 Good afternoon everybody, you're all very welcome here today to what I think will be a very interesting, it's certainly a timely event on political advertising online and effective data regulation. But before I begin, can I ask you to just sure that your mobiles are switched off and to remind you that this session, the speaker's time is on record, but the questions and answers to Chatham House rules apply. So today we've got a very distinguished group of panellists who will be looking at this, as I said, really topical issue but really one that's very important. It's not only that we've had a lot of discussion about it, but there's also issues with ongoing investigations in the US, in the European Union and in the UK. So our panel today will discuss the proposed regulation in this area, the long-term viability of data-driven business and the model that it uses, the future of data privacy in general, and the threat that that currently arises. I suppose we're all conscious that the referendum is on the 25th of May, but if this is not just about this referendum, it raises issues really in a much more general sense, and that's what our panel will be looking at. So our first speaker today is James Lawless, TD. He's TD for Kildare North and spokesperson on science, technology and development. But before he made that great leap into politics, he had a background in IT, financial services with focus on compliance and regulation, including data protection, corporate governance and conduct of business. And added to all of that, he's a barrister. So James, we look forward to your presentation. Thank you very much. Thank you Joyce, and look forward to hearing everyone's questions and comment from you later on. I think what Joyce meant about the introduction was I used to have a real job before I was in politics, but anyway that's what I do now. I suppose one of the most interesting books I've ever read is Guns, Germs and Steel, a book by Jared Diamond, historical profile that by civilizations fell, others rose, why the Incas were defeated by the Spaniards, you know, why the Setsia Islanders are conquered by James Cook and the British and all the rest of it. One of the findings of that book is that technology played a huge part in all these conquests. If you were a better donor of Sany technology, be it steel, be it guns, be it crops, agriculture, the more likely you were to survive and the more likely you were to be dominated by a more powerful opponent. The same goes for elections today. I think elections, free elections and democracy is under attack. I think it's under attack in Ireland, it's under attack in the Western world, and further afield. So I think it's really important that we defend ourselves and we take steps to understand what's happening and guard against the Setsia Islanders. There are evidences there. We're talking about social media campaigning, online campaigning. I suppose I was asked why legislation was to talk about a few moments to try to safeguard against this in Ireland. Where did all this come from? Some people say what is this happening for real? Is there any evidence? Last week the US Congress saw three and a half thousand fake ads, Russian ads that were unveiled to it by Facebook released, that's one of the two by the Iceberg, Twitter had 50,000 Russian fake accounts during the US elections, during Brexit the Guardian newspaper, which isn't really great work on this, highlighted that of the top ten influencer Twitter accounts in the Brexit debate, one of the top ten, at least one of the top ten was a Russian bot account. So Russians have certainly been very busy and it's not just them, it happens closer to home as well. Some people say what is this all very esoteric, do we really have the Russian intelligence agency working in Ireland or other such threats? Well it doesn't have to be one source, I'm sure people here remember Michael Sean Gallagher ran for the presidential elections 2011 before social media was popular or profitable, arguably the whole election hinged on their fake tweets. So this is not new, this is not esoteric, it's not international, it's happening already in our elections. We have the referendum on at the moment but we all know and we have a huge focus on that and groups like Liz from the Transparency Reference Initiative here today have been doing great work on highlighting the potential of uses and some of the activities in the referendum space with fake news and with different groups and different accounts being manipulated and opinions being manipulated. So we're very aware of that. We've seen a trend in the last maybe week where multinational corporations, Google and Facebook in particular are making their own rules so they're very, very late in the day saying we're going to ban all ads or ban international ads or whatever it is. I think this is a step forward but I think it's probably too far in the case of Google but I think we shouldn't be leaving it to global companies to decide what our electoral laws should be about. I think it's also important that we're very balanced with how we do this that we make sure all sides get a fair hearing or a fair dishearing if the case may be. It was interesting when Trump was elected, one of the first things that he said and I confess I'm not a Trump supporter so I might be a little bit biased but he came out and talked about fake news being a problem. Like this guy just put an election with us a fake news. It's not unique to any one side in any particular context I suppose is the point and I think it's important that as legislators or as potential regulators that we bring people with us across the spectrum. It's not seems being done from one side or another's advantage. That's something that applies everywhere. But it's very serious. It happens, if we look at the patterns that we saw in the American elections some of the types of advertising that we saw, much of it was divisive. A lot of the fake advertising wasn't supporting a particular candidate or a particular campaign. It was actually about voter suppression, about selling dissent, about fatherizing issues, race, immigration, social issues, kind of what we call wedge issues. And they were being driven up by the opposite sides in many cases that were being promoted by the same agents. We went to Seoul Division and tried to manipulate people and I guess what they were doing on we know was because of people who were being micro targeted because they had surrendered their data to the companies, to the platforms which then could be resold to advertisers in the case of Cambridge Analytica and the Kogan app in a dishonest way because it was gathered with their consent but in some cases people actually did give consent and did fill in forms and did fill in surveys which led to their data being manipulated. So the hope streams, thoughts and visions were out there in their houses to micro-target. I mean the discussion about this story you're on is saying it comes up a lot, is that such a bad thing? What's the bad about micro-targeting? And one way we've always had it, everyone had a supermarket loyalty card. It was 20 years ago at first, it was a super clean at the time where Tesco had their cards, we swipe every week and maybe get a discount the following week and a round of our choice. If we fly it on an airline we have a frequent flyer, frequent air miles type arrangement. So again it's not new and it can be beneficial because we should be free and fair to consent. We should know what we're doing, we should know what we're signing up to and we should be brought into that. But there are difficulties with it. The first thing is because you don't have any regulation in that space we have the equality of arms which means that the highest bidder can actually buy a democracy. So the election is available to buy, whoever can put most money on the platform, whoever can mine that data to the widest extent, whoever can roll an average high as much to the widest extent is plotting all the rules around campaign spending, campaign donations, campaign finance and the reason those are there in the first place is because we as a society, as many societies have a principle that you cannot buy an election that the highest bidder of the highest resources should not be able to just spend their way into it. The second issue is transparency. I think people have a right to know who is trying to influence them. If somebody is trying to be they small or big or bidding high or low if somebody is trying to influence you in an electoral outcome if you change the constitution or electing a new government or a new president you have a right to know who that is. And a lot of these kind of false flag ad campaigns where people pretend to be someone else maybe it's money, the name of their opponent they can pretend to be their opponents clothing or they can run something very dark or an issue based campaign that will get people thinking without associating with a particular group which is very insidious. The other thing I think why this is a problem is not so much a regulatory issue it's more a democratic issue we have to be aware of echo chambers. So if we have a situation and people like Havermass and it's a well established factory democracy that public square discourse be it Hyde Park or Speakers Corner or Twitter or wherever else we go or even here today it's positive to have opposing views it's good to have questions that don't agree with you it's good to have conversations because that's how we evolve so we progress if we come into a concentric circle if we go into a market targeting into a particular area then we cast the net too narrowly so we begin to believe we're all on the same side and become more polarized, more extreme and I think that's a negative for democracy. So what do we do about all this? How do we combat this? In an Irish context we have the Advertising Standards Association we're not responsible the BAI come out and control social media CIPPO it's outside their remit the Referring Commission has no role to play the government up until recently have been I suppose not engaged on the issue and pretty much everybody so it's part of the civic society and people like myself and Amy and Ryan and other opposition politicians have been trying to raise these issues and make some noise but that's not good enough I think we need legislation we need regulation and we need to introduce the force of law the other thing we need consistency because when we're talking today about social media I'm talking online we have other channels and we're not the moment by an atom to be a radio now I'm not sure that logic applies anymore the thinking at the time was prohibitively expensive with the advent of local and community radio stations it may no longer be the case I think that the very least it's worth revisiting that is the broadcasting ban still so logical we have roots around traditional advertising of political advertising if you put up posters, if you put up leaflets you must have a name, notice, the publisher, the sponsor on that material again there are 30 offenses on the statute books dating back to 1992 that's a very different world certainly pre-international, pre-blogman, pre-social media but also 30 years ago even outside of technology is not worth a look again now and there are quite a few offenses you can go to jail I think for 5 years if you fail to put the right printer on your poster now nobody actually in the real world does and when I inquired recently there's actually nobody enforcing that so we have 30 offenses on the statute books that are essentially obsolete if you put up a cap on election posters I wouldn't ban them but I've capped them there's no need to pose them to blight every election so there's a couple things we can do it is important that whatever measures we take are ideological neutral I talked about bringing people on board and having people with us so that they don't favour anyone's side it's also important to be technology neutral and we saw with the Google ban which came into place in ads in the referendum which kicked in last week the same ads on similar spaces so any regulation cannot be specific to Facebook or Google or Twitter or any provider technology moves quickly today's Facebook is yesterday's Bebo so it has to be technology agnostic across the board I think it can be done we also need to balance free speech I make sure people do have a say both that we know who they are that we have some measurement and some regulation of who they're doing so a final thing I think that is worth looking at is fact checking I think the journal that I read Facebook has partnered with them recently in terms of fact checking fact checking sites can be very helpful for the advent of fake news and people can challenge a particular claim and have an expert review but I don't think there are panacea because one person's fact check has another person's biased opinion so I think it's really important one thing I mentioned before the current rule is applied 30 days before a vote we all know that 6 months before a vote people are campaigning, 6 weeks before a vote even the current very modest limit we have only kick in months to go I have legislation before the door the opposition party is all when it came before the house it's moved very slowly the government I understand are coming on board I won't be having some time for the referendum but we have many more votes referendum expected over the next 12 months we have local elections, European elections general elections probably over the next 12 months as well it's really important to get it right we've looked across the international scene the Honest Ads Act in the US is a similar piece of legislation speaking to American colleagues recently the Democrats are advancing that the Republicans are blocking it they probably want to advance in the near future Germany has the time to do this they've taken a slightly different approach they've tried to censor free speech there's a free speech issue there they're talking about hate speech and how do you define that more difficult to do so at the American companies they're just looking at this Canadians have tried to do but more has given up so they can't do it so it's possible, I think it would be great if we did Ireland could actually be the flag bearer here we have my legislation with amendments and the press party doesn't have to be all about my proposals but if that legislation gets through the house which it may very well do later this year maybe the flag bearer and actually have the opportunity to get it right and just to finish on a I suppose a what if and again it's a positive story for Ireland could have been a positive story we all know about Cambridge Analytica and Alexander Cogan and the app that was exploited and I think the normal kids enrolling is that 70 million people now use this data that was mined and put into US student various electors afterwards the Irish data collection commissioner in 2012 identified a problem with third party consent on Facebook, they identified an issue with the app and they would have collected data the Irish DPC flagged that to Facebook didn't that, I think they had it two years later if the Irish DPC had been listening to 2012 Alexander Cogan couldn't have ran his app out of Cambridge University in 2013 Cambridge Analytica couldn't have mined in 2014 Brexit may not have happened in 2016 and maybe Donald Trump would still be doing hotels so look at this listen to Irish people I suppose is the message and maybe we could be a trailblazer this space that's very much much James now we'll move on to our next speaker Fred Loog and Fred is the principal with F.P. Loog and the solicitors in Dublin who specialize in technology, intellectual property, data protection and information law Fred himself specializes as you'd expect in technology, intellectual property, information law and commercial law but he's the leading information lawyer advising on freedom of information requests access to environmental information data protection and privacy so Fred we look forward to your presentation, thank you thanks very much and thanks to the institute for the invitation to speak so suppose in the late 1950s the birds stopped singing and people began to wonder why so in 1962 an environmental scientist called Rachel Carson wrote a book called Silent Spring that documented the adverse effects on the environment of large scale use of pesticides which had been developed following the Second World War with the growth of the chemicals industry she also documented the disinformation campaigns that these industries used to spread and the fact that public officials basically accepted industry claims without question and the pattern revealed by Rachel Carson in the 1960s is a pattern that we all know has played out time and time again where big industries would do something harmful so we just have to think about tobacco semiconductor manufacturing that would contaminate the natural environment and they would spend more effort and time covering it up and lobbying to prevent the harm being addressed than they would have actually had to spend to do a right in the first place and the great thing obviously it's not great to pollute the environment I don't think anyone can disagree with that but the great thing about Carson's work in the 60s and the early 70s was that the US government and particularly a republican administration took note of what had happened and brought in a new era of environmental regulation and it was Richard Nixon for all his his sins who set up the Environmental Protection Agency in the US and that took away from the US Department of Agriculture the dual role of both regulator and promoter of the chemicals industry and agriculture and put into an independent body and the other thing that resulted was that tools regulatory tools were developed most notably something called the environmental impact assessment where instead of building your factory and then seeing who died and fighting the rest of the people until they died in court that you would assess the environmental impacts of certain activities and developments that were likely to have an effect on the environment in advance so this radical idea that you actually make sure that you don't harm people before you do something risky had to be invented and had to be forced through by the Americans and now this concept spread around the world we have it in the EU law we have it in the the Arhus Convention and things like that and that's the norm and people accept that and buy into that so now we bring ourselves to the 21st century and the mental environment is under threat so the revelations about Cambridge Analytica that wealthy individuals are using shady companies to gather detailed personal information about huge proportions of an electorate and spending that money on very well crafted ads that are designed to influence people at an emotional or at a psychological level both in the US and in the UK as James has talked about in terms of the Brexit debate so in my view this Cambridge Analytica revelation which actually came out I think on the day that the data protection bill was published which was great timing actually is the silent spring of the digital age the birds have stopped singing in our heads the music has stopped playing in terms of social media and we're looking at the carcasses of birds on the ground for want of a better word I think it's clear to me and we're talking about overlaunch but it is clear to me that this type of unregulated use of personal data by powerful organisations can be and has a direct adverse effect on the political sphere but you don't have to look hard to see other examples even today there was a further revelation about the public services card which the state knows or ought to know is an illegal biometric ID card the police seem to be rolling out a national surveillance system on the roles network using automatic number plate recognition and this is despite three or four very very definitive and unambiguous decisions of the European courts that mass surveillance is illegal and that mass processing or unjustified mass processing of personal data is fundamentally incompatible with European values and European fundamental rights but not only that in my own practice we're seeing children being targeted being profiled for junk food advertising gambling alcohol I don't think there's a great leap for anyone to accept that these things are or could be happening companies that were set up with the noble purpose of connecting everyone or organising the world's information have now mutated into vast surveillance tools which are at the disposal of anyone who has enough money to access them and the vainglorious attempts of voluntary self-regulation aren't fooling anyone to be honest so the mental birds have definitely stopped singing and the question is where do we go from here so I think we have a choice we can continue with self-regulation and hope for the best but risks subverting our democracy risk turning our children into gambling addicts and making them obese or we can say that society is better than this and as James has mentioned that we need to have some form of regulation I think the model is there from environmental law which I also practice in and some of the tools are already to be seen in the GDPR so we already have now a concept of a data protection impact assessment we already have a concept of prior checking with the regulator in my view I think what we're going to see is that there will be a type of planning permission for certain types of personal data processing so not only will our regulator be an enforcer, they will also be a permitting authority so that before you do a Cambridge Analytica type of scenario you have to go to an independent state regulator you have to make information available to the public you have to give them the opportunity to comment on it and there has to be a published decision to make sure that whatever plan whatever proposal is carried out within the terms of the of the of the permission so that's my proposition I think the stakes are too high for it not to be regulated and I think my prediction is that there will be it will follow the same path as environmental regulation in the future Thanks very much Fred now with our last speaker Professor Barry O'Sullivan who's an award winning academic in the field of AI, data analytics data ethics constraints programming and operations research a wide, wide area but Barry I think has a very well not just a very good reputation, not only for that area but in research he managed to get together over 200 million euros I know with colleagues but also was named Science Foundation Research of the Year in 2016 his day job if I could say is the director of insight centre for data analytics in the department of computer science at UCC where he holds the chair of constraint programming he's also an adjunct professor in Monash University in Melbourne Australia and currently the deputy president of the European Artificial Intelligence Association which covers about 30 countries and about four and a half thousand people Barry your presentation too I really wish I could prepare some notes now I suppose we're living through an absolutely transformational time the iPhone basically didn't exist in 2007 well it just came on the market in 2007 we had no idea what this technology was going to have what kind of impact it was going to have on us as citizens we never imagined a time where people would be sharing photographs looking for people to like them if someone walked down a convent street holding a photograph of their dog and kept on putting it into your face and said do you like it? you probably arrested so our behavior online is totally different to what it is offline I suppose the other thing that's important to realize is that the corporations that are in operation in that sphere are bigger than any country so there are I think about two billion monthly users of Facebook there are only 1.2 billion people in China there are about a billion people in India the only thing bigger on the planet in terms of membership is Christianity in fact I'm not quite sure if any particular Christian church is bigger than Facebook it's a huge thing and I suppose over the last 8 to 10 years we've given away massive amounts of information about ourselves and of course we've done that very willingly because we enjoy seeing those cat videos we enjoy the fact that we can see what our neighbor is doing we like to project a very beautiful image of how our lives actually are and because my life needs to look as good as your life then I don't tell you all the bad stuff now I suppose factored into all of this is the fact that technology has moved to the point where we can do incredible things with technology so in the last 10 years maybe more than that we have built a self driving car so some people say these things are not safe they're actually statistically safer than human beings driving cars but we still don't like the idea of a car killing a person just to put it into context human beings driving cars kill about 1.2 million people every year all war, violence and drugs kills about 1.6 million people every year so we almost kill as many people driving our cars as every form of violence and sort of egregious behavior we can imagine yet we still don't like that technology we have a funny relationship with technology there's a great book written by Kathy O'Neill called Weapons of Math Destruction and it's a book about how big data and there's a lot of things we can talk about that large amounts of data images and so on can be used for all sorts of negative impacts on us now I suppose we know what we can do technically I suppose the question is is it really true that information online is actually affecting how we vote and how we behave in elections and I suppose the scientists are out on that so we don't know for sure so we don't know in scientific certainty whether that is happening or not but I think as James said in his address the evidence would suggest that that really is happening and I suppose the question is what would it take for us to actually accept that this is a possibility and that we should actually do something and I think seeking transparency is one very proper step so I think that's something that we really should do so I think I'm not a member of UNIFOR but I think it's a very fine piece of legislation I think it's the first step on a great path I suppose in terms of in terms of what is this data revolution I think Facebook and Instagram all the social media companies they provide a service to us they connect us in some way they give us free services but of course the currency in which we pay is our data and so the purpose of these platforms is essentially to get as much information out of us as possible so that they can basically advertise to us so that is the business model there's no great secret in that and I suppose what's interesting about the Cambridge Analytica story is that it's a very recent time and it's only very recent times that people have sort of sat up and taken notice because of this data set that they got true what we regard as improper means but in fact what Cambridge Analytica does every single day of the week is simply and I'm not a sympathiser or supporter of Cambridge Analytica but what it's doing is just simply that they have been designed to there are two billion monthly users anyone of those can by advertising for whatever purpose and they can transmit that message to those two billion people if they so wish to pay for it that's what the platforms do and I suppose one thing we should bear in mind about these social media platforms is that the currency is data the people who are trying to personalise to you are not necessarily trying to upset the election they're not necessarily trying to turn the referendum some of them are but lots of them just want you to engage in the platform to reveal more about yourself and so I suppose we need to sit back and sort of think about what are the motivations of these sorts of people and how do we police it and it is very very difficult when the internet was imagined as this sort of utopia where every single user was exactly the same as every other user and so no allowances were made for people who were young so the children for example as Fred mentioned no special allowances were made for children children are on the internet in exactly the same way as we are there was no concept really of age appropriate content so you know some of you may have seen not mentioning any particular groups or positions but you know if you're using YouTube say two or three weeks ago you've been seeing lots of advertisements that were very strong content and you know the conversations were taking place amongst children 13 year old sees the thing and talks to the age about what's that about and all of a sudden having a conversation about something that you really wish to have a conversation about at all so the technology hasn't actually cut up in some sense but yes this incredibly powerful tool is our fingertips and the questions are well how can we possibly begin to regulate it and is it appropriate to regulate it and I suppose what I would argue is that we need to take a very prudent approach in terms of how we deal with these sorts of things and I think transparency is exactly that it is a prudent approach so it sort of sets out a set of rules and guidelines what we have to actually figure out as well how do we now enforce it and I suppose here there's something that's very very interesting we just talked over lunch about a project that the reference transparency people are doing that's very manual a lot of the policing that's done on the internet by many of the social media companies is actually manual they have thousands and thousands of people who just look at stuff and say well actually that looks bad we're not going to allow that one we're going to allow that these technology companies still haven't caught enough with some aspects of this technology that's something we really need to bear in mind so I think we're living through an era of massive transformation we have built a tool that is bigger than any nation states that we've ever seen on the planet it is tremendously powerful there are all sorts of objectives and goals and agendas some of them are political some of them are just mischievous some of them are just simply well I want to commercialise your data a bit more so please give me more data and we have to understand that and I think it's a really challenging space it's also a space where it just simply hasn't lived long enough for us to really understand its effects so I suppose in that sense we should be using you know common sense judgement whatever to I suppose to think about well how should we start controlling or regulating in some way that doesn't get in the way but is protective and I suppose ultimately when it comes out of elections you know these are about these affect people's lives in a particular jurisdiction and frankly in my view the only people who should have any influence over an election are the people who are directly affected by the consequences of those and so any legislation that sort of protects that I think is a good thing so thank you very much