Oral evidence: Fake News, HC 363

Oral evidence, 6 June 2018.

Witness

Alexander Nix.

Chair

Good afternoon and welcome to this further session of the Digital, Culture, Media and Sport Select Committee inquiry into disinformation and fake news. The Committee is very grateful to Alexander Nix for agreeing to come back in front of the Committee today to answer our questions. We appreciate that a lot has transpired since the end of February this year. I thought it would be helpful for the Committee and for the record if you could give us an update as to what the current status of Cambridge Analytica is, who is in possession of its assets and what role you have with the company at the moment.

Alexander Nix

Thank you, Mr Collins. I would be pleased to answer your question but before I do so, I would like to make a few short clarifications. It will take only a few minutes but it is important to be able to frame my answers.

Chair

Mr Nix, what I would rather do is if we could take the questions in the order the Committee has them. I would imagine that we will probably raise questions and issues where you will want to make clarifications and statements as we go through. If, by the end of the session, there are things that we have not covered on which you wish to place your response on the record, I would be very happy for you to make a closing statement to cover that off. It is not the practice of the Committee to have opening statements from witnesses, so I would be grateful if we could start with the Committee’s questions and see how we go.

Alexander Nix

I do appreciate that and ordinarily I would respect that but these are not ordinary circumstances and so, if I may, I would like to start with a very brief statement to set out my position.

Chair

Is this to do with the substance of evidence or is this to do with your current status and the impact that might have on your testimony as a witness?

Alexander Nix

This is to do with framing my answers today.

Chair

Okay. Well, let’s see how we do.

Alexander Nix

Thank you. First, as the Committee is aware, I did not refuse to attend to give live evidence—quite the contrary. Over the last few weeks I have expressed to the Committee my desire to attend as I am keen to deal with the untrue and misleading statements that have been made about me and my company over the past few months. My request was that this hearing was postponed until after the investigations in the United Kingdom and in the US have concluded. Alternatively, as with certain other witnesses, I asked whether evidence could be given in private so that I was able to answer all of the Committee’s questions fully and openly. I have been given clear advice that discussing topics that relate to ongoing investigations in the UK and the US may prejudice those inquiries and infringes my right to silence. This is a fundamental right extended to all those who find themselves in a similar position to mine. None the less, I have been compelled to attend. With this in mind, you will understand there is a limit to what I can say in relation to certain topics that touch on those investigations. Out of respect to the Committee, I have attended today and will answer the questions that I can. Can I address straightaway the suggestion that I misled this Committee about receiving data from GSR?

Chair

Mr Nix, I appreciate the opening statement about your position and the fact that there are other inquiries with which you are co-operating. I think we are now getting into you giving a statement about the substance and this will be covered in the questions that members wish to put to you. I would rather take this on a question-by-question basis than it being dealt with as a statement at the beginning.

Alexander Nix

I do understand that but it is important that I push back on this and so I am going to have to insist, Mr Collins, that I finish my statement.

Chair

I am sorry, Mr Nix. I have allowed you to give us some opening remarks. I feel that your opening statement is now going into the substance of issues that we want to raise with you, which is based on your previous evidence to the Committee and the reason we invited you back. I am going to deal with that on a question-by-question basis. It is up to you how you respond to those questions. As I said at the beginning, if there are things that we have not covered where you wish to place a statement on the record, I am very happy for you to do that at the end of the session once the members have completed their questioning. I am not going to allow an opening statement that goes into the substance of topics the members want to question you about.

Alexander Nix

Mr Collins, you will have plenty of opportunity, as will all the Committee, to ask me as many questions as you want but I have to insist on delivering the rest of this statement.

Chair

I am sorry, Mr Nix, it is not your place to insist on anything in front of this Committee. You have agreed to come and you are here to answer the members’ questions. I have allowed you to make some opening remarks that reflect the constraints that you may be under because of other investigations you are taking part in, and that is completely proper. I want now to go on to the questions the Committee intends to put you to and to my opening question, which is about the current status of Cambridge Analytica and your current status with that company. Could you answer that question, please, and then we will go on?

Alexander Nix

As I said, I will be pleased to answer that question but I will just finish this statement, if I may.

Chair

No, Mr Nix. I will start the questions now.

Alexander Nix

I accept that some of my answers could have been clearer but I assure you—

Chair

Mr Nix, you are just reading out the statement. Can you answer the first question that I put to you? Then we will go into the questions from the members. If there are other things you want to put on the record, I would imagine that you can cover off a lot of what is probably in that statement in your answers to members’ questions as we go through. Could you answer the first question, please?

Alexander Nix

Mr Collins, I really don’t think this is appropriate. Other witnesses have given statements to this Committee and I think I should be allowed to do likewise.

Chair

No, they have not given statements to this Committee. People have been asked, as I have asked you, a relatively general opening question, which you can respond to. The character of all of the hearings so far has been that members have put questions, with me starting as Chair, to witnesses who have answered them. We do not have opening statements and that is the case for all witnesses who appear in front of the Committee regardless of the circumstances. I think it was appropriate for me to allow you to make some opening remarks about your status as a witness, which is helpful to the Committee’s work, but the way this works is that we ask questions and the witnesses answer them.

Alexander Nix

Could you repeat your first question, please?

Chair

Yes. Thank you. Since we last met, as I said at the beginning, there has been quite a lot of change at Cambridge Analytica. Could you update the Committee on what the current status of Cambridge Analytica is, who is in control of the assets of the company and whether you have any direct involvement in the running of the company today?

Alexander Nix

Cambridge Analytica is in chapter 7 in the United States. The SCL Group and associated companies are in administration in the UK. The process is being managed by the administrator and I have no further involvement.

Chair

The assets of the company are in the control of the administrator?

Alexander Nix

That is correct. [Interruption.]

Chair

Did I hear a device? There is no photography allowed in the room by members of the gallery. If people can’t respect that, I will ask that they leave the room. I hope I do not have to make that statement again. Is that the same with SCL Group? Are its assets in the hands of the administrator?

Alexander Nix

That is correct.

Chair

You will have seen reports in the newspapers today suggesting that you took an $8 million sum of money out of Cambridge Analytica before it went into administration. Is that correct?

Alexander Nix

I saw that newspaper article today. What I can say is that the allegation made in that article is false; the facts in that article are not correct. Other than that, I cannot speak to it, simply because it alludes to other matters that I am not so familiar with. What I can say is that over the last three months since this media storm erupted, I have personally invested, unbeknown to anyone except the board, millions of dollars to meet staff salaries, staff bonuses and staff redundancies.

Chair

Could you say when was the last time that staff members at SCL or Cambridge Analytica were paid?

Alexander Nix

I can’t speak to that because, as I think this Committee knows, I was suspended from my role as CEO and then resigned from the board. These matters would have been handled by the board and then the administrator.

Chair

The money that you said you put in to meet staff payments, is that money you gave to the company or to the administrator and when did you make those payments?

Alexander Nix

To the company.

Chair

How long ago was that?

Alexander Nix

I would be happy to circle back to you once I have had more facts about this article, but I don’t think that this is relevant to this Committee hearing.

Chair

I wanted to give you the chance to respond to what was in the article. Finally, has the administrator been in touch with you about that amount of money that was taken from the company?

Alexander Nix

No, he has not.

Paul Farrelly

In line with the current state of play, it is right to give you the opportunity here in public to respond to the continuing media storm. You said that you put money into the company—you used the word “invest”—to pay staff salaries and bonuses. Was any of that redeemed in any way?

Alexander Nix

By myself?

Paul Farrelly

Yes. Did you receive any of that back?

Alexander Nix

No, I did not. My intention was to try to help the company through a difficult period when we were uncertain about what was happening. I was also keen to try to help the staff as much as I could. At that time, none of us envisaged that this outcome would happen.

Paul Farrelly

Not a cent of that has come back to you.

Alexander Nix

Not a cent.

Paul Farrelly

Did that place you as a creditor of the company or how did you invest the money? Did you loan it to the company?

Alexander Nix

Again, I don’t think the details of that are important but given that the company has no assets, being a creditor to the company is probably not so important.

Chair

Aside from the financial assets, are all the assets of the company, including the data it holds, other information it holds that has a commercial value, in the hands of the administrator?

Alexander Nix

That is my understanding.

Chair

I have one final question on some of the events that transpired affecting the company directly. At the time, after the publication of the story in The Observer newspaper, Facebook sent lawyers and forensic data staff to the offices of Cambridge Analytica and they were asked to leave by the Information Commissioner. Things were also removed from the offices by Cambridge Analytica itself or under the direction of Cambridge Analytica. Can you tell us what was going on at that time, what was removed from the building? Have those documents, files, datasets or whatever it was been made available to either the administrator or to the other relevant authorities that may have been interested in them?

Alexander Nix

Could I ask you, Mr Chairman, what makes you presume that anything was removed from the office at that time?

Chair

There were photographs of boxes being removed and things being taken, which has been reported on. If that is not the case, I am very happy for you to say that is not the case.

Alexander Nix

I just wanted to check that is where you were getting your information from. I also saw those photographs. My understanding is that those photographs were from personnel who worked in a different office on a different floor that had absolutely no connection with Cambridge Analytica or SCL and that the media simply interpreted this in the way that they thought was most beneficial to them.

Chair

As far as you are concerned, there were no documents, files, computers or anything removed from the office of Cambridge Analytica?

Alexander Nix

I can tell you categorically that the people or persons in those photographs were not part of Cambridge or SCL and those boxes were not ours.

Chair

Otherwise, apart from what was reported in the media, there was no other removal of documents, information, computer files, whatever it may be, from Cambridge Analytica’s offices during that week?

Alexander Nix

As far as I am aware, the ICO were the only people to remove documents, files and hardware from the office.

Chair

At all during that period of time? They did not get a warrant to go into the offices until the end of that week. In that week, as far as you are aware, nothing was taken from the offices?

Alexander Nix

As far as I am aware.

Jo Stevens

I want to ask you, now that it is some weeks on and in view of everything that has happened, how you feel about the Channel 4 undercover footage of you and Mark Turnbull discussing what work you did. Do you feel embarrassed about it?

Alexander Nix

Well, clearly. For me to get carried away and try to impress someone who I thought was a bona fide client who had come to us with a very genuine and serious project to help the people in Sri Lanka with health and IT infrastructure solutions, to then get entrapped into talking about the sort of things that we did is not only deeply embarrassing but it is something I regret enormously because of the damage it caused to the company and ultimately to my staff and the impact that it had on them. I would not wish that on anyone.

Jo Stevens

You are not embarrassed about what you told them, the bribery, the honeytraps, the fake IDs, the websites and the extortion? You are embarrassed about the impact it has had on your reputation and on the jobs of, I think it is, 80 people who worked in the business?

Alexander Nix

It is more than that if you count the US, of course. No, that is not what I said at all. What I said was that by speaking with such exaggeration and hyperbole, I did not represent the company properly, I did not represent what we do as a company or as individuals, and that there was a significant impact from doing that. I go on to say that there were some five and a half or six hours of meetings that took place with this individual. Ostensibly these were very legitimate meetings about legitimate subjects. All of these were recorded and they were condensed to maybe six or seven minutes of footage. Even those minutes of footage were heavily edited to portray the worst possible picture of the company, myself and my colleagues.

Jo Stevens

That worst possible picture, Mr Nix, was word for word what you told the journalists. You gave them that information. You gave them that evidence. I was interested to understand how you feel about it now some weeks on. Obviously the statement—

Alexander Nix

I think I should clarify that clearly we did say those words but the context in which they were said has been massively altered and also the caveats that we attached to those words have been altered. Let me illustrate, if I may, for a moment. My colleague was absolutely crystal clear in telling the undercover reporter that we are not in the business of fake news, we are not in the business of lying, making stuff up. We are not in the business of entrapment, so we would not send a pretty girl out to seduce a politician and then film them—

Jo Stevens

Yes, I have got that.

Alexander Nix

Please, if you don’t mind—in their bedroom and then release the film. There are no companies that do this—there are companies that do this but to me that crosses the line. If I may go on—

Jo Stevens

No, I am going to ask you the question, Mr Nix.

Alexander Nix

If I may go on, please. When the undercover reporter put these issues to me—

Jo Stevens

Would you let me ask you the question, please? It is my job to ask the questions; it is your job to answer them. If you could just do me that courtesy, please. Mark Turnbull said that. I accept that entirely. I have read the transcript, but then you go on to say about beautiful Ukrainian girls and you give examples of what you would be doing—and I have referred to them before—bribery, honeytraps, fake IDs, extortion, fake websites. You were suspended by Cambridge Analytica. The company put out a statement saying that your comments “do not represent the values or operations of this firm and his suspension reflects the seriousness with which we view this violation”. What are the values, what were the values of Cambridge Analytica?

Alexander Nix

I will address your question but let me—

Jo Stevens

No, please answer my question. Alexander Nix —continue by saying that—

Jo Stevens

Please answer my question, Mr Nix. Alexander Nix —I caveated my comments very, very clearly by saying, “The answers are hypothetical and that is really important. Please don’t pay too much attention to what I am saying. I am just giving you examples of what can be done,” and, of course, this was edited out of the video. I was asked a very leading question, a hypothetical question about what sort of tactics could be employed to discredit a politician and I answered hypothetically and I caveated my answer by saying, “Please don’t pay too much attention.” Of course this was edited from the footage.

Jo Stevens

Can you now answer my question, please? What were the ethics, what were the values of Cambridge Analytica?

Alexander Nix

At that meeting we had a client that we thought was bringing a very interesting contract to the company and the client started taking the conversation down a completely uncharted avenue. In our overzealousness to secure a contract, Mark Turnbull and I were guilty of hyperbole and agreeing with the client on matters that not only have we never involved ourselves in before but we had no understanding of. We were just telling the client what we thought he wanted to hear in order to find out exactly what his intentions were and to move the contract forward.

Jo Stevens

I have just been passed a case study from your marketing brochure: “As part of an election campaign that turned nasty, we launched a targeted digital attack on the opposition candidate. Within three weeks, every single reference to him on the first two pages of Google referred to the candidate’s horrific track record of corruption, coercion, rape allegations and victimisations.” Are those the ethics, are those the values of Cambridge Analytica?

Alexander Nix

If you have a candidate who is being found guilty of corruption or guilty of rape or other such things, pushing those into the public space is in the public interest, surely.

Jo Stevens

What are the ethics and what are the values of Cambridge Analytica? Did you have any at all? If you didn’t, just tell me, but I am interested to know: did it have any business ethics or values?

Alexander Nix

Cambridge Analytica is a company that is predominantly staffed by very young, very ideological and very talented individuals who come from a broad range of political ideologies and interests. They genuinely have been working to understand how they can use data to make the world a better place on many different levels, by contributing to developing communication and how that is used, and also in the social and development sphere, in counter-terrorism and counter-radicalisation. There are huge areas of positive work that I could point to, but instead the media—and it seems this Committee—is intent to focus on this one meeting that was clearly not representative of the values nor the ethics of the company.

Jo Stevens

Let me move to the report of Julian Malins, QC, into Cambridge Analytica. You mentioned young, idealistic employees of the company. Would they be the same employees that Mr Malins talked to who described you as delusional, that you should have been reined in a long time ago, that you and Mark Turnbull were behaving like teenagers desperately trying to get a deal? Are those accurate assessments of you and Mr Turnbull?

Alexander Nix

I think the quotes that you are referring to were in their response to that particular Channel 4 investigative video. I think that they are quite right in their criticisms. As I think Brittany Kaiser said in her testimony, she did not recognise the company that was portrayed in that video. I think the staff were rightly very disappointed by the fact that, under the very skilful manipulation of the journalist, he was able to coax us into saying things that neither represent what we do nor the values and ethics that the company holds.

Jo Stevens

So why say them?

Alexander Nix

Clearly, if I could have the meeting again I would have it differently. You do appreciate that.

Jo Stevens

I am sure you won’t take exception to this but it goes to the very heart of the credibility of your business reputation and your evidence to us that you will say those sorts of things when you are telling us they are completely untrue.

Alexander Nix

I hope in due course and over time that I will be judged by my actions and not words. Clearly in this case my words were foolish.

Chair

To be clear, on the Channel 4 investigation there was quite a lot of focus on the reference to the Ukrainian girls; I am sure you will remember that. What you said there was, “They are very beautiful. I find that works very well,” so it was not a kind of hypothetical. That was saying, “I find that works very well.” Was that just a lie to impress the people you were talking to or was it something else?

Alexander Nix

That was just a lie to impress the people that I was talking to.

Chair

I am interested in whether this is out of character with the company. Jo Stevens mentioned the QC’s report that the company commissioned. In that he said that if this sort of thing happened then you would probably find examples where politicians who were on the opposite side of a campaign that SCL or Cambridge Analytica was involved with would be exposed in some sort of sex scandal. He said from what he had seen that that had not happened, but that is not true, is it? In 2015 the Prime Minister of St Vincent was caught out in a sex tape scandal and he was on the opposition side of the campaign that your company was involved with; isn’t that correct?

Alexander Nix

I don’t believe that is correct.

Chair

I have a copy of the newspaper report on the scandal itself, “St Vincent and Grenadines prime minister comments on sex tape”. The party that you were working for is the NDP and they said they had nothing to do with it but nevertheless the sex scandal took place. The QC who wrote your report was not aware of that particular example.

Alexander Nix

Nor was I aware of that, Mr Collins. That is the first time I have learnt about that. I am not sure that we—and I will revert to you on this. You said 2015?

Chair

That is the article I have here. It may be earlier than that.

Alexander Nix

I am going to say to the Committee—and I will circle back to you on this—with a high degree of certainty that we were not involved in that election in 2015.

Chair

We will check. The date I have is on the article I have in front of me. The piece of work itself may have been earlier. I think it may have been 2010 but we will double check that. You had not heard that story but to me that seemed to be a case where the QC may not have been aware of the breadth of some of the work of Cambridge Analytica.

Alexander Nix

I am sure that any of the Committee can point to a politician involved in a sex scandal, but that does not automatically mean, because I got entrapped by a journalist, that we were responsible for it. I think you are drawing a connection that does not exist.

Chair

I was referring to what was in the QC’s report. He was saying that if there was any example of a candidate being on the wrong side of a SCL or Cambridge Analytica campaign who had been exposed in a sex scandal, that might suggest that maybe there was some credence to it. It was not just something being said for the benefit of the tape or to impress a new client but that actually this sort of thing had gone on.

Alexander Nix

Let me make this absolutely categorical so that we can put this to bed. There is no credence to it. We have never undertaken any work that involved a honeytrap or the use of a sex scandal or anything like that or tried to create a sex scandal for political leverage or for work on a campaign. I can’t be more blunt or plain than that.

Chair

I have seen a document that was presented by you to some potential clients in Mongolia in late 2014. This is not about honeytraps but about forms of what the document calls “proactive campaign intelligence” that it says it can provide, including, “Investigations, stings, infiltrations designed to shed light on the opposition and to better anticipate their strategy”. Is that reflective of the sort of work your company does?

Alexander Nix

I am not familiar with that document, but the answer is no. In 2014, we probably had six or eight staff in the company, most of them were 25 to 35 years old. They came from either political consulting or academic backgrounds. We had not even the wherewithal to be able to undertake any of that work, even if we had wanted to, and we would not have wanted to and we never have.

Chair

There is an emerging pattern that in documents that have been presented and interviews when you were recorded talking to Channel 4: you were saying one thing to your potential clients about what the company does, but in reality you don’t do any of those things. You don’t do infiltrations and stings, you don’t do honeytraps.

Alexander Nix

I think you need to be careful about making sweeping statements.

Chair

Infiltrations and stings are two words that are in the document.

Alexander Nix

There was one interview and it was very leading and the questions were deliberately opaque and they led me down a certain road. I have already apologised to the board and the shareholders and the employees, and I am very sorry. I can say to everyone that I was so foolish to fall for a trick like that, but it was a trick and it does not represent what we do. It would not be fair to characterise me or the company by that one incident.

Chair

I am not trying to do it by the one incident. I am sure something we will look at as we go through the questions this afternoon is whether what you were recorded as saying on Channel 4 reflects the sorts of things that you say. It was not a one-off. Those are the sorts of things you say to potential clients and get put in pitch documents to potential clients; whether stings and attempts to acquire information about opposition candidates so they can be exploited during the course of a campaign is something that you routinely talk about when you meet new clients or whether that was a one-off. I think that is pertinent to understanding whether the Channel 4 thing was an aberration or whether these sorts of untruths were told routinely or indeed whether some of this sort of work goes on.

Rebecca Pow

It does seem strange that you are now telling us this was all just a little mistake and an aberration and you should not really have said these things. In that Channel 4 document you referred to SCL as the largest political consultancy working in the shadows. You are quoted as saying how you used the method of going to meet a potential incumbent and having video evidence of what they had done so that you could basically pop it on the internet and then they might be accused of corruption, offer a large amount of money to a candidate to finance his campaign in exchange for land and so on. It seems to me very odd that you are now saying that you did not at any time have any of these involvements, practised what I might call the dark arts.

Alexander Nix

Let me start by saying that there is nothing more sinister about what you have just read out than what Channel 4 was undertaking itself.

Rebecca Pow

But they are actually your words that we are quoting back to you, and—

Alexander Nix

Does that mean that Channel 4 also works in the dark arts?

Rebecca Pow

Another thing you said is: “you know equally effective can be…to…speak to the incumbents…offer them a deal that’s too good to be true and make sure that that’s video recorded,” so that you could use it later. It does seem really odd that you are now saying none of that was true and you were not practising these dark arts.

Alexander Nix

Let me clarify. First and foremost, I was drawing a very particular distinction about what Mr Chairman was talking about, which was the use of honeytraps and the use of—what was the other word you used? I can’t remember.

Chair

Infiltration and stings.

Alexander Nix

Infiltrations and things like that. We don’t do that. We have not done that. I think that most of this Committee would recognise that there is a difference between honeytraps and infiltrations and, for instance, what Channel 4 would call an exposé.

Ian C. Lucas

Mr Nix, when you last gave evidence to the Committee you may recall that I asked you about ethics and in particular whether the company had ever refused a proposal from a client in connection with an advertisement or a film on ethical grounds. You said you would go away and come back to us. We have received a response from you that says, “We cannot find an example of where we had to refuse to put out an advert on behalf of one of our clients”. Isn’t that a clear example that you have not applied an ethical approach to the advertisements that you put out on behalf of Cambridge Analytica and SCL?

Alexander Nix

That is certainly one interpretation of that. Alternatively, you could simply say that that is an example of where we have not had to employ those sanctions because we have not been asked to undertake anything unethical by a client.

Ian C. Lucas

There are some films about the Kenya election that were featured in the Channel 4 film, which I think your colleague referred to as being effective. Are you saying that at no stage have any of your clients used unethical advertisements or communication?

Alexander Nix

This was an allegation that I believe was made by Mr Wylie about the fact that we disseminated unethical media content in the Kenyan elections. I think that subsequent to that testimony what we have seen is that the actual company responsible for producing and disseminating that was an American-based political consultancy, digital consultancy, who had no relationship to Cambridge Analytica whatsoever and that it was simply convenient for Mr Wylie to imply that we had disseminated that when we had nothing to do with that campaign. We were not involved in that particular set of adverts or TV commercials or whatever they were. I believe a spokesperson for the campaign has also exonerated us from any involvement in that.

Ian C. Lucas

Mr Nix, isn’t it the truth that Cambridge Analytica cannot produce a single example of refusing a proposal from a client for an unethical advertisement and, therefore, Cambridge Analytica was an unethical company itself and did not apply high ethical standards?

Alexander Nix

I don’t think you can draw that conclusion from your statement. I think you would like to draw that conclusion because it fits with a certain prejudice that has been prevalent in the media but has not been borne out by evidence. I certainly have not seen any evidence that supports this, apart from this Channel 4 entrapment video. I don’t agree with what you have said, so the answer is, no, it is not true.

Ian C. Lucas

You have produced no evidence of Cambridge Analytica rejecting an advertisement on ethical grounds.

Alexander Nix

That is correct, but that does not mean that we are unethical. It simply means that we have not worked with or been asked to disseminate unethical advertising.

Chair

I think what Mr Lucas was referring to as well in the case of the extremely violent video relating to the Nigeria campaign is that SCL was running the campaign and AIQ was asked to disseminate it. What AIQ told us was that they refused to do so. They did not say that the video did not exist. They said that they refused to have anything to do with disseminating it. We only have their word to go on for that, but that was the evidence that they gave. I think that is the reason why Mr Lucas asked that question.

Alexander Nix

There were two incidents. One was in regard to Kenya, which I hope I have addressed to your satisfaction, and the other was in regard to Nigeria. When I appeared before the Committee last I was unaware of that video. I still have not seen that video so I can’t speak to its content, but I did listen carefully to the testimony of Jeff Silvester. I also read the testimony of Brittany Kaiser. I have spoken since to other employees who were in the company at the time. This was another allegation that stemmed from Mr Wylie against the company’s unethical practices, but what has become clear or what is absolutely clear to me is that that video was not produced by SCL Elections or Cambridge Analytica. Not only did we not have the inhouse capability to edit or produce videos; we certainly did not have that capability in Nigeria for the team that was on the ground. My understanding is that that video was presented to us by the client who asked us to disseminate it and that a discussion took place between the members of the campaign team, some of whom were SCL, some of whom were subcontractors like AIQ, and a decision was reached not to disseminate it. I can’t speak to this directly because I never went to Nigeria on this campaign, but that is my very clear understanding from the other parties and members of the company who I have spoken to. At the time when you asked me that question, Mr Lucas, I was not aware of this. I am now aware of it so I can point to a video or a client campaign that we were presented with that we did not disseminate.

Ian C. Lucas

Why didn’t you tell the Committee this before?

Alexander Nix

When I came in front of the Committee on 27 February that was prior to Brittany Kaiser’s testimony, Christopher Wylie’s testimony and Jeff Silvester’s testimony and I was not aware that this video even existed.

Ian C. Lucas

In your response, which was subsequent to you giving evidence, on this specific point you still did not produce it as an example. Today is the first time we have heard about this.

Alexander Nix

My apologies. I am telling you now. When did that response come to you? Was it before Brittany’s testimony? I really can’t remember.

Chair

Your response was before Brittany’s testimony.

Alexander Nix

There you are. That is the answer to your question.

Christian Matheson

Listening to your testimony a little while ago, you were a bit unlucky with that Channel 4 video, weren’t you, Mr Nix? Of all the business pitches that you have done and that you have done ethically, the one that you did unethically was the one that was recorded.

Alexander Nix

I don’t know that I was unlucky. I think other business meetings are not steered by an undercover reporter who is deliberately trying to entrap you and, therefore, the conversation never moves into those areas and towards those subject matters. It was not luck; it was a very deliberate and co-ordinated programme of deception by Channel 4, together with other newspapers, where they deliberately seeded questions to elicit certain answers. I don’t think luck had anything to do with it.

Christian Matheson

But that is the business that you were in, of course, isn’t it? It is behavioural change by deliberately seeding people’s thoughts.

Alexander Nix

I know that the Committee would like to believe the media’s impression that we are this large, nefarious, multinational company that influences politics and other things around the world. The truth of the matter is that we are a very small advertising agency that happens to work across a number of sectors, one of which is political campaigns that constitutes only about 25% or 30% of the revenue of the business. Most of our time is spent selling toothpaste and automotives and things like that.

Christian Matheson

If I can come back on to the question of substance in this session. You have told us previously that in respect of the organisation WikiLeaks, “We have no relationship with WikiLeaks. We have never spoken to anyone at WikiLeaks. We have never done any business with WikiLeaks. We have no relationship with them, period, full stop, in this country.” Does that remain the case?

Alexander Nix

That remains the case.

Christian Matheson

I see that this afternoon there is an article on The Guardian website suggesting that when Brittany Kaiser was a Cambridge Analytica director she was visiting Julian Assange in February last year and told a friend that it was to discuss what had happened during the US election. The article goes on to say that she had also claimed to have channelled cryptocurrency payments and donations to WikiLeaks. Is that correct?

Alexander Nix

I simply cannot speak to that. That is the first time that I have heard that. I have not read the article. I have absolutely no knowledge of the fact that Brittany Kaiser had met with Julian Assange. I cannot speak to the veracity of that reporting. If it is anything like the majority of the reporting that that newspaper has said about our company, I can only be sceptical about the truth that it is based on.

Christian Matheson

Their source is visitor logs from the Ecuadorian Embassy obtained by The Guardian and an organisation that I don’t know called Focus Ecuador. It sounds as though, if the visitor logs are correct, it might be the case.

Alexander Nix

It might well be the case. You are probably better off speaking to Carole and asking her directly.

Christian Matheson

I might do. It is just that I could not ask her before because the session had already started, for which I apologise.

Alexander Nix

I really have never heard this before. This is completely new to me, but I am assuming that Brittany did that in a personal capacity. It certainly had nothing to do with Cambridge Analytica or SCL Elections.

Christian Matheson

The one concern about that is obviously that she was doing it on your time. Also that is the second instance now, following on from what my colleagues have asked, where things have been done in the company that you were not fully aware of. Are you a little bit concerned that perhaps you did not have a complete grip on the activities of your colleagues and what they were doing around the globe?

Alexander Nix

If you are asking me if I micromanage 120 staff in offices across the US and the United Kingdom such that I understand how they spend every minute of every day when I am travelling extensively, you are absolutely correct.

Christian Matheson

But in terms of WikiLeaks and Julian Assange, who obviously is a very high profile individual and has undertaken some fairly high profile publication of leaks, including those that are used for political embarrassment—and we have seen from the United States, for example—it might have been useful, particularly since you had work in the United States, if you had been aware of that.

Alexander Nix

I don’t know how I can say this more clearly. I was not aware of it. I did not know that Brittany Kaiser had been to the embassy. I can’t speak to her motives; I can’t speak to what she was doing. She certainly was not there representing Cambridge Analytica or SCL.

Chair

Regarding Julian Assange, I wanted to ask you a couple of questions about Brittany Kaiser’s evidence to us, which I am sure you have seen. Did you hire Brittany Kaiser personally to come and work for your company? Was she someone you knew before she joined the firm?

Alexander Nix

I was introduced to her personally and then she would have been interviewed as were other candidates for a job.

Chair

She obviously knew Julian Assange before she came to work for you. She spoke about that and in particular through her relationship with the legal chambers that represent him. She said that you knew that she knew Julian Assange as well. Is that correct?

Alexander Nix

It is certainly correct that Brittany Kaiser had worked—I don’t know for how long—in human rights for the QC John Jones and I believe that he represented Julian Assange at some point.

Chair

Yes, that is correct.

Alexander Nix

I can’t speak to whether she knew Julian Assange directly or had ever met him but I knew that she had a relationship with John Jones.

Chair

You were aware of that connection and the connection on to Julian Assange?

Alexander Nix

That is correct, yes. I met John Jones once.

Chair

It came up in Brittany Kaiser’s evidence, because you spoke to us about Julian Assange the last time you came, saying that you made an attempt to reach him via a speaker bureau or some other agency.

Alexander Nix

That is right, yes.

Chair

Given that you knew Brittany Kaiser had a route into Julian Assange, at least through his legal representation, why didn’t you just ask her if you wanted to get a message to Julian Assange? Why did you go through that convoluted route?

Alexander Nix

One of my colleagues put that question to me the other day. I really think this demonstrates how much of a spur-of-the-moment idea it was to reach out to them. It was not something that was thought through. As I think I told this Committee last time very plainly, I read an article in The Guardian, I believe it was, that Julian Assange had access to these emails and I simply reached across to my colleague and said, “Would you mind seeing if you can contact him to see if he will share those with us?” That was the extent of it. We did not sit down and strategise this; we did not think it through. She very kindly followed up for me and 10 days later we got a response saying that, no, he was not interested.

Chair

But your motivation was to seek to get access to the Hillary Clinton emails that Julian Assange had in order to benefit your client, the Trump campaign?

Alexander Nix

I do not think we had really thought it through that far. The point was these were very contentious emails, potentially, and we wanted to understand what was contained in them, as did every journalist and I would say most political consultants on both sides of the aisle in the United States. I don’t think that curiosity is indicative of anything nefarious.

Chair

I was asking just to clarify what the motivation was.

Julie Elliott

Mr Nix, in February when you appeared before this Committee, you said that you had no relationship with AggregateIQ apart from using AIQ to develop some software.

Alexander Nix

That is correct.

Julie Elliott

Would you like to reassess your original answer now in light of what we have heard in the last few months about all of this?

Alexander Nix

I want to double check. I think the question was, “Do you have a relationship with AggregateIQ?” I answered no, because we do not have a relationship with AggregateIQ and we have not had a relationship with AggregateIQ for many months, or indeed for well over a year or a year and a half. I was absolutely correct and I stand by that. I then volunteered some additional information to you to help you understand that we had had a relationship with AggregateIQ, principally in 2014 when they built a software platform for us.

Julie Elliott

Which platform was that?

Alexander Nix

Internally we used the name Ripon for it. In addition to that, they worked as a subcontractor to our firm on a number of projects that involved building other smaller pieces of software and they undertook some digital marketing for us.

Julie Elliott

If we are talking about the Ripon platform, can you tell us a little bit in detail about how that was built?

Alexander Nix

I couldn’t speak to that. Do you mean technically how the code base was put together.

Julie Elliott

Yes.

Alexander Nix

No, I can’t.

Julie Elliott

Who was it available to and that type of thing?

Alexander Nix

I can speak to that, certainly. We used AggregateIQ as an independent company of specialist software engineers to help us build what was called a voter engagement platform. The easiest way to think about this is a piece of software that allows you to visualise and manipulate data so that you can make decisions based on those data.

Julie Elliott

What was the intention of using that data, of using that Ripon platform? What was the outcome you wanted?

Alexander Nix

The intention was to be able to provide this as a tool to clients to be able to use so that they could identify and target audiences for campaigns.

Julie Elliott

Where would they have got the data from to use in that programme?

Alexander Nix

The vast majority of data that was used in this platform was a combination of voter registration data, which is something that is commercially available in the United States, and consumer and lifestyle data, which is something that you can go out and commercially acquire or license from data vendors.

Julie Elliott

The tech website Gizmodo revealed the entire code base for Cambridge Analytica’s Ripon platform. The security firm Upstart revealed the integration of Cambridge Analytica and AIQ through another base code. Why was so much of your source code kept in such unsecure locations?

Alexander Nix

That source code was written and developed by AIQ, as I have just made clear, and so you would have to address that to them. It would have been their responsibility to keep that code safe. But that code—

Julie Elliott

Did you not have any oversight on that?

Alexander Nix

We tasked a subcontractor to build us a product.

Julie Elliott

I would have thought data security was quite important.

Alexander Nix

We tasked them to do that because we did not have the capability or wherewithal inhouse to do it. It would have been difficult for us to be able to micromanage or police that at that time. But I think more importantly—

Julie Elliott

What information did they give you or what did you ask them to make you satisfied that that data was secure?

Alexander Nix

I think there is a confusion here and I want to explain this as best I can, given that neither of us are technical software developers.

Julie Elliott

I might be. I’m not.

Alexander Nix

The confusion seems to be that this code base was the same as the data that was contained on the platform. My understanding from the article, which we both seem to have read, and from the evidence that Chris Vickery gave was that the code base pertained to the platform. It was the code for the development of the platform and it was only part of the code, not even the entire code. My understanding is that it was just a small piece of the code. There is no correlation between the code base for the platform and the data that went into it. The best way to explain this using a metaphor would be that AIQ were tasked to build a car for us. They happened to have left a door lying around that somebody found but possession of that door does not at all give you access to the petrol that is put into the car. Does that work for you?

Julie Elliott

Yes, but do you think it was acceptable that the source code was in such an insecure place?

Alexander Nix

Other than the fact that it is our IP or IP that we commissioned, there is nothing sensitive or put at risk by the code for a programme being left on an open forum. Many developers choose to leave their code in open forums so that it can be shared with other developers for the benefit of the good of the community.

Chair

I don’t think it is quite as straightforward as that, is it? What we have seen is evidence of joint project working, people at AIQ and SCL having access to the same databases, working on the same databases. It was not as simple as AIQ being hired to design tools that the rest of the company used. People were working together on the same data on the same campaign.

Alexander Nix

My analogy was very specifically to illustrate the point that Ms Elliott was making about the code base that had been left on the internet. It could be better but I think that was a fair analogy. Your point stands, Mr Chairman, which is that after they had built the software platform for us, AIQ continued to work with us as consultants, not least to help some of our clients to interface with the product that they had built and to teach them how to use it.

Chair

AIQ worked with you before they built the platform. They worked in Trinidad, which I think was the first engagement.

Alexander Nix

That is correct.

Chair

I believe you were introduced to AIQ by Chris Wylie; is that correct?

Alexander Nix

That is also correct.

Chair

You obviously were very impressed with them because after meeting them they were very quickly involved in an important campaign with you. People have said some fairly disparaging things about Chris Wylie and his role in the company, but he introduced you to AIQ and you hired them pretty much straightaway. Would that be a fair reflection of what happened?

Alexander Nix

I don’t think that is fair at all. First, let’s talk about Chris Wylie. Chris Wylie had an absolutely instrumental role in the company.

Chair

That is not what it says in the QC’s report that your company commissioned. It basically makes him sound like someone who was an intern, a freelancer, who had no significance at all.

Alexander Nix

I think those are the facts of the matter.

Chair

What you have said now and what the QC said in his report—

Alexander Nix

The facts of the matter are that Chris Wylie was on a tier 2 student visa. He was limited to working 20 hours a week, I believe. You will have to check that. As such, he could only work two days a week or so at the company, which is probably why Mr Malins alluded to him as a part-timer or a consultant or an intern or whatever he said, but that is to undermine his value. Chris Wylie brought some very significant value to the company. Not only did he introduce us to AIQ but he was also the person responsible for introducing us to Dr Kogan. He was the architect of that relationship and the idea of asking Kogan to collect data and to use that sort of data on modelling personalities. He went on and managed that relationship right up until the point that he left the company.

Chair

I think what you said there is probably a fairer reflection of the value he brought to the company than what was in the QC’s report.

Alexander Nix

I can’t speak to the QC’s report. It was independent.

Chair

But it was commissioned by Cambridge Analytica. AIQ worked in Trinidad. They are then engaged to design the Ripon system that you have described. They worked on the campaign work that was done for John Bolton’s pack in the election races in America in 2014. They are then working on Ted Cruz’s campaign in 2015. Jeff Silvester said to us that probably 80% of their revenue from the beginning of 2014 through to the start of their work with Vote Leave was with SCL. I think many people would say that during that time their teams were totally integrated with yours. While they may be a legally separate entity, they were effectively part of the same team. Isn’t that the reality of the situation?

Alexander Nix

Well, certainly not in my mind. I never considered AIQ part of our company. It was a subcontractor that was brought in over a period of time to undertake very specific tasks for the company. They came in, they undertook a piece of work, such as building the Ripon platform, and then they left. Possibly later they would come in and undertake a different piece of work. So, the answer to your question is no.

Chair

But they had access to the same data during that period of time. I think what you want us to believe is that they are just designing discrete tools. They make those tools, they give them to you and they go away again. They might come back for another project. They did design and create a very important tool, the Ripon tool that you have described, but other than that they are working continuously with SCL, sharing data, helping improve targeting in campaigns, helping to run and execute those campaigns. They are a separate company. What kind of protocols or rules did you have in place about data sharing? Were there agreements signed with them on data sharing and data handling, given they were handling your company’s data?

Alexander Nix

I think that you are not correct when you talk about data sharing. AIQ is a software engineering company and Cambridge Analytica was a data analytics company. They built a product. If we go back to my rather trite analogy of the car and petrol, they built the car and we put the petrol in. The petrol that was put into the car was not raw data. It was modelled data, so they would not have had access to the raw data. The only way that they could have had access to that data, other than in the work that they were undertaking for us, would have been if they had taken a copy of those data illegally, and obviously against the contracts that they were bound by.

Chair

The raw data would not be much use to them anyway, would it? Their job is not to handle the raw data; it is to take the profiles that have been created by the algorithms your company worked with, and then turn those profiles into customised audiences, and then to execute advertising campaigns. That is what they do. To do that, they have to have access to all the profile data.

Alexander Nix

That is not what they did. Their role in regard to Ripon was largely to maintain the platform and to work on the platform and to ensure that it is scaled, that it did not have any bugs, and that if the client requested a particular functionality, they could then do that. Their role was very much a functionary role in order to make sure that this platform that they had built for us was a usable tool for clients. Yes, that is it, really.

Chair

If that is the case, then—

Alexander Nix

That is the case.

Chair

What surprised us is evidence Chris Vickery has given to us from the AIQ files he found on GitLab, which shows evidence of employees from SCL and employees from AIQ having access to the same datasets. We will see if there is more evidence of that, but that would seem to sit at odds with what you have said to us, because there would have been no need for that if people were not working on datasets that were being used in common.

Alexander Nix

I do not really understand the thrust of the point that you are making, but maybe I can colour this in in a different way. It would not be unusual for us as a firm to undertake data analytics on behalf of our clients where we would be working on their data. Sometimes we would take receipt of that data to work on it, and sometimes we would work within a client’s own firewalls—say, a bank or somewhere where the data is more sensitive. Just because we are working on clients’ data and, therefore, we have access to the data that we are working on, does not mean that we have a right to that data, we can make a copy of that data or otherwise own that data. It simply means that we are a contractor working on clients’ data, and that was the relationship with AIQ.

Chair

There certainly did not seem to be any policies in place in terms of the handling of data, and you talk about their not having access to raw data that were gathered by SCL. Just for the record, can you explain what that raw data would include, what datasets? I know you have given some talks and presentations about the methodology that Cambridge Analytica has used, but what sort of raw data would be processed by SCL?

Alexander Nix

I have just explained to your colleague—and I explained when I was here last time, I think, if you would like to check my testimony—that this is largely consumer and lifestyle data. This is data on, for instance, loyalty card data, transaction data. This is data that pertains to lifestyle choices, such as what car you drive or what magazines you read. It could be data on—I don’t know—consumer habits, and together with some demographic and geographic data, and obviously the voter data, which are very important for US politics.

Chair

Facebook. Do not forget Facebook

Alexander Nix

Obviously there is the GSR dataset, but I think you need to understand that those data, in the grand scheme of things, were really a very, very small component of the larger dataset.

Chair

We are going to move on to that subject now.

Paul Farrelly

Just before we do, can I just clarify one thing? These newspaper reports have been very, very recent. When you appeared in February—and so much has gone on since—you made the distinction between Cambridge Analytica and SCL Elections and the SCL Group as separate entities, apart from certain common shareholdings. It is the case that all of the SCL companies have now gone into either chapter 7 or administration.

Alexander Nix

That is correct.

Paul Farrelly

As far as the SCL Group companies are concerned, has it been the case that what has happened with all the media publicity surrounding Cambridge Analytica and SCL Elections is that the whole SCL brand has been toxified and, therefore, extinguished?

Alexander Nix

As I have already stated, it is certainly the case that all the SCL companies are in administration.

Paul Farrelly

The SCL Group was carrying out defence work, wasn’t it, traditionally?

Alexander Nix

That is correct.

Paul Farrelly

What has happened with your companies has made the situation so toxic that the shareholders have decided also to put SCL Group into administration.

Alexander Nix

This was a commercial decision based on the negative press and the allegations that have been made against the company, many of which have now been proven to be false. As a result of these, the board decided that the company was no longer a viable going concern and decided to put it into administration.

Paul Farrelly

Are those the different boards of the different companies?

Alexander Nix

Indeed.

Paul Farrelly

I just want to come on now to where I think you left off in your opening comments. It is regarding the relationship between your company or companies and GSR and Aleksandr Kogan. When you came to us in February, the Chair asked you this question: “Does any of your data come from Global Science Research company?” You answered, “GSR?” The Chair said, “Yes.” You then said, “We had a relationship with GSR. They did some research for us back in 2014. That research proved to be fruitless and so the answer is no.” The Chair then asked you, “They have not supplied you with data or information?” Your answer again was, “No.” The Chair then asked again, “Your datasets are not based with information you have received from them?” Again, you said, “No.” The Chair, “At all?” Your final response was, “At all.” Since then, we have had some rather conflicting evidence from Dr Kogan himself, from Christopher Wylie, about what data was supplied and used. Indeed, Mr Wylie described those data obtained via the Kogan app from Facebook as the foundation dataset of the company, which may have collected data on up to 87 million users. Also, clearly, there has been action by Facebook following these revelations or allegations. I think you were perhaps going to do it in your opening statement. Do you want to clarify or amend now the evidence that you gave in February with respect to data supplied by GSR and Dr Kogan?

Alexander Nix

Thank you. I am grateful for that opportunity. I certainly would like to continue with the very short statement on this, so I will do that and then we can move on to questions from there. Clearly, I accept that some of my answers could have been clearer, but I assure you that I did not intend to mislead you. It was my firm impression that Mr Collins’s questions were focused on whether we were currently in possession of data from GSR following the research they had carried out and which I was very clear in telling the Committee about. My belief was, and remains, that we deleted all such data licensed in good faith from GSR under that research contract, so I answered “No” to his questions on the topic. When I read the transcript, it became clear that Mr Collins was also asking about whether we had ever been supplied with such data. Of course the answer to this question should have been “Yes”, but my focus was on whether we still held the data from GSR Research, which is the issue that Facebook had been most concerned about. There was certainly no intention to mislead the Committee. The fact that we received data from GSR was already in the public domain from as long ago as December 2015, when The Guardian published an article detailing Cambridge Analytica’s relationship with Dr Kogan. There was and is no need to hide this fact. Indeed, on 20 March, three days before the Committee first raised this specific inconsistency, I appeared on “Newsnight” and made it clear that we had licensed research data from GSR. I also stand by my evidence that we do not work with Facebook data and we do not have Facebook data. The only Facebook data we ever received was via GSR and Dr Kogan, and that was subsequently deleted at Facebook’s request. At the time we had no reason to believe that data had been provided to us improperly by GSR.

Paul Farrelly

In short, you misunderstood the Chairman’s questions?

Alexander Nix

When you look at the transcript, it is very plain or it is certainly clearer what the Chairman was asking. When you listen to the video, and I have several times, the conversation is much less clear, and indeed we spoke over each other, and he did ask about whether we had data from Kogan. He then reiterated that point, and then he returned to that point. That was my understanding, and it was a genuine misunderstanding.

Chair

For the record—everyone will have transcripts in front of them—I did ask you. I said, “Does any of your data come from Global Science Research company?” You said, “GSR?” just to clarify. I said, “Yes.” You said, “We had a relationship with GSR. They did some research for us back in 2014. That research proved to be fruitless and so the answer is no.” I then said, “They have not supplied you with data or information?” You said, “No.” “Your datasets are not based with information you have received from them?” “No.” I said, just to clarify, “At all?” and you said, “At all.” There is no ambiguity in the exchange there, I do not think.

Alexander Nix

“At all”, our datasets are not—

Chair

“Datasets or information”, I said.

Alexander Nix

“Or information”, yes, are not informed by that data because it was deleted.

Paul Farrelly

Chair, that is exactly the same exchange as I have just read out, so Hansard will have it now in duplicate. He was mixing his tenses up and he misunderstood the questions, and you were talking over each other. Why didn’t you take the opportunity later just to write and clarify with your follow-up supplementary information?

Alexander Nix

I do not believe that when I submitted the supplementary information—and you can certainly correct the timeline—that inconsistency had been flagged.

Paul Farrelly

No, but you flagged it yourself.

Alexander Nix

I beg your pardon?

Paul Farrelly

You flagged it yourself. You have read the transcript and realised that—

Alexander Nix

I did not read the transcript until after it had been flagged by this Committee, so after whenever it was, 23 March or something like that.

Paul Farrelly

Still a few months back. You did not take the opportunity to write and say, “I want to clarify what I have told you.”

Alexander Nix

Mr Farrelly, I have agreed to come in here and talk to you in person.

Paul Farrelly

You have now given an account of it. You might say it is not up to you to check whether Dr Kogan was complying with Facebook’s terms and conditions, which he clearly was not, but the methods he was using were a neat trick, in a sense, because you could get friends’ data and, therefore, multiply greatly the data you got from people completing surveys. Was that a neat trick at all that Cambridge Analytica or SCL Elections or any of your companies repeated with respect to any other surveys that you did directly, which involved any logging in at the end on Facebook?

Alexander Nix

Unfortunately, as you will recall from my opening statement, there are some topics that, as much as I am wanting to talk about them and willing to talk about them, I am simply not able to because they refer to live investigations. Unfortunately, this is one of those topics. I think you are all aware that the ICO is conducting an investigation at the moment and so I am not able to speak to any matters in relation to GSR and to that project.

Chair

Just to make it clear, that is your answer, but the Information Commissioner has made clear to us, and I believe to your representatives, that it has no objection to you discussing these issues today.

Alexander Nix

I am sure it does not, but the Information Commissioner’s Office took advantage of my parliamentary privilege the last time I appeared here to use my testimony to this Committee as part of its application for a search warrant of our premises. While in theory this privilege exists, clearly it is open to abuse.

Chair

It depends what you say, I suppose.

Alexander Nix

It should not depend what I say. It is either a privilege or it is not.

Chair

It is not open to abuse. If they are interested in something you have said and they therefore wish to talk to you about it, that is a matter for them, but if you are declining to answer those questions, that is a matter for you, and if you are doing that under advice, I understand that.

Alexander Nix

Thank you.

Chair

Just to make it clear for the record, the ICO itself does not feel that you answering these questions would in any way jeopardise or prejudice its investigation.

Alexander Nix

The Committee should be reacting to findings from the ICO and not the other way around.

Chair

We are two completely separate bodies. There are very clear sub judice rules that apply and that prevent us from asking witnesses questions relating to matters before the courts. This is not currently a matter before the courts. We did, as a matter of procedure, seek guidance from the ICO itself, so that was its guidance to us. You have given your answer to Mr Farrelly. There is nothing more to say about that, but I think it should be clear for the record that, as far as the ICO is concerned, it does not have any concerns about you answering any questions on that topic.

Alexander Nix

Thank you.

Paul Farrelly

I just have one final question on this strand, but I think the refusal to answer the question will be taken in the record as meaning it was a relevant question to ask you, Mr Nix. Again, this is a question in relation to potential use of data, which you may feel you do not want to answer, and it is really just a question to try to exemplify your approach to data. You had a staffer from your companies working on data from Breitbart, which has a strong connection with Steve Bannon, I understand. Was any data gathered from that secondment to Breitbart, and, if so, how would those have been used by the company?

Alexander Nix

I am sorry, you are going to have to help me understand. You are saying that one of my former employees went on to work for Breitbart?

Paul Farrelly

No. You had someone seconded, I understand, to Breitbart.

Alexander Nix

I do not believe that we did.

Paul Farrelly

You did not?

Alexander Nix

I do not believe that we did. Certainly I cannot think of anyone who was seconded to Breitbart, and I am certainly not trying to mislead you in this regard.

Paul Farrelly

No. It was just a suggestion.

Alexander Nix

I have not heard about that before, no.

Paul Farrelly

Then you cannot answer it.

Chair

Before I go to Rebecca Pow, Chris Vickery has kindly just sent us a message in relation to the AIQ data, just on whether they had access to raw SCL data or not. He says that, based on the work he has done looking at the AIQ files, it was almost certainly accessible—the usernames, passwords, network locations were all present, and that anyone who had access to the AIQ GitLab store could have had access to all the SCL data that it had worked with. I note as well that from the email exchange that we published there is a document given to us by Chris Wylie that relates to the Bolton Super PAC work as well. There is an email that was sent to AIQ that lists a series of next steps to be undertaken during the campaign, and in one of the items of that it says, “AIQ needs two days to push all information made available to them into a form that can be accessed and used in the platform. Therefore, all currently available data would need to be ready and sent or provided to them by the close of business on Saturday, UK time, so they have the weekend to do what they need to do.” We know as well, obviously, from Facebook that it found evidence of shared audience files on SCL for SCL and AIQ campaigns that it had run where there was 80% correlation in the members of those audiences, and AIQ said to us that it had been given audience data by SCL. Are these things that you were aware of? If they are, I am not sure that the answer you gave to us was truly reflective of the nature of the relationship between AIQ and SCL.

Alexander Nix

What I explained to you, Mr Chairman, was that the example that I gave you was that it would not be unusual for SCL or Cambridge Analytica to work on a client’s datasets, whether it was a bank or a corporation, and to have access to the data while we were working on them, but that did not entitle us to have any privileges over that data or any wherewithal to make a copy or retain any of that data ourselves. The relationship with AIQ would not have been dissimilar to that, where, as a subcontractor who is brought in to assist us on projects, they would have had possibly access to some of the data, as it is indicated there, whether those were model data or otherwise. Again, that would be covered by the contract relationship that we had with the—

Chair

That is fine, Mr Nix. You said there they would have had access to model data or otherwise, including raw data. It is clear that they did have access to raw data.

Alexander Nix

I cannot answer that with certainty but, again, I do not want to mislead you, because I did speak to my technical team, my data team, prior to this hearing, and they assured me that there were no raw data that went into the Ripon platform. Again, I can only defer to their expertise. Again, this could be a topic that we could circle back to you on.

Chair

Yes. Again, there is a difference between raw data going into the Ripon platform, which they almost certainly would not have done, and raw data being accessible to AIQ, which they certainly were.

Alexander Nix

Based on what Chris Vickery says.

Chair

Yes, and I think based on what is in that email, based on other evidence that we have seen, that they are working on common sets of data.

Alexander Nix

That email does not suggest raw data. It could certainly suggest model data.

Chair

What we have seen, both in Chris Vickery’s and other evidence that we have seen from the AIQ files that we have had a chance to look at, is shared access to raw data. Chris is saying that it is certainly technically possible. He said it would have been easy for anyone to do that.

Alexander Nix

The premise remains. Let us agree that we need to establish what data they had access to, but, regardless of that, that does not give them any rights to that data, as it would not any contractor working with anyone else’s data.

Chair

I am not saying they did. I am just saying you said earlier on that they had no access to raw data, and that seems to be a matter of dispute, and I think even you are now saying that that is not a statement you could stand by.

Alexander Nix

I was told very firmly and unequivocally that no raw data was put into Ripon, and so I stand by what I was told. What I have agreed with this Committee is to circle back to you and confirm this point.

Chair

Who puts data into Ripon, when you are working on that? Will AIQ put the data into Ripon?

Alexander Nix

Data scientists, I believe.

Chair

Who do they work for?

Alexander Nix

For SCL Elections or Cambridge Analytica. Mr Chairman, sorry, do you mind if we take a break for a few moments, please?

Rebecca Pow

Could I just ask one very quick question because I am afraid I have to leave the Committee for a second? It will be very quick.

Chair

I am very happy to have a very short break. If Rebecca Pow would ask her question—I believe it is quite short—and then we will take a break.

Rebecca Pow

Just a very short point. One was you said, Mr Nix, you were going to refer back to your team. I just wondered what technical team that was, because SCL is in administration, is it not?

Alexander Nix

That is correct, but I still have a relationship with some of my colleagues.

Rebecca Pow

Just for clarification, that was all. I was just interested.

Alexander Nix

I can obviously ask them as a friend for their advice.

Rebecca Pow

It is just a few points relating to your last appearance here, where we dealt a lot with Facebook and those surveys on Facebook. This is where a lot of our discussion began, which seems like a long time ago now. At that first meeting, where we talked about those surveys that had a Facebook login at the end, you said that if they had a login, it did not mean that your personal data was then going to be accessed by Cambridge Analytica, but Brittany Kaiser told us that the purpose of the surveys was to gather information and that by completing it with your login, Cambridge Analytica would also get access to your data. I just wondered whether you might like to change your mind and amend your comments about that?

Alexander Nix

I also was surprised by Brittany’s answer. She is also a non-technical person, as am I. I did discuss this with my technical colleagues and they assured me that my original testimony reflected what was the case. Therefore, what I would like to do, because obviously I do not want to run the risk of misleading the Committee, is to agree to circle back to you on that point, but my understanding is that no data—

Rebecca Pow

It would be quite nice if you could just come up with an answer now, rather than us having to get back to you again.

Alexander Nix

My understanding is that no data was collected, but again I would like to confirm that because obviously we do not want to have a miscommunication.

Rebecca Pow

That is a pity. I want to clarify one other thing. You told us at that first meeting that Cambridge Analytica did not work with Facebook data. Those were your exact words. I think they might have been to me, actually. Do you want to change your view on that? Would you just like to put on the record, in the light of everything else that we have heard since, whether you have changed your mind on that?

Alexander Nix

Sorry, I must stop myself from continuing because, again, this is another topic that unfortunately falls under the investigation by the ICO, so we are going to have to let that investigation play out before I can address that. Sorry.

Rebecca Pow

My final point is something that you mentioned earlier. You said, “Very little data would have been collected anyway,” but Brittany Kaiser referred to 87 million people having their data gathered in this way.

Alexander Nix

I can address that because this is in the public domain, and I urge you to revisit the testimony of Dr Kogan because I think his testimony was a very fair reflection of what happened, and that is in contrast to people who were not necessarily so involved, such as Brittany Kaiser. I think what Dr Kogan made amply clear was that while he collected data on potentially 87 million people and a great deal of data, Cambridge Analytica or SCL Elections only received data, and a very small amount of data, on about 26 million or 27 million individuals in the USA, and that the only person to receive the entire dataset, not just in its entirety in terms of all the people that it included but also all the data that it included—which I believe Dr Kogan said was about 96% more data than that that Cambridge Analytica or SCL Elections had received—was Christopher Wylie for his company, Eunoia. Before we have a break, this would be a good opportunity to talk about that. A lot of the allegations—in fact, I am going to say 90% of the allegations—that then gained traction in the media have all come from a single source. They have all come from Christopher Wylie, and it is really important that this Committee understands that and understands the motivations behind this and what was going on at the time. If you will permit me, I would just like to give you a little bit of background because I think it is very relevant and extremely important, where so many allegations have been made and so many of them have now been proven to be false, that you understand what is going on here.

Rebecca Pow

Can I just interrupt you there? On Christopher Wylie, when you were talking about him just now, I think you suggested he was just on some kind of simple student’s visa, but he told us that he was on an exceptional talent visa. You are dissing him with all these comments but he was clearly, in his own mind, a very intelligent man.

Alexander Nix

No one has suggested that Mr Wylie is not intelligent, least of all me. I stand by the fact that when he worked with Cambridge Analytica and SCL Elections he was on a tier 2 student visa. The reason I stand by that is because I applied, or my company applied, for that visa. If he went on to change his status at a later date after he left me, I cannot speak to that. What I can tell you is that Christopher Wylie was at the company for 10 months. He was instrumental in defining the relationship with Kogan and for being the architect of that particular programme of data-gathering. As the company grew and started to undertake business with the methodology that he felt that he had designed and was the architect of, he grew more and more resentful, to the point where he was openly discussing with clients that he wanted to replicate SCL Elections or Cambridge Analytica “without Nix”—i.e. without me. He also stated things like he wanted to create “the NSA’s wet dream”. He had his own ideas about a company that he wanted to build. At the time that he left us, which I believe was in around June or July 2014, he had already set up a company called Arg.us, and he developed a pitch that he took to Silicon Valley to raise $15 million or $20 million for some 20% of his company. What was clear was that he was totally agnostic about where the money came from and even courted Russians, stating that he found the idea of working “crazy evil Russians” quite intriguing. One San Francisco-based investor who spoke to Wylie about his start-up in 2014—please.

Chair

No, sorry, Mr Nix. I do not see how this is relevant to the questions.

Alexander Nix

It is very important. It is very relevant because most of the allegations that have been put to this Committee stem from one source. They have gained traction in the media. If you do not understand why these allegations have been made, they are going to be given unnecessary credibility. It is important—

Chair

Your assertion is not true that the only evidence we have about the work Aleksandr Kogan was doing and the significance of it comes from what Christopher Wylie has told us. It comes from documents that the Committee has received, contracts that you signed with GSR that bear your signature that set out very clearly what the work was for.

Alexander Nix

Sorry, that is not what I said. What I said was most of the allegations that have been made against my company come from one source.

Chair

The issue you totally glossed over when you came to us. You could have chosen to give a more honest and more full answer to the questions I asked you about your relationship with GSR when you gave evidence before, and you did not mention any of it, and you wanted us to believe that there was a very small, as you said, short-term academic project that had gone nowhere. Yet what we have is not Christopher Wylie’s opinion but facts, documents, detailing the work that was to be done, what was to be paid for it, what it was to be used for, not on an ad hoc basis but commissioned to support elections being run in specific states in the United State of America and how you would remunerate it. That is what we are basing our questions on, and our questions are to you about whether you were clear with us, not just what Christopher Wylie has said, but by evidence that you do not dispute and that is in the public domain.

Alexander Nix

You will remember that at the time I appeared before you the issue of Facebook data, the issue of GSR and Kogan, was not a sensational news story.

Ian C. Lucas

You did not tell us about it. You did not tell us about Facebook.