at the time, and more broadly, this narrative of consent is slightly problematic in the sense that when people have to use these platforms, it doesn't matter whether or not they understand, if they have to use to get a job, eywill still use it. we are coursing in compelling people to hand over a lot of information which not only now could be dangerous, but you have to mention the future, what the developments of the technology are, and will be moving forward, and what kinds of risk we will be exposing to people 10 years from now, when the data still exists. >> i would conclude by saying that companies now and can
4:51 am
market their services and products based upon people's expectations, of privacy, and so if consumer -- if consumers are fully informed about what they're doing, and what the consequences are, they can ke their chces, that might create markets for alternative platforms were data will be more protected. thank yovery much mr. chairman, and thank you for your assisting dancers, mr. wiley, we introduced a bill t protect consumers privacy online, and include in that, giving consumers the right to control their own data by allowing people to opt out of hadata collected, and requiring companies to notify consumers of a privacy violation within 72 hours, you have previously expressed support to allow users to opt
4:52 am
out of targeting criteria in three clicks or less, and rules that the data collected, be proportional to the apps actual purpose. would cambridge analytica have been able to harvest that data, if they had opted out of that data be tracked by facebook? >> the problem was that facebook had set up applications that physically allow the collection of data, one of the things i would say is that in addition to giving consumers rights, we should be putting obligations on companies themselves. a principle of privacy by design, as an engineering problem, would be incredibly beneficial, because privacy is not just a governance issue or a terms or policy issue, it when it comes to ware. >> do think they would do that on their own?
4:53 am
>> no. >> that is my response to one of my friends on the other side, and why we have proposed this bill. secondly, the honest ads act, with senator mccain and senator warner, i understand you supported the idea that there should be more transparency of political ads? >> absolutely. >> no requirements in place, he stated that if a foreign actor dropped propaganda leaflets over florida or michigan, that would universally be condemned as a hostile act, but this is happening online. facebook, twitter and microsoft are supporting my bill, taking measures to disclose these ads, especially facebook, do you think that this patchwork of voluntary measures will be the answer?>> no, just the same way that we require safety standards in everything we care
4:54 am
about, we should be requiring them and transparency in software and online platforms. >> i asked misters zuckerberg of facebook had determine whether the 87 million facebook users whose data was shared with cambridge analytica were concentratedn certain states. misters zuckerberg said that he would follow-up with a state-by- state breakdown, i want to ask if you have any knowledge as to whether those facebook users were mostly in any particular state? >> i can say off the top of my head what the density was each day, but i do know that there were particular focused -- particularly focused on states from the company's activities on swing states, and ates that were winnable by republicans. >> states likes -- states like wisconsin, michigan, and pennsylvania? i await that information from facebook. i also asked misters zuckerberg, were any of the roughly 126 million people who might've been shown content for me facebook page associated with the internet research
4:55 am
agency, the same users whose data was shared with cambridge analytica, he believed it is entirely possible that there will be a connection. what can you say about the potential for any overlap between the facebook users whose data cambridge analytica obtained, and the user shown content from the internet research agency? >> the thing i would say is, firstly, i did not deal with the internet research agency, so i can't comment specifically on that entity. but to your point, my concern is that information either may have been shared, or indeed misappropriated, as a second instance, by russia through cambridge analytica, what i would say is that it is not just whether or not these individual records were then targeted, because if they were used to build algorithm, whether it was built at cambridge analytica or another
4:56 am
entity, other users who share similar profiles and patterns in the data, could have also been exposed in a way that if you simply looked at were these records targeted specifically, even if the answer is no, it doesn't mean that they weren't used to build a targeting algorithm. >> i get it, i will follow-up in the second round. one last question, use the word profiling, voter disengagement, suppression, one of the most horrifying things that i saw from these hearings, with these ads that were clearly made to suppress the vote of afric americans, do you have any knowledge aboue scope of this activity and how often it occurred? >> my knowledge result -- refers to the tail end of my engagement with cambridge analytica, one of the things that did provoke me to leave was the beginnings of discussions about voter disengagement, i have seen
4:57 am
documents that reference voter disengagement, and i recall conversations that it was intended to focus on african- american voters. >> can we get those documents? >> i will discuss with my lawyers the best way to get that informaon >> just tobelear, just as you said it would be illegal and having an airplane come in with those pamphlets, under american law, where you are specifically suppressing votes is also an illegal act. thank you. >> to be clear, i did not partake myself in that. >> understand, that is why you left in part, and i appreciate you coming to testify very much today, think you.>> the chair has asked me to chair the committee, and i'm in the next order, it is my turn, thank you for being here. this is a very helpful hearing, i intend to be here for the second round as well, and mr.
4:58 am
wiley, i just leaned over to the ranking member, she was complementing you on your technical expertise, i love having a witness i can understand, i can understand lawyers, but i n understand data analytics people, because i was one. i hope that the result of this hearing is ashley try to figure out what if anything congress should do from a regulatory framework and compliance framework with respect to the new avenue for data use. all of us in these committees have already used data from aggregators, we take the voter data, we know what propensities are, use that as a basis for targeting voters, and then data aggregators god so that you can overlay people's affiliations with associations, the magazine subscriptions, that has all become passi in terms of data bashing, having for 10 or 20 years, in our campaigns, and i
4:59 am
would dare say that every single member here who has run for election as had that. now that is where they by data, they aggregated, and then they build their proprietary platforms around it. now with the advent of social media, we have entities who have come into play that really don't want to sell their data, they want to sell the analytic result of the data so that they can target people in certain social media platforms, would you agree with that analysis? >> is the age of axis rather than the age of transfer, in terms of data. >> to believe that data, there baseline, part of their intellectual capital, a part of their institutional value? >> yes, data is incredibly powerful, it is like the new oil. >> i would be happy for either of the other two witnesses to chime in, a part of one mr. zuckerberg was here a few weeks
5:00 am
back, i try to focus on what we should all be looking at as policymakers, as other alytica, and i should say that my firm engaged cambridge analytica, i met them the day that they proposed it, and i saw the begin on election day, i do have some questions about the focus on trying to get psychographic data from people through in-person interviews, it may or may not have been captured through some of the techniques we are discussing today, but i want to go back and talk about, if we are going to do a thorough, impartial, nonpartisan review of the facts, that we really should go back over the last 10 years, i don't know if you've had an opportunity to, but i would commend it, to read in the mit technology review, a three-part series, it was titled how obama's team used the data to rally voters. and if you go through that, very well written review, and
5:01 am
through articles associated with it, there were quotes from campaign workers who said, we literally took the whole social graph from peyton book -- for facebook, and able to download it through the use of an application, there is a term, they talk about gameifying the apps that they would use, and the alabama -- the obama campaign, they asked if you are a facebook user, if you wouldn't mind, clicking on a button, be lling to sha information about all of your friends. in my case, i have about 4900 friends, by clicking that but, i was actually giving axis to thousands of people's information without their knowledge. and this was a documented practice that i think also has to be looked at in the context of creating good policy that is
5:02 am
based on various uses, contemporary uses of data, social media platforms. in my maining time, the first run, i would just ask you, do you believe that some of the technology players today, simply have grown so large so quickly, that some of what you thought would've been captured through just good corporate governance and code of conduct, having somebody from a social media platform who leads conservative orleans liberal, kind utting their thumb on the scale, and giving people information, that they should if they were good stewards of the data in the social media platform that they were on, i will leave that as an open question for anyone of the panel. >> i would just say that i don't hold a view that facebook is somehow necessary for allies,
5:03 am
as an instructor of political science, i encourage my students to never share information about politics, ads or news online, because they are not professional editors, they should ct r atteion to new sources that i've editors, the idea that best -- that you must contact your friends and family and upload pictures, and that there is an option, it doesn't strike me as something that is true. facebook has a lot of data, it has acted really inappropriately in terms of how it has conveyed news and sold ads to pretty much who was to coffee -- with hate groups, it is not our fault, it is a terrible way that they have conducted busine and again, i would encourage people to use it, but i am confused about people's continued interest in the company. [ laughter ] >> i concur a lot and what you saying, it is difficult to
5:04 am
argue that facebook per se is necessary for people's lives, with about half of the internet users are donna, and lots of other types of social media. to your point about would someone with a particular ideological view tilt the scales one way or another? i don't know if people would delibera, ey might, t but the nature of it, they will, what i view as true is influenced by how i think ideologically, things that i might screen out, and say this is bad, this is disruptive to the community, would be influenced by how i think. >> thank you for the chance to be with you, the reason this matters is because we are talking about the intersection of several important developments that are difficult for the average american to understand, big data and social media. and for interference the 2015 election, trying to tease out what actually happened or
5:05 am
didn't happen. mr. nine, the mr. with you, michael kaczynski said on the average of the facebook like spy user, what you liked or disliked, you could predict sexual orientation, political party affiliation, race, with 85% accuracy. further factors like religious affiliation, alcohol or drug use, even whether your parents were divorced, could be deduced. is that your understanding of that analysis, and you think that cambridge analytica and the work of professor dr. kogan, used it to weapon eyes differences between americans for electoral gain?>> -- in order to weaponize it.
5:06 am
>> so that work that we were doing, the firm replicated his approach, they sought to improve it. >> d did i correctly summarize the incredibly high correlation that you could show in terms of really knowing the individual facebook user, if you had access to their likes and their friends, and their social media activity? >> yes, the particular paper that you are citing, the number of likes once you pass 100 or 200, you can get the same level of accuracy at predicting for example personality traits, as your spouse would, if they were asked to -- answering questions about you.>> to a point that professor hirsch made earlier, all of us that it stood for election have struggled with the difficulty of targeting voters effectively, the data says we provide -- we have had access to, the available, thin,
5:07 am
narrow, very little we have. the data sets that facebook has access to, that is why it is a multibillion-dollar company, unbelievably deep and rich, unlike anything we have had to confront before? >> yes, i don't contest that presr rsch was saying, persuading 70, compared to motivating them to turn out, much more difficult, the data sets that traditionally were used, are often very sparse and not the surly reliable, but -- >> but the ones available now are orders of magnitude larger. >> that is why cambridge analytica ended up pursuing that path, because it found that in comparison to traditional marketing data sets, the data that you could procure from social networking sites was much more dense and actually much more reliable, to create a precise algorithm.>> a
5:08 am
billionaire american supporter of trump funded a shell corporation, cambridge analytica, still run by foreign nationals, to take advantage of this research, this act -- this understanding, and gained access to 87 million americans faceok information and developed some of the algorithms that came out of that, i want to ask you, about your experience working with steve bannon, one of president trump senior campaign advisors, and the gos that he used cambridge analytica to achieve, was one of the goals to suppress voting or discourage certain individuals from voting? >> that was my understanding, yes. >> was voter suppression a service that people could order?>> yes. >> so steve bannon is running an organization where you could as a client request and contract
5:09 am
voter suppression using this remarkable data set? >> i don't know if it was referenced in contracts, but i have seen documents that make reference to it in relation to clrvices. >> testifd thatck in 2014, cambridge analytica said of focus groups, message testing and pulling, on americans views on the leadership of vladimir putin, and russians expressionism and eastern europe. you've also testified that it is entirely possible, relatively easy, for russian intelligence to put a keylogger on professor kogan's computer and get access to this full data set. why do you think cambridge analytica was testing president imirputin's policies, and all the algorithms, if they are all currently in the hands of russian intelligence? >> i don't have a clear answer as to why the company was so engaged in testing russian
5:10 am
expansionism and the leadership vladimir putin, that is a question better put to steve bannon. other democracies around the ot world, data is powerful, if put into the wrong hands, becomes a weapon. we have to understand that companies like facebook, an atforms like facebook or twitter, a not just social networking sites, they are opportunities for information warfare, not just by state actors, but nonstate actors, we really do have to look at detecting cyberspace, as a national security issue,ju in the same way we have agencies to protect our borders, land, sea and air.>> thank you very much. thank you, welcome to each of the witnesses, thank you for being here. i think americans are rightly concerned about privacy and the
5:11 am
security of our data. we are also concerned about the power that is being collected in silicon vall, a fuof companies hang enormous troves of data that they are -- they are able to use and control without much attention. much of the media attention has focused on the data operation of the trump campaign, but the trump campaign was hardly the first, to employ data in a very significant way in a political campaign. doctor jamison, you talked about the obama campaign, in 2008 and 2012, and their data operations, can you share with this committee with the obama campaign did regarding data in 2008 2012? >> sure, during t 2008 campaign, what they did is that they had an advisor, a cofounder of facebook, who help them understand how to use it
5:12 am
and obtain data from facebook, and a lot of that work was done by consultants, but that was the center of it. in 2012, they change the strategy, they ed it in- house, making it more effective, combining facebook data, more cleverly, more carefully with other sources of data, and do a much better job of understanding what individual voters are like, who was connecting with whom, and how they could understand his conversations. >> has facebook approached access to data on a fair and evenhanded matter, allowing candidates from whichever party the same access to data, or have they been more political and partisan players that regard? >> i have no knowledge of that. >> i will note that carol davidson, the director of data integration and media analytics for the obama for america 2012, she said quote, facebook was
5:13 am
surprised that we were able to suck out the whole social graph, but they didn't stop us, once they realized that was what we were doing. she also said that facebook quote, came to office in the days following election recruiting, and they were very candid that they allowed us to do things that they would not have allowed someone else to do, because they were on our side. so that is the head of data analytics for the obama campaign saying facebook was giving preferential treatment to the obama campaign, because that is the political side thereon. did they give the romney campaign the same access to data that lection in 2012? -- in that election in 2012? >> not to my knowledge. >> there's been a lot of talk about what they did with data,
5:14 am
did the hillary clinton campaign have a operation quick >> i assume so, but there was not very much written about her campaign on any of this. >> does anyone the panel know what the hillary clinton campaign did on the data side? does anyone of the panel think that there is a chance in 1 million years that the hillary clinton campaign did not have a substantial investment in dental analytics -- data analytics quick >> so campaigns across the spectrum in the right states use data, and just to your point, when we look at -- a lot of people are concerned about for example, the ability of big government to inhibit our choice, and actions, limiting our choice, data can do that, it is not a partisan issue. to your point about other
5:15 am
parties using types of axes, on facebook, i actually agree with you in the sense that there is a substantial risk of distorting the electoral process of a company like facebook decides to pick a side, whether that is democrat or republican, so the thing that i would hope that this committee and others really internalize that we are talking about cambridge analytica, but it is not a partisan issue, we are talking about the future of how these technology companies operate, and the risks to ordinary american citizens, and the risks to our democratic prses,here in the united states and around the world. that is not a partisan issu >> i would very much agree with that, i would note that there is an overlay that facebook and otr social media companies are now the vehicle through which some 70% of americans get
5:16 am
their political views, and so the specter of censorship, i think, is a profound threat, to liberty, so i appreciate this panel for being here. senator blumenthal? >> thank you, welcome to you all and thank you for being here. mr. wiley, are you aware of conversations between cambridge analytica executives and any representatives of the russian government or people associate with the russian government?>> i amaware of meetings the company had ith lu, has close connections with the russian government. >> are those conversations documented anywhere in letters, emails, or any other kinds of evidence? >> there are documents
5:17 am
pertaining to the conversations, and presentations made to lukoil, and i have passed some of those documents on to the authorities.>> to this committee? >> not yet tois committee, no. >> would you be willing to provide them to this committee?>> i will consult my lawyers as to the best way to get you that information. >> thank you, my understanding is that you are aware of the founders of funders, including robert mercer, providing monies so that they would not be quote, necessarily considered the clear will campaign contributions, are you aware of conversations to that effect by mr. mercer or anybody else? >> what i'm referencing is what
5:18 am
was explained to me, after i inquired of to the relatively convoluted setup in the united states, with respect to the setup of cambridge analytica, this is not the primary reason, it could be an ancillary benefit, but when you invest money as an investor to a company that you own, it doesn't necessarily constitute an electoral contribution which is declare bowl.>> me just cut right to the intent of my question. was there specific explanation to you that the purpose of structuring these funds as investments was to, in effect, avoid the reporting requirements or other provisions of united states election laws? >> it was explained to me as a benefit of the setup. >> were there any firewalls
5:19 am
during the time you worked at cambridge analytica that separated work on different campaigns, or were the funds in effectmingle in all the campaigns? >> whilst i was there, i did not see firewalls being set up, or any sort of barriers between people, or conversations, if 'squestion. >> so there was no recognition of the legal responsibility to separate campaigns? >> i am aware of memos, from the company's lawyers to some of the executives of the company that outlined the responsibilities in the united states to separate contact between staff and activities that relate to different campaigns or pacs, etc., those
5:20 am
were instructions or advice to executives, but when i was there, i did not see that. >> in practice, they were not followed? >> no.>> can you provide some specific examples of your direct knowledge of either focus groups, or other efforts to suppress voting? >> in terms of specifics, i'm happy to work with the committee to give them a more full expedition, there are limits on time, but i am aware of research that was being looked at about what motivates and indeed what do you motivates different kinds of people. so you focus on messaging that demotivate certain types of people, you decrease the amount of turnout. >> thank you. doctor jamison, you served as part of the agency landing team
5:21 am
in the transition for the federal communications commission for president-elect trump, yes? >> yes. >> during that time, did you have any contact with eichel: about fcc policy? >> did you ever meet him?>> not to my knowledge. no. senator harris. >> thank you. so we are here today to talk about how cambridge analytica obtain sensitive data about millions of americans from facebook without their knowledge or consent and then use that to target voters and influence our relations. and there are broader issues of privacy that are highlighted by this incident, and i think it is worth stepping back to pull all of this into context for the american public. to put it plainly, most americans have entered into a bargain with facebook and other web service providers in which users unknowingly give those companies huge amounts of
5:22 am
personal data in exchange for the free service of social networking. intern, facebook uses this data to show its users carefully targeted ads which are the source of 90% of facebook's revenue. interestingly enough, this business model makes the facebook user the product, and makes the advertisers the customer. but let's be clear. it is not always working in the best interest of the american people. first, users have little to no idea just how facebook collects their data, including tracking the users location, the device they are using, their ip address, and activities another websites. to be clear, this occurs whether or not they are logged into facebook. and whether or not they even use facebook. let me put this in perspective. in the real world, this would be like someone following you every single day as you walk
5:23 am
down the street, watching what you do, where you go, for how long, and with whom you are with. for most people, it would feel like an invasion of privacy. and they would call the police. and yes, social network sites technically lay this out in their terms of service. but let's be honest, few americans can decipher or understand what this contract second, as this hearing illustrates, americans do not have real control over the data collected on them. and there is almost nothing that users can do once data is shared with third parties. i believe it is therefore government's responsibility to help create rules that yield a better and more fair bargain for the american people, one that respects their rights as consumers, and their privacy.
5:24 am
in the meantime, i have a few questions for the witnesses. mr. wiley, you mentioned a discussion about how cambridge analytica discouraged african- american voters from participating in elections when steve bannon was the vice president, what specifically did steve bannon decides motivates or d motivates african-americans to vote? >> it is not just focusing on racial characteristics of people, when you pull a random sample of afghan americans, they are not a the same people, very different, different lives and motivators. when you are looking at any program, boat -- motivating or not, understanding their internal characteristics is a powerful thing, because you don't treat them just as a black person can't you treat them as who they are. that can be used to encourage people to vote or discourage people to vote. >> so how specifically then did
5:25 am
they target african-american voters, understanding is you do that the african-american population is not a monolith, how did they decipher who was african-american so that they would target their intent to suppress the vote? >> racial characteristics can be modeled, and i am not sure about the studies that my colleague here was referencing, but we were able to get an auc score, measuring accuracy for race, that was pointed 89. >> area under the characteristic, a way of measuring precision. it was.89. very high. -- 0.89. to be clear, i did not participate on any voter suppression programs, so i can't comment on the specifics of those programs, i can
5:26 am
comment on their existence and my understanding generally of what they were doing. but those questions are better placed for steve .>> and i am joining the request for any doctor -- documents to evidence what you have described. what has facebook done to verify that either you or cambridge analytica had deleted the unauthorized data? >> i can't speak for cambridge analytica, i can speak for myself. in 2016, they sent me a letter that said we are aware that you may still have data from this hostinpr it informed me that doctor kogan didn't have the validation for commercial purposes, only academic purposes, that was new information to me, and he requested that if i still had the data, to deleted, and then
5:27 am
sign a certification of that, that i no longer had the data. >> didn't require you to sign that under the provision of a notary, or just sign it and send it back? >> it did not require a notary or any sort of legal procedure, so i sign the certification. i sent it back, and they accepted it. >> senator harris, just as a housekeeping, we will have a vote called about noon. on so the time, we need to be there in about 25 minutes. so i think we will be able to get the first run, we have three more centers, i am happy to stick around to the end of the vote, if we can agree to 3 minute rounds before the vote is called. senator durbin? >> senator harris, i thought ur prentation on thissue
5:28 am
of privacy was spot on, it really describe what we are about, or one of the things that the should be about. i asked a question of mr. zuckerberg, whether he felt comfortable telling me the name of the hotel he stated last night, and after couple of minutes, after couple of seconds of hesitation, he said no, i've asked a lot of questions, that got a lot of attention. because it got to the heart of the issue. in terms of mr. zuckerberg, and his own feeling about personal privacy, and where he would draw the line. we know of course from what senator harrison said, and life experience, the last hotel i stayed in is probably a matter of some record with my name attached to it, somewhere, who knows? but privacy is a critical issue here, and the right of facebook or any entity to use my formation without myex provision, i think is over the line. i said to senator cornyn, my
5:29 am
friend, earlier, when he said he thought consumers were aware when information was being gathered on him, but i'm sure he's wrong. we have now put a little piece of electric tape on the camera on my laptop. most people do now, because they are being watched, and they may not know it. there are two other issues here, and one of them relates to mr. alexander nix, a british national, is that correct?>> yes. >> he was clearly involved in some of the campaign activities of cambridge analytica? >> he was the ceo of the company, so he was the point person for all the clients. and he often made the presentations and recommendations to those clients. >> the federal election commission says expressly that
5:30 am
regulations prohibit foreign nationals from directing, dictating or directly or indirectly participating in t decion-makg process of any person with regard to election related activities the united states. so there is a red flag, or red union jack, that should make it clear that we are in a territory that it may be, a violation of law, and you have said that the involvement of cambridge and successor organizations in russia, came after you left, is that correct? >> sorry, can you clarify slightly what you mean by -- my experience, not related to trenton, as it relates to understanding the research that was being done, in terms of the focus groups and all that? >> the use of information by the russians in the election campaign?
5:31 am
>> oh sorry, i misunderstood, yes. >> that happened some of the place or after you left? >> i was not aware at the time that there was activities in russia to inuence united states election >> i would add that the second category, the first question privacy, the secondtion quesis the involvement of foreign nationals the nine states campaign, whether alexander nix personally, or russia as a company -- personally tried to impacted. getting to the heart of another important issue, the issue of the secrecy of this activity. the fact that we know robert mercer was engaged in this is because of something that is called open secrets, and other sources, that were not discordant -- disclosed in the ordinary course of business on this. did you have any guidance on cambridge analytica while you were there in terms of the secrecy of the clients that you are working for?
5:32 am
>> to your first point about non-us nationals working in u.s. elections, cambridge analytica received formal legal advice, which i disclose to the media, and that legal advice did make clear that the company should not be funding non-us nationals to work on american elections. with respect to the secrecy, everybody had to sign a nondisclosure agreement, a very thorough nondisclosure agreement, and in fact, after i left, when the company engaged in the protected -- protracted legal correspondence with me, ensuring tha others who left as well, an undertaking of confidentiality, which is a higher threshold of nda. >> those two documents, the
5:33 am
legal opinion about the involvement of cambridge employees -- >> i made it public, and -- >> if you wouldn't mind sharing it, and also a copy of the nda that you or others might assign, thatwould be helpful. professor hirsch, elections very much turn on voter turnout, soyou would agree that efforts to suppress voter engagement or turnout should be a matter of serious concern to us?>> certainly. >> you testified that it is easier to demobilize people rather than mobilize, the impact of voter id requirements on voter turnout. so if we were to consider regulating anything, and i realize that there are privacy issues, it's all very competent, but if you want to focus on some kind of regulatory scheme, should we focus on regulating ads or messaging needs -- messages
5:34 am
that demobilize people? >> it's cupcake, i worked for the aclu in the obama ministration, voter id laws, this is really different, and it is different because campaigns do things that are not nice all the time, if a democratic campaign reminded trump folders -- voters, of all the moral failings, is that demobilization or not? >> is not the easiest thing to figure out whether a message actually does it, but let's assume that we can find a way to define it, because that is a matter of serious concern, that is easier to affect you said, so that might be an area for us to pursue in terms of any kind of regulation in this very complex space that we are in right now?>> that might be right, i am not a regulatory expert.
5:35 am
in terms of first amendment issues and so forth, if i were to ask somebody who were they were voting for, and they said president trump, don't vote for them, i would really rather you don't vote, it doesn't seem to me that is a form of voter suppression. >> we are basically talking about people pain to suppress votes, steve bennett was running an operation where clients could request voter suppression messages, which they would then pay for, we are talking about a different kind of circumstance then saying don't vote for so-and-so. i think you understand the differences. mr. wiley, you obviously have a knowledge of the use and misuse of a lot of information. the u.s. immigration and customs enforcement has proposed a new vetting initiative, hiring a contractor to exploit publicly available information, such as media, blogs, public hearings, conferences, social media websites, such as twitter, facebook, linkedin, to extract pertinent information regarding
5:36 am
targets. to determine who will be a productive member of society and who will commit crimes and terrorist acts. we are talking about people who want visas to come to the country. untoward -- according to the nonpartisan center for justice, deportation or visa, denial bad on the exact criteria from the original muslim band, so is the original muslim band, that -- the original muslim ban, do you think that prediction of human behavior, where that person will be upstanding or terrorist, is that even possible from the sources of information that cou be available to u.s. government?>> two points to that, there is no you -- there is no universal
5:37 am
definition of a bad person, therefore it is a social construct, it is laden with people's moral judgments, so there is no mathematical way to determine whether or not somebody is a bad person in the abstract, in the sense that there is no universal definition of a bad person. so the second point that i would make though, is that just because you are using data science and mathematics, you could have the most advanced neural network made yet, but if the underlying training set for the algorithm uses systemically biased information, so for example if you're looking at criminal justicstatistics, and you have a model that ends up predicting that this particular african-american is more likely to commit a crime, because more african americans in the prison, you create an algorithm that has the biases
5:38 am
of the information making it. >> i think it is problematic, in terms of all kinds of issues, privacy, they are pursuing it, by the way. i think it is problematic, the government getting its hands on information whether someone will commit crimes, does that make sense to you? >> i have a lot of concern about ing th to pit people against eaother, one of the greatest values, indivisibility, the the lines that ties together are stronger than the lines that divide us as a
5:39 am
country, when i woke up, i listen to the new york times podcast, all about sri lanka, when facebook rumors inside real violence, it was about how the platform was being used by sinister forces, to incite hatred between groups in fear, and ultimately violence in the circumstance, i was just so deeply disappointed and angered when my jokes, the founder of the sco group, would say things in recorded conversation, where he says things like this a resnick, to attack the other group, and know that you are going to lose them, reinforcing and resonating your group, and that's why hitler attacked the juice, he didn't have a problem with them, but people didn't like the jewish people, he leveraged an official enemy. that is exactly what trump did, he leveraged a muslim, and i mean really the balls to say what people wanted to hear.
5:40 am
this is a frightening reality, the threat to a very idea of a nation that was unfounded because we were alike. that united the country.
5:41 am
5:42 am
5:43 am
5:44 am
5:45 am
5:46 am
5:47 am
5:48 am
5:49 am
5:50 am
5:51 am

21 Views

info Stream Only

Uploaded by TV Archive on