UNCORRECTED TRANSCRIPT OF ORAL EVIDENCE
To be published as HC 729 -iii

House of commons

oral EVIDENCE

TAKEN BEFORE THE

Culture, Media and Sport Committee

Online Safety

Tuesday 19 November 2013

Simon MilNer and Sinéad McSweeney

Tony Close and Claudio Pollack

RT HON Damian Green, Mr Edward Vaizey and Claire Perry

Evidence heard in Public Questions 125 - 222

USE OF THE TRANSCRIPT

1.

This is an uncorrected transcript of evidence taken in public and reported to the House. The transcript has been placed on the internet on the authority of the Committee, and copies have been made available by the Vote Office for the use of Members and others.

2.

Any public use of, or reference to, the contents should make clear that neither witnesses nor Members have had the opportunity to correct the record. The transcript is not yet an approved formal record of these proceedings.

3.

Members who receive this for the purpose of correcting questions addressed by them to witnesses are asked to send corrections to the Committee Assistant.

4.

Prospective witnesses may receive this in preparation for any written or oral evidence they may in due course give to the Committee.

Oral Evidence

Taken before the Culture, Media and Sport Committee

on Tuesday 19 November 2013

Members present:

Mr John Whittingdale (Chair)

Mr Ben Bradshaw

Angie Bray

Conor Burns

Tracey Crouch

Paul Farrelly

Jim Sheridan

Mr Gerry Sutcliffe

________________

Examination of Witnesses

Witnesses: Simon Milner, Policy Director, UK and Ireland, Facebook, and Sinéad McSweeney, Director, Public Policy, EMEA, Twitter International Company, gave evidence.

Q125 Chair: This is the further session of the Committee’s inquiry into online safety and we have three panels this morning and I would like to welcome our first witnesses, Simon Milner, the Policy Director of Facebook, and Sinéad McSweeney, the Director of Public Policy at Twitter.

Q126 Tracey Crouch: Good morning. I will start by asking if you could outline what it is that you already do to prevent harmful or illegal material online.

Sinéad McSweeney: In terms of illegal content we have a zero tolerance for any child sexual exploitation material on the platform. So when that is reported to us, we review those accounts. They are suspended and they are, in addition, reported to the National Center for Missing and Exploited Children in the US and they then have channels of communication with the relevant law enforcement agencies, who can take action both on the offenders but also ensure that if there is a child at risk, that they can be identified, located and rescued as well. We also have relationships with law enforcement and police in different countries. So we have good working relationships with the police here in the UK not only so that they are aware of our processes and procedures in order to request information of us and work with us, but also so that we would participate in some education and awareness raising with them about how the platform works and how to contact us when they need to, whether that is in an emergency or when they are investigating crime.

Q127 Tracey Crouch: What about harmful content, before I move to Simon?

Sinéad McSweeney: Yes. Sorry, did you say Simon?

Tracey Crouch: No, no.

Sinéad McSweeney: Okay. In terms of content generally, clearly, there are rules as to the type of content that is allowed on the platform. So when content is reported to us that is in breach of our rules, we take action on those accounts. In addition, because there is such a diverse range of discussions and contact that take place on the platform, we also have built into the product the ability for users to label the media that they may be uploading as sensitive, and the default setting for every user is that they do not see sensitive media without a warning.

Q128 Tracey Crouch: You do not have a report abuse button though, do you?

Sinéad McSweeney: We do, we do.

Tracey Crouch: Okay.

Sinéad McSweeney: We have it within each tweet because Twitter is built around the small bursts of information that are tweets. But right within each tweet there is a report button and once the user hits that, it takes them to the range of possible issues that they may be reporting, whether that is abusive behaviour, impersonation or illegal content.

Tracey Crouch: Okay. Simon?

Simon Milner: From our own perspective, we also have a zero tolerance approach towards both illegal and harmful material. On the illegal side, as well as having clear policies about what people can and cannot use our platform for, we also use technology. So there was much talk yesterday in terms of the announcements made at the PM Summit around technology. We use one of the pieces of technology that people are talking about, photo DNA, a piece of technology created by Microsoft, and we use that to scan every photo uploaded to Facebook. That are some 350 million photos a day against a database of known images of child exploitation imagery, and we will not allow those to be uploaded and will take action against those accounts, including notifying the authorities about them. Then we work closely with CEOP in respect of their access to that data via Netmec in order to track down the offenders. When it comes to harmful material, that is pretty much very hard to spot by technology so that is where our reporting processes come in, and we too have extensive reporting processes via both our Help Centre, which anybody can access-they do not have to have a Facebook account for that-or from individual pieces of content. They can report it to us or indeed report it to somebody else who can help that user take action so, particularly, in the areas, for instance, of bullying and harassment, we do not tolerate that. Those reporting processes are incredibly useful to someone either who themselves is feeling a risk of harm or if somebody sees somebody else on the platform who they think is at risk, then they can use those reporting processes.

Q129 Tracey Crouch: This week, as I am sure you are aware, is anti-bullying week and there has been a great deal of discussion about cyber bullying over the past few months, especially as a consequence of some rather tragic events. What are you doing to work with organisations to try to raise awareness about cyber bullying in particular?

Simon Milner: So perhaps if I take that one first. We work with a number of anti-bullying organisations. Indeed, yesterday, I was at an event in London with 450 schoolchildren and their teachers with the Diana Award. The Diana Award has an anti-bullying work stream and they have a fantastic champion there in a young man called Alex Holmes. He does great work around the UK, working with schoolchildren, enabling them to become anti-bullying champions so we support that work. We help to fund what they do and I was there yesterday to talk mainly to teachers to help them understand how to deal with these issues on Facebook, because it is something that we absolutely recognise they need help with. But we also provide lots of help via other people who reach out to schools. So Childnet are another great charity for doing this, because we recognise bullying typically does not just happen on one platform and will happen across a range of different media as well as in the school. Therefore, working with people who can not only talk to the schools about Facebook but also about Twitter, about Snapchat, about Ask.fm and other services that young people are using is much more effective than having Facebook necessarily coming to the schools.

Sinéad McSweeney: I think we would have a similar approach in terms of partnering with the people who are talking to children all the time in this space, so the South West Grid for Learning, the UK Safer Internet Centre, participating not just in the Safer Internet Day or the anti-bullying weeks but on an ongoing basis, highlighting the resources that are available both within the platform and more generally in terms of breaking down the advice that we have to ensure it is suitable for each different category of people who need to know and understand what is happening. So, for example, in our Safety Centre we would have tips for teachers, tips for parents, tips for teens, recognising that the language and the context may need to be slightly different for each audience. That is material that we have highlighted again, as I say, to organisations who are working in this space on a daily basis, whether it is Parent Zone or BeatBullying or whatever, so that they can link to our material and also so that we can link to their material so that there is a constant shared learning. Also, as Simon says, recognising that the context for the behaviour that you may be seeing on one platform may be behaviour and activity that is occurring in other platforms and also in the offline world, it also about encouraging how to deal with online conflict, encouraging empathy and also trying to educate young people that the consequences of their online behaviour are the same as offline. If you say something online, it has an impact on somebody in the real world although for young people it is their real world, so sometimes we have to be careful even in terms of that kind of language. But it is a question of ensuring that we are using our own communication channel effectively by working with people like the Safer Internet Centre on their Connect with Respect campaign but then also getting into the other worlds in which young people are accessing information and ensuring that those who are guiding them through the world, whether it is their teachers or parents, are as equipped to deal with the questions and activities that they are involved in as the young people themselves.

Q130 Tracey Crouch: You both mentioned that quite often harmful behaviour is happening across multiple platforms. Do you think that all social media providers are equally good at dealing with this or do you think some are better than others? I have in mind, obviously, the controversy that came out over the summer about Ask.fm and I just wondered what your views are on that. Do you think that that, perhaps, poor or lacklustre response from Ask.fm had a very negative impact on the wider social media providers?

Sinéad McSweeney: I think we can only speak for our own platforms but I know that our experience, whether it is ourselves and Facebook who are here today, is that it is an area that we work on together. We are not territorial or proprietary about the way in which we work with young people to ensure their safety online. We all participate together in, for example, the planning day for next year’s Safer Internet Centre and there is another planning day coming up in December. So, from our point of view, it is about sharing the expertise and ensuring that together we are giving young people the skills and resources they need.

Simon Milner: In fact, we work so closely that when Sinéad and her colleagues recruited our Head of Safety for Europe, we were genuinely pleased. We thought this was good for her in terms of her development but also good that Twitter sees the value in having somebody who has been working on safety for four years come and become part of their team. I think that we have to recognise one of the things about young people is they love new services and they love to try them out. Usually, they are the first ones to adopt new services. It would be unrealistic to expect that those new services have invested to the extent that our two companies have in safety. Nonetheless, I think we have a duty as citizens but also as part of this industry to help them, so that is what we try to do. I certainly do that around, for instance, the UKCCIS Board that I sit on. We are always trying to share intelligence around what are young people using and how can we help them to be well-educated in using those services.

Q131 Tracey Crouch: How many people do you employ to combat illegal or harmful content?

Simon Milner: We have a user operations team of several hundred people. It is not a public number. I am afraid I cannot tell you. Our company as a whole has 5,500 people. We have several hundred people who work in user operations, including hundreds of safety experts, so we have people who have seen everything before and know how best to deal with it. We are always also looking to work with outside organisations to try to ensure that we are consistently learning about what is the best way of helping our users.

Sinéad McSweeney: Similarly, we are a much smaller company obviously. We have 2,000 employees worldwide. We have trust and safety teams based in San Francisco and also here in Europe in Dublin, so those teams are providing 24/7 coverage right across the globe across the different time zones. The trust and safety and user services teams are divided into sub-teams to address the different issues that can arise across the platform, the different rule breaches. There is a team looking at abusive behaviour or user safety behaviour through to impersonation through to illegal content. So those teams are covering, as I say, the different time zones on a 24/7 basis.

Q132 Tracey Crouch: Final question, Mr Chairman. Clearly, although you might like to think so, neither of your platforms are perfect in dealing with particularly harmful content. What do you think you can do better to try to stop or prevent future harmful content, because I recognise that illegal is very different and probably being dealt with quite well?

Simon Milner: That is a really tough question. I would say we never rest on our laurels. We are always looking to innovate and I think the moment when we start to think we have this cracked, it would be an unfortunate moment. It certainly would not fit with the culture of the organisation, which is to constantly innovate. So I really would be reluctant to predict what is the next best thing we could do but I would be very surprised if I came back here in a year’s time and said, "We are just doing the same as we did last year". We are always looking to innovate.

Sinéad McSweeney: I think like anything in life, there is always the potential to do something more effectively, more efficiently and in a better way for users. I spent 10 years working in police communications and in that context, of course, you also in some ways have the division between what was illegal and things like antisocial behaviour, which was harmful. So it is about constantly working with the community, and in our case that is our users, to know and understand how they want the platform to be a safer and better place because at the end of the day, it is there to enrich their lives. It is connecting people to events and information and news and so on.

Q133 Tracey Crouch: But you both recognise that you still have weaknesses?

Sinéad McSweeney: It is a question of understanding that as the technology evolves, just like in the world of crime or antisocial behaviour, people think of new ways to breach your rules so you have to ensure that you are one step ahead. So we have a dedicated engineering team on the trust and safety side constantly alive to the possibility that there are issues within the product that we will need to address when somebody finds a new way of using the platform. So I do not think anybody would ever put their hand up and say, "We are done on safety", and particularly not when it involves children.

Q134 Chair: You said you each have lots of people sitting there who are safety experts. If I push the report abuse button, how long before somebody looks at what I am reporting?

Simon Milner: In Facebook, it would depend frankly on who you are and what you are reporting. So if you were reporting that somebody you know is suicidal, we are going to look at that very, very quickly. It could be within minutes. If you were reporting that there is some spam that you are seeing, we may not get to that for a week so it would depend on the seriousness of the harm that you are reporting. Also, frankly, how old you are. So teenage reports we are going to look at more quickly than we do others. But one of the things we always hesitate is to make those hard-and-fast rules because, otherwise, we might find that people think, "I have this problem. I will report myself as being suicidal because that means I will get looked at sooner". So we always aim to get to reports within 48 hours and most of them we do that. As I say, for those that are most serious we will get to them very quickly.

Sinéad McSweeney: Similarly, depending on the nature of the abuse or harm that is being reported. Again, threats of self-harm, suicide, threats of violence and anything under child sexual exploitation category is looked at within a very short period of time. Again, other issues may take longer.

Chair: Are we talking minutes again?

Sinéad McSweeney: It can depend. It could be a minute to an hour. Again, you cannot be hard-and-fast in terms of the issues but we do respond to those first, before other issues.

Chair: For instance, there is a choice. If I press the report abuse button, I am given the choice of different categories?

Simon Milner: Yes.

Sinéad McSweeney: Yes.

Q135 Chair: I see, and depending which I choose will determine how long it takes to get back to me?

Simon Milner: It could be that but, also, it may be dependent on what is going on in certain events. One of the things about our teams is they are national teams so they are interested in what is going on in the news. So, for instance, around the time of the August riots 2011, the team was on high alert for content associated with those riots. So antisocial behaviour or people organising that kind of thing might not have been flagged otherwise but because we were particularly on the lookout for that, then we will deploy more resources. So it will depend on the circumstances, the public circumstances as well as the circumstances of the individual who is reporting.

Q136 Angie Bray: In terms of antisocial behaviour, the debate has also moved on to whether you should be showing certain footage of things like beheadings, which I think, probably, just showing that does cross into what I would consider to be fairly abusive. I know that Facebook has taken quite a strong line on that. Has it Facebook’s view changed as this debate has raged or are you still adamant that this is freedom of information and people should be allowed to see these?

Simon Milner: I think the way I would put it is that we have refined our approach to this kind of content. I am sure we are grateful every day is that we live in a country where beheadings are abnormal, where random acts of violence do not happen to most people we know, but there are lots of people who are using our platform in countries where this is a normal part of life. They want to use our platform to highlight what is going on in their neighbourhood, in their country, in their city. Sometimes that means posting graphic content because that is the only way to get the world’s attention. We recognise there is a place for graphic content when people are condemning it, so absolutely not to glorify it and not to take pleasure in it. When people are using our platform to condemn this and to bring the world’s attention to it, then we should be able to do that. However, we also recognise that given that we are a platform that also has young people, people under 18, we need to enable people to get more prior warning and we are taking steps to ensure that people share responsibly. There are different ways in which we can do that on the platform and we are developing those, frankly, in the light of some recent events. So you are right to highlight it and I am glad I have had the opportunity to address it, but it is something where we are refining our approach because we think there is a place for people to share that kind of content in the right context and with appropriate warnings.

Q137 Angie Bray: But do you set the context? It is one thing to show footage but you are not suggesting, are you, "We are showing this because it is demonstrating what terrible things are happening"? You are simply showing it, so are you setting an adequate context?

Simon Milner: We are not setting any context.

Angie Bray: That may be the problem.

Simon Milner: No, just to be clear, because we do not post anything to Facebook, it is our users that post things to Facebook and it could be a government, they could be a parliamentarian, they could be a campaigning organisation and they could be an individual. An example is after the terrible bombings in Boston at the Boston Marathon, a number of people who were witnesses posted footage to say, "Oh my God, look at this horrendous thing that has happened". We enable people to do that and we enable people to share that with the world, and we think that is part of how we bring the world’s attention to some horrible things that are going on. What we do is provide the tools to enable them to share responsibly and we are refining those tools as we speak to enable people to put warnings, and also to share in a way that young people are much less likely to see it. But just to be clear, if you go on to Facebook and search for beheadings, you would never have found them. Our search facility does not allow you to do that whereas if you go into many other parts of the internet and search for things like that, you can very easily find it. In the recent examples, we had no incidence of reports to us from young people or others with responsibility of looking after young people saying, "Oh, my goodness, this young person has seen this video", because it was not being shared in that way.

Q138 Angie Bray: Twitter, of course, has been under some kind of criticism for the fact that it was quite a useful vehicle during the recent terrorist outrage in Kenya where it would seem that some terrorists were using Twitter to get all sorts of messages out. Again, does that make you uncomfortable that Twitter can be used in that way?

Sinéad McSweeney: As a platform, we do not knowingly allow terrorist organisations to maintain accounts on Twitter.

Q139 Angie Bray: But were you aware that it was being used?

Sinéad McSweeney: Where we have reports to us from law enforcement, where it is brought to our attention that terrorist organisations are maintaining accounts on Twitter, we take action on those accounts. We do not allow accounts to be used for direct threats of violence or for unlawful purposes. I think the flip side of the events in Egypt, in Kenya and events right across areas of conflict is that Twitter is used by human rights activists and by the media to get information out to the world about events like those and ensure that people in real time can follow the events. But to make it clear, we do not knowingly allow terrorist organisations to maintain accounts on Twitter.

Q140 Angie Bray: But it was over a period of two or three days, not just a flash in the pan, that these terrorists were able to send out to the world what they were going to do, what they were not going to do, that they were coming again, and they were basically using it to send out their philosophical message. So at what point would Twitter have woken up to the fact that they were being used in this way? Really, the world does not need to know about that, does it?

Sinéad McSweeney: Drawing on my experience from policing and the way in which police approach the investigation of certain types of crime, there are different approaches to the existence of information in the public domain and the way in which it can assist investigation. So, as I say, we work closely with law enforcement. Where we receive reports of terrorist organisations maintaining reports on Twitter, we take the action that is required of us.

Q141 Angie Bray: Did you get any reports on that, because it was certainly all over the newspapers?

Sinéad McSweeney: I am not going to talk about individual accounts. We do not talk about anybody’s individual accounts or indeed individual engagements that we have with law enforcement.

Angie Bray: So was action taken?

Sinéad McSweeney: Action was taken in those circumstances but we do not comment in detail about the interactions or communications we have with law enforcement.

Q142 Mr Sutcliffe: If I can just turn to the anti-bullying theme, in the evidence that is given to us, 30% of youngsters under 16 expect to be bullied on social media. What do you say about that? What is your view of that?

Simon Milner: I think it is sad that young people feel that. I expect if you ask them also, "Do you expect to be bullied at school?" probably a similar number would. Most bullying that we see on Facebook starts for young people in the classroom, on the school bus, in the canteen. I have spent a lot of time travelling around the UK meeting with teachers and that is a regular thing that they tell me. Indeed, because of that, we have allowed a feature to enable teachers to resolve issues on Facebook. They can both report things to us without having an account but, also, a young person who is feeling bullied by somebody else in school can report that person to a teacher. All they need is an email address and that teacher will get a report saying, "Simon Milner feels he is being bullied by-forgive me-Sinéad McSweeney and this is the content that he feels is bullying".

Q143 Mr Sutcliffe: But does that happen?

Simon Milner: Roughly half of all the young people who report bullying use social reporting and that is a global number. They find it very effective for resolution. The reason we introduced this was often it was very difficult for us to understand the context of a relationship between young people and, typically, bullies are friends of those who have been bullied. Therefore, it was very hard to understand why that person felt a particular piece of content was bullying because we did not understand the wider context of their relationship, and that is why we enabled this feature for somebody who does understand that context to be able to get involved in it.

Q144 Mr Sutcliffe: Is it just teachers? One of the other things that concerns me is that children or young people are on the internet without their parents being involved. What do you do to help parents identify potential problems?

Simon Milner: I am a parent of teenagers. I absolutely understand where you are coming from although, like fellow parents, one of the things you recognise as children grow older is they do become more independent. I do think part of their digital citizenship, indeed their general citizenship, is about having an ability to be independent online, particularly around teenagers. One of the things we do provide is, firstly, parents can report to us. So if they think there is a problem on Facebook, they do not have to have a Facebook account to report to us. They can do that through the Facebook Help Centre. We provide information through Parent Zone. You mentioned the good work that Vicki Shotbolt and her team do earlier. I have done a video for Parent Zone explaining to parents about how they can deal with issues if they have concerns about what is going on on Facebook. But as with everything, including some of the things that we have discussed yesterday in the Prime Minister’s Summit, the key thing is getting young people to talk to someone. That is where we found the biggest problem in the terrible events of young people committing suicide. Typically, they have not told anybody about the problem and we cannot help somebody who does not tell us, just the same way as a teacher cannot help a pupil if they do not tell them. So that is the main message we give to parents, "For goodness sake, even if your child will not tell you, get them to tell someone who can help".

Q145 Mr Sutcliffe: I think the point I am trying to get to, Simon, is that because we have social media, it has not suddenly become a problem. That problem has always been there, as you say, bullying at school, that sort of thing. I think that is the issue that I am trying to get across, that you have mechanisms in place for people to get to.

Simon Milner: Also, I think one of the things that distinguishes bullying online from bullying offline is that there is a footprint. There is a record so this business of being able to report to a teacher is not saying, "Sinéad said this". It is, "Here is what Sinéad has said to me. Here is the photo she has posted", and that is physical evidence that can be used then to help, not to necessarily punish the bully but to help educate the bully about, "This is the consequence of what you are saying and the words that you think may be funny actually are very hurtful".

Q146 Mr Bradshaw: Sorry, I am having trouble finding your report abuse button. Could you help us? Report abuse, where is it?

Sinéad McSweeney: It is within the tweet. If you are in a tweet the three dots indicate more content. At the bottom. It depends on which operating system you are using.

Mr Bradshaw: Don’t worry. Perhaps you can show us later.

Sinéad McSweeney: At the bottom, it says, "Report tweet". I am happy to talk you through it.

Q147 Mr Bradshaw: It did not seem to me to be in an obvious place. You said earlier that you felt you did respond in a timely and effective fashion to reports of abusive tweets but that was not the experience of our colleague Stella Creasy and the woman who was involved in the save a woman’s face on the British bank note campaign. They are quite high-profile people so how can we have confidence-they were very unhappy about how this particular issue was responded to and the abuse they suffered-that ordinary people who do not have that level of profile and voice are having their reports taken seriously and dealt with quickly enough?

Sinéad McSweeney: I think without getting into too much detail of individual accounts-I appreciate that while Ms Creasy has given evidence in public here that Ms Criado-Perez, while she has spoken publicly, has not given evidence in this forum-in some ways, what was experienced in that number of days in August is not the commonplace normal experience of a Twitter user. The levels of activity that were directed at those individuals were unprecedented in my time with Twitter, so that is the first thing to mention. Since then we have done a lot of work to highlight the reporting resources that are there because part of the issue was accessing the reporting forums. It was not a case of non-responsiveness to our report. I think the issue that they were talking about was the ability to report so the in-tweet reporting, greater ease of access around forums, better tools for users are things that we have improved since and continue to improve, going back to the other member’s question, that that job is not done.

We have increased the resourcing around our trust and safety team. Simon already made reference to the person we have recruited to head up user safety for Europe and she has started on our team since then. In addition to that, again, we are working with the organisations in this space like South West Grid for Learning, UK Safer Internet Centre, so that when people contact them about issues, they immediately are able to give them the advice and the resources that they need. But publicly at the time there were a number of statements from Twitter about what happened and the fact that we took on board the responses that we received. I have spoken personally on numerous occasions in the time since to the people involved in that and I continue to take on board the feedback they have.

Q148 Mr Bradshaw: You mentioned the South West Grid for Learning as part of the UK Safer Internet Centre-in fact, both of you have more than once-and in your evidence you have been full of praise for this organisation. How important do you think their role is in helping to mediate and report abusive comments and getting them taken down?

Simon Milner: I think they are particularly helpful for schools so there are professional hotlines. They are there mainly for teachers and they can help people understand problems because they have seen them before and, particularly, Laura Higgins, who runs the helpline there-it is a shame you are coming to the end of your evidence, she would be a great person to bring in to talk about it-she is an expert in all the different services that young people are using and she has seen every different kind of issue in the school before. I was with her yesterday at this event with the young people and their teachers and not one of those teachers had heard of her helpline, so it was a great opportunity for her to tell them about it and it is wonderful for me to have this opportunity to tell people about her helpline. They have a phone number that teachers can call and say, "We have this problem. Can you help us?" Then what she has is a hotline to us. Often she can help people resolve their problems without needing to come to Facebook but sometimes she cannot. She has a hotline to our team in Dublin to help resolve things where she cannot do it.

Sinéad McSweeney: That is the same approach that we have, that we ensure that she has all the information she requires and in the situations where she cannot in the moment resolve the issue that the school is bringing to her, she can contact us and we give her the additional information she requires.

Q149 Mr Bradshaw: I should perhaps declare an interest, Mr Chairman, because she is a constituent of mine. Given the high praise that you both have for her and her organisation, is it not surprising that it is run on a shoestring funded by the European Union?

Simon Milner: Indeed, I was talking with her about this yesterday. We have provided some resources to help her and her colleague David Wright run a road show around the country, so we helped to support that. But I think you are right in saying this is something that we as an industry should look at, as Laura is looking to develop that helpline, as in, "Can we help her further, including giving additional resources?" It is a good point to raise.

Sinéad McSweeney: Similarly, we have assisted them with training workshops for hotline managers and so on, so we have been asked and have been able to contribute to the way in which they provide their service.

Q150 Mr Bradshaw: But this valuable service, I think, is being done by one and a half people with very little money. It is running out next year. You have deep pockets. Is this not the kind of thing that you should be funding as a matter of course?

Simon Milner: Laura has not actually asked us for money to do that yet but I am sure when she does, possibly as a result of this, we shall be listening with a very open mind.

Q151 Paul Farrelly: You are both global businesses so, presumably, you share with each other in your organisations experiences from around the world.

Simon Milner: Yes.

Paul Farrelly: Presumably, if something happens in one part of the world, as with any big global organisation, if you are approached by the press or committees like this, you will have notes on how to deal with questions?

Simon Milner: I will not put it as strongly as that. We tend to talk about what are the issues we think might come up and, frankly, who is going to answer the question first.

Q152 Paul Farrelly: Sinéad, if you were asked about things that are happening in one part of the world, would you expect to be briefed by an organisation?

Sinéad McSweeney: It depends. My responsibility goes beyond the UK so I would be familiar with issues that are going on in the EMEA region, which is Europe/Middle East basically.

Q153 Paul Farrelly: Okay. I was following the build-up to the All Blacks match at the weekend and, as we were doing this inquiry, I was asked whether we have come across the case that is causing huge controversy in New Zealand called the Roast Busters. Have you come across that case?

Simon Milner: No, I was at the recent Family Online Safety Institute conference in Washington DC with NetSafe, the New Zealand equivalent of the UK Safer Internet Centre, and they did not mention it, so I am not aware of that, I am afraid.

Q154 Paul Farrelly: Have you come across it?

Sinéad McSweeney: I am not familiar with it.

Q155 Paul Farrelly: No. I am just reading from a CNN report and reports have been all over New Zealand press over the last two weeks so this is from a couple of weeks ago. Just a synopsis of this case, it is about a group of teenage boys who were boasting online about raping drunk and underage girls, getting so incoherent they did not know what they were doing. What came to light, which was exposed by a Channel 3 television programme in New Zealand, was that for some months they have had a Facebook site on which they were posting these boasts. Days after the exposure of the site, it was shut down and there was apparently another site that had started on Facebook. Facebook is not the only social medium to be used by this gang. Just again from the CNN reports, Twitter was also a conduit for their boasts, Ask.fm and also YouTube. I clearly cannot ask you what lessons you are currently learning from that experience because you do not know about it.

Simon Milner: Mr Farrelly, I am not aware of that case. I am happy to ask my colleague, who does my equivalent role in that part of the world, about it and see whether there are some lessons we might learn. I do not really want to comment further. It is clearly not appropriate because I do not know any more about it.

Q156 Paul Farrelly: Would you both drop us a note on it?

Simon Milner: Sure.

Q157 Conor Burns: Can I just ask in relation to this report abuse facility within Twitter, does it alarm you that you have three relatively sophisticated and regular tweeters on the Committee and we cannot find it? I was not aware that it existed. When did you launch it?

Sinéad McSweeney: It was rolled out at the end of July and the plan was to roll it out across all the platforms by Christmas but, in fact, we achieved that by the end of September. If you are saying to me that we need to highlight it more, that is something that I can take away. We have done blog posts. We have tweeted about it. We have spoken to various safety organisations who work in this space. I think every time that I have been at a meeting, I have mentioned it but as I know only too well, having spent 10 years in Comms, that you can never tell people something often enough so we will keep telling people about it.

Q158 Conor Burns: I think the message from us is that with the nature of what we do, we are probably slightly more prone to abuse sometimes than your average user and I was not aware that it existed.

Sinéad McSweeney: It is not the only means by which you can report abuse. You can also report abuse through the Help Centre.

Conor Burns: I know. I have just googled that. I have seen that there is another mechanism to do that.

Sinéad McSweeney: Yes.

Q159 Conor Burns: Facebook. When we were in California, I raised an instance where someone had posted something that was profoundly libellous against someone else. It was reported and absolutely nothing happened. Facebook came back and said they deemed that it was legitimate comment. Would you like to comment on that?

Simon Milner: I do not know the details of the individual case but, as you can see, we looked into it and we do not always agree with reports that we get. Frankly, a lot of the reports we get are frivolous. The classic one is somebody says they like Justin Bieber and other people report them as being suicidal so I am not belittling it but we have looked at it.

Q160 Conor Burns: What about somebody saying that they believe somebody else is a child abuser?

Simon Milner: Again, we will look at the context and because one of the things we have learnt is just because somebody has said it, that this is suicidal or child abuse, does not mean it is. We have to look at the context and the details.

Q161 Conor Burns: No, I am putting to you that when somebody posts that they believe somebody else to be a child abuser and that was not taken off.

Simon Milner: I do not know the circumstances to the individual case, and perhaps you and I can correspond on it afterwards, but one of the key things that is open to people if they think we have made the wrong decision is they can go to the courts.

Conor Burns: Which is out of reach for a lot of people.

Simon Milner: That is a matter for Parliament as to how the defamation process works in the United Kingdom. I do not think that is a matter for me.

Q162 Conor Burns: But there are often instances where there are lovely, warm words that come forward about procedures, hundreds of people on help desk, but we encounter real people who are finding that there are no options open to them and you hide behind, "That is a matter for Parliament". But you are the publishers. You are allowing these things to be published and said about other people.

Simon Milner: We are a platform for free expression. We have a mechanism for somebody to report to us something that they think is defamatory or libellous. We look at those very carefully, we are risk-averse as a company, so I am sure that the team that have made that decision have done so with full knowledge that having made that decision to leave that content up that things are open to an individual to not only take action against the person who posted it but also against us. But on the matter of, "These are warm words", these are not just warm words. We see examples week in week out of particularly young people and people in distress being helped as a result of reports by their friends-young people who are reporting that they are feeling suicidal or indeed that they have taken some pills or cut themselves. In situations where we see that somebody who thinks somebody is in imminent risk of harm, we will alert the local law enforcement who would go and find that person. Quite typically, they would get to them before they do the deed or in order to help them so we help people in the real world enormously week in week out.

Q163 Conor Burns: Yet, as a publishing platform, you allow comments to remain up that a newspaper would not in 100 years dream of publishing.

Simon Milner: I do not think that is fair in the context of I do not know the details of this case but I am absolutely certain that a team would have properly looked at that and made the decision based on the available evidence in the context. It is open to that individual to take action if they wish.

Q164 Chair: Can I pursue with you some of the measures that you have in place? First of all, what is the minimum age for somebody to subscribe either to Facebook or Twitter?

Simon Milner: It is 13 for Facebook.

Chair: 13.

Sinéad McSweeney: Yes, our services are not directed at under-13s.

Q165 Chair: 13, and how do you verify that somebody is 13?

Simon Milner: With our platform, when people sign up for Facebook, there is quite an extensive new user process and one of the things we require people to do is provide their date of birth. If somebody puts in a date of birth that is under 13, we will not allow them to open an account and we will also use technology to prevent them from trying to put in a new age from the same device, trying to change their age, so that is one mechanism. We recognise that people, unfortunately, lie about their age. Did you see this recent report from the NSPCC, which came out on Friday, about this? They reported a pattern that we have seen in other studies that in the majority of cases of 11 and 12-year-olds joining Facebook, their parents have helped them set up the account so the parent knows that their child has a Facebook account. One imagines typically the deal is, "You can have a Facebook account but you will be my friend on Facebook so I can look after you". In those situations, it is very difficult for Facebook to spot that that person is an underage user. We still also encourage and indeed find reports from our users of underage accounts incredibly helpful and indeed when teachers ask me about this, I say, "You can report it. You can let us know that somebody is underage".

Q166 Chair: Okay and what about on Twitter?

Sinéad McSweeney: We do not collect age information on sign-up. I think Twitter has established a reputation in the area of privacy. We minimise the amount of information that we require from users to sign up so we do not collect age or gender or other details about our users. Where it comes to our attention that somebody under the age of 13 is using the platform, their accounts are removed.

Q167 Chair: You say you have to be 13 to use Twitter but you are also saying you do not ask anybody if they are 13 or over 13?

Sinéad McSweeney: Yes. That is a balancing between the premium that we put on our users’ privacy and the issues that you are raising.

Q168 Chair: How does somebody know they have to be 13?

Sinéad McSweeney: Sorry?

Chair: If I joined Twitter, am I told that, "You have to be 13"? How do you make people aware?

Sinéad McSweeney: It is in our privacy policy that this service is not directed at people who are under 13.

Q169 Chair: So you have to go into your privacy policy and read it and then you discover that you are supposed to be 13?

Sinéad McSweeney: Yes.

Q170 Chair: Right, okay. What about fake identities, people who open accounts in false names?

Simon Milner: With Facebook, that is against our rules. You have to use your real name on Facebook and, indeed, we think that is the core of Facebook’s success. We would not have reached 1.2 billion monthly active users had we not have that policy because Facebook is all about connecting with people you know in the real world. So it is against our terms and we very much encourage people if they see somebody who they think is not using their real name, they are an imposter, to tell us about it and we will then look into those accounts and remove them if we conclude that is the case. We may require people to prove their identity if they claim that this name that appears very fake is real. That is something that we take seriously and indeed is at the heart of our mission.

Sinéad McSweeney: We do not require users to use their real name. Most people do because they want other people to be able to find them and connect with them on the platform. We allow people to use pseudonyms. We allow people to set up parody accounts that range from emergency cute puppy accounts through to political satire. More importantly, we have seen the platform used in areas of conflict by human rights activists, journalists, people who are at risk of harm from the information that they are attempting to get out into the world publically. That is very important to us and that is why we facilitate that kind of political activism and content on the platform.

Chair: As I understand it if I set up as Donald Duck there is no problem with Twitter.

Tracey Crouch: That has already been taken.

Chair: Donald Duck the 52nd. Twitter, you are perfectly content. Facebook, it is only if somebody tells you that I am not Donald Duck and then you will do something.

Simon Milner: No, it is not if somebody tells us they are not Donald Duck, it is if somebody else tells us.

Chair: Yes, I understand.

Simon Milner: There are also certain names that we will look out for so you cannot set yourself up as David Beckham, for instance. We also, like Twitter, provide a process for people to have their accounts verified. This is particularly useful for people in Government, for instance, so you know that when you look at David Cameron’s page on Facebook it is David Cameron and not somebody pretending to be David Cameron. That is something that we are rolling out to help people who are public figures to ensure that people understand yes, this is the real person.

Q171 Chair: Let us assume Donald Duck52 indulges in highly offensive behaviour, trolling or even posting child sex images, how do you identify them? Do you keep records and ISP addresses?

Simon Milner: If somebody is using our platform to try and distribute child abuse imagery then that is going to be spotted through photo DNA assuming it is a known image. As you may be aware, we are working together with other technology companies to try and ensure that we share even more intelligence around these known images.

Chair: I am asking how you identify whose account it is.

Simon Milner: Reports are incredibly powerful. We kind of have a neighbourhood watch community of scale, and our users really care about their safety and the safety of others in the platform. They will be reported to us and then we will deal with it.

Q172 Chair: That does not address my question. It may be important to you, you may find it but if it is being posted by somebody who claims to be Donald Duck52, how would you identify it in that instance?

Simon Milner: I see, I am sorry. In the situations where, for instance, law enforcement are on to somebody like that and they want to find out who they are and where they are living and what have you, we have a process for them to request communications data from us to enable them to do that. That would include the IP address.

Chair: You will be storing the IP addresses and the messages for how long?

Simon Milner: We store the IP logs for a period. I would have to come back to you to tell you exactly how long. The messages, it will depend. Typically it depends on how long a person wants to retain those messages but most of the time law enforcement will have the messages because somebody has come to them saying, "Look at this thing on my account". They have the messages, they have the content, what they need to know is who is this person and that is when they can come to us to ask for the communications data.

Chair: Twitter, do you keep that same data?

Sinéad McSweeney: Yes, we also retain communications data. We have protocols and guidelines for law enforcement on how they can work with us to request information, and they can also request that we preserve information pending receipt of further requests from them.

Q173 Chair: But you are keeping records of tweets and IP addresses from where they originate.

Sinéad McSweeney: The tweets are public anyway.

Chair: They might get removed or deleted.

Sinéad McSweeney: In the main most of the activity is public and remains public, and we will retain a certain amount of communications information.

Chair: You retain IP addresses?

Sinéad McSweeney: Yes, we retain communications data.

Q174 Chair: You would allow law enforcement access to that with a court order, would you?

Sinéad McSweeney: We receive law enforcement requests. They range from emergency situations where if there is an immediate threat of life we respond to it straight away, and also in other areas of serious crime we deal with requests from law enforcement on a case by case basis.

Chair: Civil actions?

Sinéad McSweeney: Civil actions, again it would depend on circumstances. It is difficult to talk about hypotheticals but on receipt of valid legal process we deal with those requests.

Chair: I would need to get a court order.

Sinéad McSweeney: Yes.

Q175 Chair: A court order from the UK, would you accept that?

Sinéad McSweeney: Again it is not my exact area of expertise the actual legal process that is required. I would need to come to you on that but we would need valid legal process.

Chair: It was suggested to me you had to get a court order in California, which is not always terribly easy for a civil action for say privacy invasion or defamation.

Sinéad McSweeney: Yes, but there are various treaties and so on that cover civil litigation and cooperation between jurisdictions also that would be invoked in those circumstances.

Chair: I think my colleagues have exhausted their questions so thank you very much.

Simon Milner: Thank you very much.

Sinéad McSweeney: Thank you.

Witnesses: Tony Close, Director of Content Standards, Licensing and Enforcement, Ofcom, Claudio Pollack, Group Director, Content Consumer and External Affairs Group, Ofcom, gave evidence.

Chair: For our second session this morning may I welcome Tony Close and Claudio Pollack both representing Ofcom?

Q176 Tracey Crouch: Could you perhaps just start by telling us what regulatory powers you do have in relation to the internet, please?

Claudio Pollack: Yes, our role is limited to television and television like content broadcast over the internet in the UK, originating from the UK. We do not have any powers that go beyond that, the regulation of content.

Q177 Tracey Crouch: That would include video content for example, child abuse videos or rape pornography, or something like that? That would include that?

Tony Close: If that content was included in a regulated service. If it was included in a linear television service provided over the internet or if it was included in an on demand service provided over the internet and established in the UK, yes, it would include any content.

Tracey Crouch: But if somebody posted up a homemade video, for example, this would not be included within your regulatory powers?

Tony Close: Probably not, no. Not if it is not part of a broader regulated offering, a broader service is regulated.

Q178 Tracey Crouch: Given the extent of Ofcom’s powers and the obvious expansion over time as technology has changed, do you think there will be a greater role for Ofcom in the future regulating internet content?

Claudio Pollack: Shall I take this one? I think at the moment there are a mix of instruments around making the internet safer for people. There may or may not be a different role for us in the future, and that is obviously a matter for Parliament. What I would say though is that the mechanism that exists today, the long established mechanism that exists today for the regulation of television services, which has been very effective, I do not think it would be appropriate or possible to extend that form of regulation across all of the content on the open internet. I think a different mix of instruments will be required over time to make the internet safer for those people that use it.

Q179 Tracey Crouch: You do have people working at Ofcom on harmful and illegal content on the linear platforms that you were talking about?

Tony Close: Yes.

Tracey Crouch: How many people do you have working on it?

Tony Close: Within the standards enforcement team within Ofcom we have around 40 people but I think if you included some of the other teams within Ofcom that have a role in monitoring, researching or developing policy around potentially harmful content, it would be around 50 people. They, of course, work on a range of different content related issues at any one time.

Q180 Tracey Crouch: If your powers are extended into wider internet use, perhaps like draft Communications Bill at the moment in front of the House, do you foresee any problems with Ofcom having further powers or wider powers to do with this issue?

Claudio Pollack: It really depends what it is that that extension means.

Q181 Tracey Crouch: Given obviously this inquiry has been looking at harmful and illegal content on the internet, and yet you by far the most powerful regulator on these matters in general terms for television and radio do not have any regulatory powers, do you think that it is time that you did and that you were able to help in terms of regulating content in a way that is obviously very much needed?

Claudio Pollack: I wonder if it is worth saying a little bit about how we carry out protection on television and radio services and then maybe we can explore whether there are aspects of that that would be appropriate or relevant in the open internet.

Tony Close: Yes, I think it is worthwhile reflecting just quickly on that. We do have powers in this area. They are limited by statute. They do require us to regulate linear television or on demand services when provided over the internet as well as via other mediums or platforms.

Chair: This is what you delegate to ATVOD?

Tony Close: In relation to video on demand, yes. Across those services that we currently regulate on the internet we have a comprehensive set of rules. We are required in relation to video on demand with ATVOD to implement those rules. We continuously monitor compliance with those rules through complaints or proactive monitoring. We refresh those rules where we think that the industry has moved on or public policy concerns have moved on. We have very significant powers currently to either fine providers who have broken the rules or to stop or to suspend their services online or not online. So we have a comprehensive set of powers and a comprehensive set of rules that apply to these services even when provided over the internet already. I think the challenge is identifying the best aspects of the way Ofcom currently operates and identifying the best aspects of the existing framework and seeing how they could be applied to a broader set of services potentially online.

They are obviously limited by geography, our powers, and there is a significant challenge around regulating content over the internet that comes from outside the EU of course.

Claudio Pollack: For example it is not possible at the moment, as I understand it, to broadcast in the United Kingdom without a licence that is granted across Europe. On the open internet content can come from anywhere in the world. of course.

Q182 Tracey Crouch: The on demand services, so the XXX channels for example that are broadcast on Freeview, presumably if, for example, they have specific licensed agreement that they can only provide programmes after midnight or something, if they were to bring that forward to 10.00pm that is where you would step in and remind them of their duties?

Tony Close: Yes, in relation to both linear and on demand adult services, services providing pornography that are regulated by Ofcom and by ATVOD as well there is a broad range of interventions and measures based on time, based on preventing access to children, double, triple PINs and obstacles for gaining access. Obviously there are limits on the content as well that are already in place, limiting the explicitness in particular environments. For example, you cannot have hardcore pornography on Freeview even behind considerable PIN and other access restrictions.

Q183 Chair: In terms of what is and what is not possible, there has been a lot of press speculation about the need to have greater control over the internet and I suspect a lot of it simply is not possible but one of the things that particularly you have looked at in a different context is website blocking, URL blocking particularly. When you last looked at that you threw up a whole lot of technical reasons why it was very difficult to do. Is that still the case in your view?

Claudio Pollack: We looked at it in the context of online copyright infringement and whether website blocking could play a role in that. We absolutely did highlight a number of technical challenges and limitations but for me the report was very clear in saying that, to making it more generic and applicable, if this about lessening the availability of content that is considered harmful, harmful in different ways, that website blocking can play an important in that. It is not the answer to everything. It is not 100%. It is not watertight. It can be bypassed and it is not perfect in the sense that, given the amount of content out there, it can over block. It could block things that you did not mean to block and it can under block, it can allow certain things through.

In both of these examples, public policy examples, if you take blocking or filtering as part of a suite of measures then it absolutely can have an impact and can play a role.

Q184 Chair: It can play a role but if I am a technologically literate child abuser seeking to overcome technical measures, can I do so?

Claudio Pollack: I am hesitating because the conversation I was having was focused more on content that is generally lawful but accessed by children that we do not want children to access. If we are talking about illegal content, for example child abuse content access on the internet that is slightly further outside our area of expertise. The simple answer is at the moment, yes, my understanding is that there are technologies in place that allow those that wish to access these images to bypass controls that are put in place.

Chair: We are told that a lot of this material is now lurking in the dark net as they call it, Tor is particularly indentified. Is it your understanding that is something that it is possible to intercept and identify or is it so hidden that it cannot be reached?

Claudio Pollack: I think we are reaching the limits of our expertise in terms of what Ofcom does. We are far more focused, as I say, on understanding how people can be protected or protect their children from content that they do not want to be viewed in terms of this aspect, these two aspects of child protection. Tony, your turn.

Tony Close: The only thing I would add-and Claudio is right, this is outside our remit-there have been some limited but notable examples recently of prosecutions or investigations of people using Tor or other dark web browsers and routers. It is possible and it has happened on very limited occasions.

Claudio Pollack: It is obviously a big, big area of focus for CEOP, as was, and the IWF. When I have met with them we have explored and also assisted with some of the technology expertise that exists within Ofcom in understanding the mechanisms around identification and blocking. There will be, given the current state of technology, imperfect mechanisms but I suspect they have a very strong role in law enforcement as well.

Q185 Chair: In terms of the role Ofcom plays therefore in encouraging the provision of child protection filters and that kind of thing, how much power do you have to lean on the ISPs for instance?

Claudio Pollack: Beyond the areas of broadcast and on demand content on the internet, our role is not about powers. We do not have any powers to mandate or in particular we do not even have a function, I would say, to engage with internet service providers. The real legal route and where our remit fits here is around our media literacy duties. It is very much about understanding and encouraging media literacy among, in this case, children and parents, and we will carry out a lot of research and therefore be the keeper of the facts. I would add that as part of that role, our main role on online safety, we have a seat on the executive board of UKCCIS, which is the main institution that leads activity in this area. As part of theirs and the Prime Minister’s negotiation or agreement with the main internet service providers to provide network level filtering by the end of the year, they have asked us to carry out a particular role in understanding how that implementation goes.

Tony Close: Shall I add something to that? Before talking about the specific reports that we are carrying out into parental views of online behaviour and safety, it is worthwhile just making absolutely clear that in our role as a board member of UKCCIS we provide all of the data that we harvest as part of our media literacy programmes so that public policy can be developed, public debate can be had about the risks and benefits of the internet and of child internet use. Claudio touched upon some reports that we will be doing over the next year for the Government around child internet safety.

Next month we hope we will be reporting to Government on parental awareness of, take up of and confidence in parental controls and other strategies, mediation strategies, available for parents to protect their children in an online environment. We will be following that up in one year’s time with a follow up report to look at how things have progressed. Just quickly in between the two reports we will be producing a third report for the Government that looks at progress made by the four major ISPs in implementing network level filtering.

Q186 Chair: You say you do not really have much control over the ISPs, that your main area is TV companies and radio, but the Government has said that it wants to ensure that internet enabled devices are supplied with filters and that you are going to be involved in overseeing this. What does that mean?

Tony Close: I think it is what I was just talking about. I do not think we have a specific role.

Chair: But this suggests that you are going to have one in the future.

Tony Close: I think that that is just another way of saying that at least for the next year we will provide a number of reports that provides the Government with our insights into how network level filters have been implemented, whether or not ISPs have delivered on the promises. More than that they will provide our insights on how parents feel about the level of protections that are available to them and whether or not they are confident or not confident in protecting their children in an online environment.

Chair: You see Ofcom’s role essentially as one of research and opinion surveys.

Claudio Pollack: Within the current statute that is exactly right. Beyond the areas that we have identified and described, there is no regulation of the internet in this country in the statute and we do not have a role in that nor do we have a role, a formal role in policing or mandating filtering. The conversations that happened were all about a voluntary commitment to deliver a form of network level filtering and Ofcom, because of its media literacy duties and because of its expertise in networks, has been called upon by the Government to help in a degree of evaluation of the extent to which the filters have been properly introduced and also the impact that that is having on the way that parents and children use the internet.

Q187 Chair: But you do have a role in licensing telecoms companies, do you not?

Claudio Pollack: We have. I was going to say that when you said we only do television and radio that is in terms of protection of UK citizens from harmful content on radio, television and TV like content on the internet. Separately, of course, we have a role around the regulation of telecommunications companies as providers of electronic communications, networks and services. It is not through a licensing regime, it is through a general authorisation regime but that explicitly does not cover regulation of the content that is delivered by those networks.

Q188 Chair: It does not but if the Government chose to say, "We are now going to make it a requirement that you also have child protection measures", you could do that?

Claudio Pollack: Yes. To be very clear, what we are describing is what our role is within the existing statute. Of course there could be a different role if the statute was changed but that is obviously a matter for Parliament, not for us.

Chair: That would require primary legislation to change the terms of the regulatory regime for telecoms.

Claudio Pollack: I believe so, yes.

Q189 Mr Sutcliffe: Just on that theme, I think you are pointing out you are a regulator without powers in terms of this area and clearly that is a decision for us as Parliamentarians to make, but I think your point us to a direction. What concerns me then is if the ISPs know that you do not have any regulatory powers what is your relationship like with them in terms of how are you seen by the ISPs in this sort of area? Do they work with you in a proactive way or do you have to react to what the agenda is? I am trying to get a feel really in terms of what the relationship is like without you having the powers to be able to say to them, "You must do this, that or the other".

Tony Close: I wonder if I might answer first in relation to the specific task we have ahead of us in the next year, and whether Claudio may want to add anything about our broader relationship. We are conscious that because we are limited by statute that we do not have specific powers to require ISPs to do certain things for us in order for us to carry out a process of reporting to Government on network and filtering. As part of the agreement with Government on doing this work, we secured agreement that the Government would contact the major ISPs themselves and ask them to cooperate with us in order for us to deliver this set of reports. I think we all agree that it is important and we cannot do it without the cooperation of the ISPs. We have no broader role with ISPs but, Claudio, I wonder whether you want to add anything?

Claudio Pollack: I have a lot of experience in dealing with the internet service providers in those areas where we do have power and that is a particular type of relationship as you would expect. In this area it is quite straightforward. We do not have any powers but nor do we have any duties or functions so there is not a gap if you see what I mean. Where we would be coming in and saying, "There is a problem here in terms of what we are expected to deliver", would be if there is a gap between our duties and our powers to deliver that.

To be very clear, the provision by the four largest ISPs of network level filters to be used by parents by the end of this year was an agreement between the Prime Minister and the internet service providers. That agreement is about what they deliver but also as part of that agreement and in the exchange that we have it must be that they are required to cooperate in any form of evaluation of whether they have delivered to what is essentially a voluntary deal. If there is not cooperation, for example if we find that it has not been implemented properly or substantively or there is not cooperation in our ability to carry out that task then of course it is an agreement between the Government and the ISPs and it is for the Government to decide, if this voluntary arrangement has not delivered what it hoped to deliver, whether it wants to consider other avenues as well. At the moment it is a voluntary agreement in which we are assisting.

Q190 Mr Sutcliffe: Tony, to go back to the survey, I think, you are doing of parents, this report on parental involvement, which I am pretty keen on because I think that part of the problem is that parents do not know what is going on in their own households. How do you go about it? What are the terms of reference? What is the background to it?

Tony Close: You are going to have to forgive me if I cannot provide you with full details of this because I am the guy at Ofcom that looks after setting rules and enforcing those rules. I am not the guy that looks after the quite complex and comprehensive programme of research that we do but I will have a stab at answering. Claudio, you might want to follow up.

We carry out a range of different quantitative and qualitative pieces of work. We are doing them on an ongoing basis so that we can benchmark and track changes in parental attitudes. The qualitative pieces are obviously limited in terms of numbers of people that we talk to but really gets under the skin of parents’ and teens’ attitudes to internet use, online safety, what concerns them or worries or distresses them, what they like about it, what they understand about parental controls. The quantitative piece is obviously different. It is bigger, it is less detailed but tries to canvass a significant range of views and attitudes to what the risks are online, what steps parents take to mediate or mitigate those risks, what strategies that they develop. It is quite comprehensive. I cannot go through all of it.

Q191 Mr Sutcliffe: I think somewhere in the report it talks about there is evidence now that people are more aware of what the issues are because of experience, if you like, online. How often are these reports commissioned? How often are the surveys changed?

Tony Close: Okay, one of the reasons the Government were probably keen for us to do the work is because they are aware that we already do this on an annual basis as part of our media literacy programme. Those two reports, and there are other reports, programmes of work that we carry out as well that might feed into this are a rolling programme of work. I cannot tell you how long we have been carrying them out for but I think it is a considerable period of time.

Claudio Pollack: I think it is since 2005, so annual surveys since 2005.

Mr Sutcliffe: There is an opportunity now to see a pattern of things happening.

Tony Close: Yes. It enables us to track people’s behaviours and attitudes as their internet experience changes, as they become more confident or as they become less confident, as they become more aware of risk or less aware of risk.

Mr Sutcliffe: I do not know whether it is feasible, Chairman, but it might be worthwhile getting a copy of the sort of questions that are asked of parents, if that is possible.

Claudio Pollack: Under the media literacy duty every year we have a research report, a very, very comprehensive research report one on adult media use and attitude and one on children and parent media use and attitude. It is the primary source, as Tony said, used by UKCCIS to understand changes and it is absolutely vital in understanding how things are moving in this area, with limitations. There are a number of things that it shows around attitudes. One of them is that most parents are doing something, there is a number of things they could do, but most parents are also reporting some difficulty in using certain tools. No majority is using, for example, filtering but a combination of using filters, being in the room when the child is using the internet, talking to their children once a month, the combination of those three mechanisms is 85%. So 15% are doing nothing, but within that 85% there are a number of people that are doing just one or just two of those things.

The other thing that I think is fascinating-and, yes, we will absolutely share the questions of what we are doing annually-is the specific targeted report that we have been commissioned to do using that data. What makes this area so difficult is just how quickly it moves. One thing that I am going to quote to is if parents were saying and taking comfort from the fact that the PC was in the room that they were in and they could observe what was happening, in just 12 months tablet computer use at home among five to 15-year-olds between 2012 and 2013 has gone up from 14% to 42%, and that gives you a real sense of the challenge of protecting children. Even when parents are investing the time and effort to do so, it is very difficult because things move so quickly. There is a lot in the research that will help understand that better.

Tony Close: Just very quickly, Chairman, of course we will share the questions with you and we will share the report once we have finished it as well.

Q192 Chair: Just to be clear, the Government has waved sticks at Google, at Microsoft and at the ISPs. The Prime Minister has basically said, "Report back to me within so many months" with the implicit threat that unless they do these things action is going to be taken against them. The position is that at the moment there are not any powers to take action.

Claudio Pollack: Sorry, say that again?

Chair: The powers do not exist at present to take action.

Claudio Pollack: In the area of regulating internet service providers in having a role in policing content the powers do not exist today. Of course if there are individuals who have the capacity to introduce legislation then they have a different response.

Chair: Of course, but therefore the Prime Minister and Ministers’ implicit threats are that they will have to introduce new legislation to take powers to require internet companies to do these.

Claudio Pollack: Yes, that is right.

Q193 Chair: You do not have at the moment really any provision within the licensing requirements that would cover this kind of area?

Claudio Pollack: There is nothing. It is the general conditions of entitlement that is through legislation that derives from Europe around electronic communications. Not only is there nothing in the general conditions nor do we have the capacity to introduce regulation that is around content using that instrument.

Q194 Chair: Television like material is covered by ATVOD?

Tony Close: Yes. The powers are vested in Ofcom. Ofcom contract out the function of regulating television like video on demand services to our co-regulator, ATVOD.

Chair: Would that cover YouTube?

Tony Close: It would cover parts of YouTube. YouTube is a mix of services. It is in large part and most notable for user generated content and hosting user generated content but increasingly it is a platform for professional services provided from different corporate entities. There are a number of professional YouTube channels that are established in the UK that meet the definition of a television like service and are regulated by Ofcom and ATVOD.

Claudio Pollack: To be clear, in order to be regulated they both have to be TV like and they have to be established in the UK. Content that is visible on the internet that is neither TV like nor established in the UK, we would not have jurisdiction.

Q195 Chair: But if I post a video on YouTube that I filmed on my handheld camcorder or whatever, is that covered by ATVOD?

Tony Close: No, probably not, not if it is user generated content and it is just a part of broader offering. It will be covered by YouTube’s own community standards.

Q196 Chair: Sure, so how do you draw a line between YouTube content that is covered and YouTube content that is not covered?

Tony Close: Partly on terms of jurisdiction, to put them on to the site you have to be established in the UK.

Chair: All right, so assume we are in the UK.

Tony Close: Then you have to identify a service that exists, something that you can see, feel, something that is whole, and then you have to run through a series of potential criteria or characteristics to determine whether or not it is TV like or whether or not its principal purpose is the provision of TV like material to consumers. This statute, I think, contains a number of conditions, all of which need to be passed in order for us to determine, or ATVOD in the first instance, if something should be subject to regulation.

Q197 Chair: Do you feel that this whole area needs further Parliamentary examination, because I was involved in the Bill that set up Ofcom and none of these services existed at that time?

Tony Close: I think it is fair to say that there is a challenge. There are different degrees of protection offered for different types of service online, very comprehensive rules set for linear services online, less comprehensive rules set for television like on demand services, and no statutory rule set or intervention for certain other types of material. There is a challenge there. It is a legitimate debate to be had about whether that is an ideal scenario for consumers or whether or not we should be thinking about a more coherent framework for media standards that spans a greater set of audio visual media. I think there are certainly some benefits in thinking about that.

Q198 Chair: Will this feature in the Communications Bill, do you think?

Claudio Pollack: You have someone giving evidence shortly that I think would be better able to answer.

Chair: Indeed, and that is a question we might put to him. I think that is all we have, thank you.

Claudio Pollack: Thank you very much.

Tony Close: Thank you.

Witnesses: Rt Hon Damian Green MP, Minister of State for Policing, Criminal Justice and Victims, Mr Edward Vaizey MP, Parliamentary UnderSecretary of State for Culture, Communications and Creative Industries, and Claire Perry MP, Special Adviser to the Prime Minister on Preventing the Commercialisation and Sexualisation of Childhood, gave evidence.

Q199 Chair: Good morning, thank you for attending our third panel this morning. It is afternoon now. May I welcome Damian Green, the Minister of State for Policing and Criminal Justice from the Home Office, Ed Vaizey, Minister for Creative Industries and Communications, and Claire Perry, the Special Adviser to the Prime Minister on Preventing the Commercialisation and Sexualisation of Childhood? We have just been having a discussion with Ofcom about what powers are available to them should ISPs and other internet involved companies not voluntarily adopt some of the measures that we have been encouraging them to take, to which the answer was there were very little powers. Do you think we need to give Ofcom more powers in this area?

Mr Vaizey: We await your guidance, Chairman, on who you would like to answer.

Chair: I think possibly, Mr Vaizey, you are best placed.

Mr Vaizey: I knew I should not have said anything. I think Ofcom obviously has some powers in this area. It is clearly a regulator of broadcast services and coregulates what we call video on demand with the Association of Television and Video On Demand. Anything that is a broadcast like service that is broadcast over the web, Ofcom has the powers to regulate. As you know, it has also been asked by the Secretary of State to report on parental awareness and how comfortable parents are with the safeguards that are now being put in place. They were also behind creating ParentPort as a one stop shop to allow parents to get the kind of guidance they need. We are working, related back to video on demand, on common media standards because we take the view that as television and internet services converge anyone, parents, consumers of any kind, will assume that the same standards apply to what appears on their living room screen, which I think you can still call a television for these purposes.

Ofcom also sits on the UKCCIS Board so they have a great amount of input into what we are doing in this area. I would be open to any discussion on an extension of their powers. I have taken the view that we can achieve a lot without necessarily imposing legislation or widely extending Ofcom’s powers at the moment, for example the whole issue of filters where we, I think, have made some rapid progress, has been by engaging in a dialogue with the ISPs. Self regulation, if I can put it that way, seems to me a way of making the kind of rapid progress one needs, not only just rapid progress but also keeping the flexibility so that we can react quickly to events. Some of the changes that Google have made, for example, would not necessarily have succeeded, I think, as quickly if we had had endless debates about the appropriate legislation and so on.

But it is quite clear that Ofcom has a role in this space. The key issue, I think, will be always to be open to ideas and suggestions about whether or not Ofcom should have its powers extended and if so how, what its remit will be, particularly considering you are dealing with global companies who effectively deal with different regulations all across not just Europe but across the globe. I do not know whether Claire wants to add anything to that?

Claire Perry: Thank you. The independent parliamentary inquiry that I chaired raised the issue that there are numerous regulators in this space and I, like the Minister, think it is an interesting and valid question as to whether Ofcom’s remit should be extended. The two areas where, I think, the issue most arises is one around to whom does one complain if one has problems around content or treatment, and of course ParentPort is a good attempt to try to centralise the complainant process. Perhaps more work could and should be done around publicising that. Indeed we will move onto the internet safety campaign that is being funded by the ISPs. One of the asks is to direct parents where to go, users where to go, when there is a problem.

I think it is right and good that Ofcom has been given an implementation role in the roll out of the policy around filter adoption. One of the questions on the filter process is what does success look like? With a sense of about 30% of the households in the country being those with children one might argue that is a good, soft target though there are not, of course, official targets for this. Asking Ofcom to look at the implementation of that and also perhaps to distinguish between ISPs and the success of their implementation is a very valid thing to do.

Like the Minister I am very minded to think that if we were trying to regulate such a fast moving space in any form we would still be sitting round looking for parliamentary time on the first debate on filtering implementation. Instead of which we have extraordinarily effective and dynamic filter roll out, and indeed companies are competing to offer their customers the best filters and the most family friendly packages. I think an extended role is possible, and I welcome the Minister’s commitment to looking potentially at how that could look.

Q200 Chair: The whole thrust of the pressure that has been placed on internet companies to date has been rather, "Unless you agree to do this then the Government will make sure you have to do it". Do you see any advantage perhaps, and we have a Communications Bill coming, in the Government taking powers rather as the Government took powers in the Digital Economy Act to require ISPs to do something if they failed to do it voluntarily?

Mr Vaizey: As I say I am open to any suggestions and I am looking forward to the Select Committee’s report. I think one has to be careful. If one took too wide a reserve power, if I can put it that way, to "regulate the internet" that would create an element of uncertainty. Businesses need certainty and need to know what regulations they would be required to comply with. We need to know what problem needs to be fixed and why legislation rather than working with organisations and encouraging them would be the answer. As you say, Mr Chairman, I do think that the potential of legislation, and the Prime Minister made that clear on a number of issues in terms of his speech in July and continuing dialogue he has with the industry, is always there.

My message to all the companies in this space is that there is potentially an appetite in Parliament for regulation and that they should not be blind to that and therefore co-operation-and to use a terrible American expression-getting ahead of the curve is a much better place for them to be but they must understand that is, as it were, the mood in Parliament that I don’t think MPs would shy away from legislation should they feel that they were not getting the adequate results that they needed.

Claire Perry: Mr Chairman, just on that point, if I may challenge you slightly, sir. The idea that companies are doing this or else I think perhaps is something that we felt a couple of years ago. I may just be starry-eyed through lack of sleep after yesterday but I genuinely think that the growth of corporate social responsibility among those who are making their living from anything to do with internet activity is really quite impressive. Certainly the British ISPs are well ahead of the ask in terms of delivering and, with yesterday announcements, we can see that some of the global internet companies who perhaps dance to a very different legislative tune have also woken up to that. So again perhaps we are in an awakening of a new dawn of social responsibility on the web, which is to be welcomed, but I would perhaps slightly challenge that the companies are living in fear of regulation. I think we are beyond that point.

Q201 Chair: That is encouraging. Can I turn to the illegal material because that is particularly going to be I suspect more for the Home Office. The area of data retention in order to identify perpetrators of distribution of child abuse images and so on, that strays into what has been a difficult territory for the Government in recent months. Are you satisfied that enough data is being retained and is available to the police to identify people?

Damian Green: I think in this instance it is not a problem of data retention because the civil liberties aspect of the debate does not pertain here. If people are producing, exchanging, storing child abuse images then they are committing a serious criminal act. So, as I say, both legally and morally it does not seem to me that whatever civil libertarian arguments one may wish to adduce about data retention in other spheres apply in this case. I think the key is identifying the images, blocking them, trying to get into some of the more hidden networks, moving on peer-to-peer exchange and so on. Those are the areas where activity both by the industry and by this Government, and indeed Governments overseas with whom we act in concert, is concentrated.

Q202 Chair: Are you getting complete co-operation from every single internet service provider for instance?

Damian Green: In a sense this is more a matter for search engines than ISPs and obviously it is both because the ISPs will host the sites. One of the parts of yesterday’s announcement after the Prime Minister’s summit, was that Google and Microsoft have changed their algorithms so that they can now both identify and block child abuse images. People are using search terms, 100,000 search terms and more, so they now have a dynamic algorithm that will be much more effective than original ideas of just producing search terms and saying block all of those. People who both produce and search for child abuse images are not only evil, they are cunning and it does not take much cunning to decide to change a search term slightly. So you do need the full extent of the technological expertise that is available to companies like Google and Microsoft to ensure that the algorithms can cope with that. The tests that the Child Exploitation Online Protection Centre and other bodies have been running show that these are very effective, what is coming in. Perhaps most encouragingly we learnt from Google yesterday that instead of getting back a child abuse image, getting back a warning page saying, "What you are seeking may well be illegal and do not do this" essentially, that has already reduced the amount of searching by 20%. So it has had a very significant immediate effect.

Q203 Chair: Obviously the progress that has been made through search engines is welcome but a lot of the distribution of material that is being done by the hard core is not going to be done using search engines. Do you think enough has been done and to what extent can you assess the scale of the distribution that is occurring either through peer-to-peer networks or through the dark net?

Damian Green: You are absolutely right, Mr Chairman, there are three elements to it. There is the blocking of the initial search, which we should not dismiss because what it will do is stop people embarking on this particularly horrible journey so it is worth doing in its own right. But you are right, people who are hardened may well be using peer-to-peer or may well indeed, for the worst type of abuse images, now be using now the dark net, in the current jargon. The answer to your question is yes in both those areas. One of the other things we announced yesterday was a pilot project that CEOP, along with the Internet Watch Foundation and the industry companies will be working on to see how we can deal with peer-to-peer. That is obviously different from using a search engine algorithm for reasons the Committee will understand. So that is one project. One of the key areas where the UK/US taskforce, that I will coach here with an Assistant Attorney General of the American Government, will be to bring together the best industry technologists along with the best people, GCHQ and their American equivalents who are actually very good at encrypting and therefore decrypting secret networks, to use all those skills precisely to get in to the dark net and so on. So we are working on all three of those areas.

Chair: I think that probably leads neatly in to Gerry Sutcliffe’s area.

Q204 Mr Sutcliffe: I do not know if Mr Green has had the opportunity to read the evidence that was given to us by the existing Chair of CEOP and the former Director of CEOP, quite opposing views about where CEOP sits. An interesting perspective from the Minister about how CEOP’s role is developing and what are the strengths and what are the weaknesses?

Damian Green: The big change, I agree, having read the evidence, there was a difference in emphasis, if I can put it like that, between the former view and the current view of CEOP, is that CEOP has changed very radically over the past couple of years and has had its most radical change in the last few months by becoming an arm of the National Crime Agency. That is a game-changer for CEOP because it is not now a small very effective, very good outfit, it is now a very effective good outfit that is part of a very large outfit. For example, the NCA will have 4,000 officers, all of whom will be trained in child protection because it has its national links, it has a border arm as well as taking over many of the international policing functions that we need in this country. It can, if you like, look the FBI in the eye and say we can now work with you as an equal, and since this is clearly a global problem it requires global solutions. The fact that CEOP itself is now part of a global size police organisation, the NCA, I think gives it opportunities and access to expertise that it did not have in the past.

Q205 Mr Sutcliffe: Regardless of others’ personal views, what they were unanimous on was that there were about 50,000 potential areas where people could be convicted on internet abuse. What are the resources going to be like for CEOP in terms, and I accept what the Minister is saying in terms of being part of that bigger organisation, there may be some possibilities of it being not downgraded but diluted in terms of its ambition and in terms of where it needs to go?

Damian Green: Absolutely not. You are quite right that the CEOP threat assessment, which it inevitably has to be-it is the best guess we have that is what it is-was up to 50,000 individuals involved in exchanging child abuse images. That was this year’s threat assessment. But, no, I think the exact opposite of dilution is happening. It is being amplified that CEOP now has access to people, expertise, resources, international access in a way that it did not have by itself.

Claire Perry: Just to add some colour to the point the Minister made about exchanging, the interesting behavioural factor of those engaged in this traffic is that they connect. They seek reaffirmation for their desires from connection and in a way that makes the problem easier in that if you are going after a particular node or interrupting a particular site, and we have seen this in several busts recently, you have potentially access to a large group who have connected through, even in the dark internet, a node. Mr Chairman, if I may just throw back a point about the issue of the dark net, I mean things like Tor, The Onion Router, these are often set up by Governments and indeed funded by Governments as a way of providing anonymity to people for various reasons. One thing that would be very interesting to flag up to those Government departments who are setting up these organisations is what the unintended consequences may be. Internet anonymity is both a blessing and a curse, particularly if people are using it for illegal purposes. I don’t think we think enough about that when we are looking schematically at the internet.

Q206 Mr Sutcliffe: If I can just return to the regulation point with Ed and Ofcom. What was interesting this morning, Ofcom’s evidence was that the regulations were not there but one can understand that for the rate of progression and rate of change that takes place. I think the negotiated route is probably the best route to be taken and it is heartening to hear that the ISPs are dealing with that. One of the things that I think that we need to look at is parental involvement in all of this. Again, Ofcom this morning said that they are conducting surveys of parents going back perhaps to 2005. What can the Department do or what can Government do in general terms to try and raise awareness of the potential of these issues affecting vulnerable young children?

Mr Vaizey: As I say, Ofcom was behind setting up ParentPort because it felt you needed a one-stop shop for parents to go to get all the advice they needed. The Secretary of State has asked Ofcom to report regularly, three times over the next year or so, on parental awareness. The other thing that I think has been of benefit from the negotiated route is that we have got the ISPs to agree on a marketing campaign. I took the view these are organisations that know their customers, they have to sell them services, they will know them back to front and, dare I say, each company will have pretty well paid marketing directors who should know their trade. So I think it is a massive opportunity to have the four of them work together. I know what Claire says about internet service providers competing in terms of providing services to parents, that is good because it will mean innovation, but there needs to be a strong degree of co-operation and to have them come together for a marketing campaign is a good thing. My understanding is that it is about £25 million worth of advertising that they are planning to put in place and I think next Spring is the target date. Hopefully there will be a strong and high profile campaign to make parents aware of the filters and the kind of things they can do to protect their children from inappropriate content. It is also important to say that should not be a one-off, you cannot just have a blitz once and then forget about it and assume that people will know about it for ever and a day. One has to keep refreshing and reminding parents of what issues are out there. Again Claire, who has been involved in talking to them, might have something to contribute.

Claire Perry: I think it is interesting that this question comes up as to why is this not a Government sort of public health campaign. Indeed the last Government I believe spent £3 million on the same campaign, which was called, "Zip it, block it, something it" and frankly nobody can remember what it was. It almost sank without trace I would submit, and part of the problem was it was written by us, by people who are not out there every day working at how to get marketing messages to their users. There is a collective commitment of the ISPs to do a £25 million campaign over three years to alert parents to the new filtering technology and what they can do but also the broader issue of conduct questions. Technology only solves some problems, it does solve cyber bullying, it doesn’t solve sexting. There is a lot of stuff out there. I have always felt that parents on the one hand tell us that they are technologically overwhelmed and they do not understand but on the other hand are paying for these contracts. It is like, "If you do not want your children to be sexting then take their mobile phone off them and stop paying for the contract or switch the router off at night". We somehow have abrogated this space to our kids and I think it is time to remind parents that they are part of the solution as well.

I will say that one of the big breakthroughs on technology-forgive me for talking so much-is moving device level filters to the home network level is a massive step forward in making life easier for parents. The ask now for parents to install a filter on every single internet enabled device in their home is a joke. The fact that we have persuaded the ISPs to implement a home network level filter, which we were told initially could not be done for technical reasons, until TalkTalk did it, is massive because it means that parents will only have to set the safety settings once and know that everything in the home is protected. That is an enormous step forward in helping parents in my view.

Q207 Chair: Where are we on your suggestion that there should be a default setting that is a strong filter and that people should then be required to opt in to receive stronger material?

Claire Perry: Having proposed that, we of course had the Government consultation on that, which was very welcome. We heard that people were all over the place on this default on. What we have arrived at is that we are going to ask every account holder in the country, which is 20 million accounts, so this is a very significant ask, whether or not they would like to activate parental filters. Of course two thirds of them will not be parents, so they are going to be asked for something that perhaps is not relevant. We will have the box, "Would you like the filters installed?" pre-ticked. To me, in a way, that satisfies the desire to have people make a choice. We basically present them with an unavoidable choice.

Chair: Who is presenting them?

Claire Perry: I do not know who your internet account is with, but you will have an interrupt at some point in the next year and they will say, "You can now install fantastic family friendly filters would you like to come and install them?" The box will be pre-ticked, "Yes". You can’t just do nothing, you have to choose to get off that page. That to me satisfied this requirement to ask people to choose, which is the classic tension. Do you ask people to choose or do you make the choice for them in quite an elegant way. It was a proposal from the industry that I felt satisfied where we wanted to go. We ask everybody and if you do nothing the filters are automatically installed or you cannot move on.

Chair: Just to be clear, one day over the course of the next year or perhaps-

Claire Perry: One year.

Chair: Right. Every single customer of every single ISP is going to turn on their computer and suddenly get a page come up saying, "Do you want filters?"

Claire Perry: To be clear, the top four ISPs who have around 85% of the market, we would like the others to commit to the same thing. They will ask you in different ways. Some might ask you when you go to query your bill online, some may interrupt your browsing session, which is a first for the UK to do that. This is the commitment that Ofcom will be monitoring. Every household will be contacted and asked whether or not they would like to switch on the filters, and the box, "Yes, please" will be pre-ticked.

Q208 Chair: You have conducted your inquiry, you have talked about the change in attitudes of the companies who are now much more willing to pursue these kinds of things, are you pretty much satisfied that everything you wanted to see happen has happened?

Claire Perry: Well, yes and more in some ways. For example, the Google commitment yesterday, hopefully to be matched by Microsoft, to block illegal returns for search terms globally was huge. So we will basically be protecting the global internet with those changes. However, I think one can never declare victory because the technology moves on at pace, which is why regulation of the space is so difficult. I think we have a solution that is going to be what parents have been asking for, what the country’s been asking for. There may be a new way of accessing the net or a new form of conduct or a new device that pops up in the next couple of years that changes again, which is why having the company self- awareness and involvement is so important. You want them to be thinking about these dangers rather than relying on politicians to constantly be jumping up and saying, "What about this?"

Q209 Chair: Given we expect more and more internet browsing to be done on mobile devices, can I check, are the mobile providers also going to be putting up this question of saying, "Do you wish to fit filters?"

Claire Perry: Interestingly, they already do that. The mobile operators had an opt in system where you had to prove you were over 18 in order to have that filter dropped. They have both expanded that, so Three has become involved and also they have moved to BBFC filtering technology, so their filters are even better and some of the issues like false positives have been removed. In a way we have layers of protection around young people now with mobile devices in the home, and of course with public wi-fi.

Q210 Chair: Damian, can I come back to the hardcore that you are concentrating on, the figure that Gerry quoted and you confirmed, up to 50,000, when we took evidence on this earlier it was suggested to us that a number of those 50,000 who can expect a knock on their door is pretty small. Is that satisfactory or should we not be deploying more resources into this?

Damian Green: Yes, and we are is the brisk answer to that question. That is one of the points of having CEOP as part of the NCA is precisely to improve our enforcement capability. You understand that that 50,000 figure is a pretty broad brush. One of the interesting things, as Claire just alluded to in another context, is that you do tend to break open groups of people, it is not a question of 50,000 isolated individuals. The successful law enforcement operations that take place, both here and overseas, have two characteristics: firstly, lots of people are involved, and, secondly, there may well be lots of people in lots of different countries. You break open networks.

One of the points made at the summit yesterday was that the big sites that are effectively child sex abuse image multiple sites, there are probably not that many of them. There might be 10 or 12 of them based around the world and when you get the law enforcement capability to break into them you may well find very large numbers of people who are attached to them. That is in a sense the way you will get the greater number that I accept we want to see.

Chair: Jim Gamble suggested to us that you could perhaps have special constables who would receive training and then be sent out to look for this material and identify it.

Damian Green: To some extent a lot of that work, in terms of identifying where the images are and from then to the individuals is now being done more and more by the Internet Watch Foundation, and the IWF has a significant increase in its capacity coming with the extra money from the industry. The big change that will happen in the coming months is that instead of being passive and receiving information from the public the IWF will be able to go out there and actively search. The key from that will be to make those searches active, in terms of the individuals as well as the individual images. I think we are meeting this desire to have more pairs of eyeballs looking at these images and from that will flow through the information that hopefully enables us to go and finger the individuals.

Q211 Mr Bradshaw: Damian, do you know how many prosecutions there have been for threatening and abusive behaviour online?

Damian Green: I don’t off the top of my head, I am sorry.

Mr Bradshaw: Would it be possible for you to let us know?

Damian Green: Certainly, yes. We will find out about that.

Q212 Mr Bradshaw: Do you think it is a serious issue?

Damian Green: Yes. It is a different part. I suppose it develops out of cyber bullying in a way, so in a sense it will be networks. Facebook is obviously the biggest one, but we have seen a lot of the other ones where this kind of behaviour has become-prevalence is the wrong word, but we have all become more aware of it simply because that is the arena in which people exchange views and information. Therefore those who are predisposed to be threatening and abusive can now be threatening and abusive online.

Q213 Mr Bradshaw: Do you think it is important that the criminal law is used as a deterrent, in terms of our learning process and how we view this sort of behaviour?

Damian Green: The principle one has to operate is that if something is illegal in the real world then it is illegal online as well. You can cause as much distress to someone by threatening behaviour online as you can offline. To that extent, yes, absolutely. There are various stages one can go through online, and just as criminals can use technology, therefore law enforcement can use technology against criminals as well. Similarly, the social networks know their audience very well, they have the capability at the first sign of this of enabling people to switch individuals off, to report bullying and those kind of instant buttons on screen that you can press. They seem to me to be very good trip wires. At the end you need the criminal law, but I think you need all those trip wires on the way up as well.

Q214 Mr Bradshaw: Do you think the CPS shares the seriousness with which you take this problem?

Damian Green: Yes. I think across law enforcement. The CPS will be the recipient, that is what it is for. It receives reports from the police. Again, you have to go through the procedure, the police have to know it is happening, you have you be able to collect the evidence, they have to receive complaints, they have to have people who are prepared to give evidence, all those kinds of things. If they have cases that are best resolved, as it were, at the end, as a criminal law case then they will approach the CPS.

Mr Bradshaw: I have a question for Mr Vaizey, I do not know if anyone else wants to come in before that?

Chair: Conor, do you want to come in now?

Q215 Conor Burns: I just want to ask very quickly on the social media, do you think the social media companies-Facebook, Twitter in particular-could do more than they are currently doing?

Damian Green: They are now much more aware than they were even a few months ago of the importance of that. I have been dealing with this portfolio for about four or five months and I am struck by the change throughout the industry of some of these very big companies-based and with their origins in the USA with its different tradition-where freedom of speech is such an absolute right that any kind of reduction of it is regarded with extreme suspicion. The journey that the companies, including Facebook and Twitter, have taken over the past few months is to realise that in these areas, particularly with child sex abuse where there is no argument, but also in terms of bullying and abuse that we have seen can lead to tragic incidents of teenage suicide and so on, they do now get it in a way that I suspect, until it was brought to their attention, they didn’t quite.

Q216 Conor Burns: Earlier this morning when we had evidence given from Twitter we were told that there is a report abuse function now on Twitter that was brought in in the end of July. None of us on the Committee who use Twitter were aware that that had been brought in. Do you think there is more they can do to promote how to report abuse, because the reporting of abuse means it is taken more seriously if it is in volume?

Damian Green: Yes, I think that is a very good point. The timing is interesting, it was in July Twitter had got the point. All these companies have the technological capacity to solve these problems, the trick is to get them to recognise them as a problem. There is then a further problem of letting users know. I suspect in that world, once you have established something new, so much of it does get out by osmosis that people just know it. It may reflect our age that we are not instantly aware that something new is available on Twitter now, but you can guarantee that around the world hundreds of billions of Twitter users probably do know about it now, just because that is the way things operate. But, yes, a bit of marketing to the elderly might be useful.

Claire Perry: Could I come in with a couple of points of reassurance and perhaps a suggestion? One is I had a meeting with the Solicitor General on this very issue because obviously I have attracted quite a lot of third party cases on particularly bullying or stalking and harassment issues. He may be worth taking a statement from. I was very reassured that the law has been looked at and expanded wherever necessary to include online conduct as well as offline conduct. Having been on the receiving end of quite a storm of Twitter abuse over the summer relating to my campaign, I have to say I don’t think companies do do enough. I think part of the problem is the anonymity of usage.

I was encouraged with Ask.fm-having spend a lot of time with Hanna Smith’s father, who was one of the young women who did indeed committee suicide-that company did set up a facility where users could choose to be anonymous, but you would know if the user was anonymous when you were exchanging information with them. People Tweet, people post abuse about how they would like to rape you and kill you because they think you do not know who they are. If there was some way of the company knowing and being prepared to verify that identity and to show you that that identity was verified I think it would lead to a diminution in that sort of behaviour. I don’t think the companies do enough and there is a great concern around it, as the Minister said, given the US legal framework within which much of the global companies operate.

Q217 Mr Bradshaw: Do you think it is important that we see some prosecutions so that there is a deterrent out there?

Claire Perry: Yes, I do. I think it is deeply misogynistic as well and it was interesting over the summer because there was a bit of flurry of push back from a number of people in the public over this. I don’t know that we saw prosecutions, we saw cautions and we saw apologies but prosecuting, for what is pretty vile behaviour, would be helpful.

Q218 Chair: On the point you raised, we explored this point with both Facebook and Twitter, Twitter basically told us they are perfectly happy for you to use any alias you choose and there is no requirement for you to tell them who you really are. Facebook said to us that it was their policy that you should use your real name and if somebody told them that somebody was not doing so then they might take action about it, but they would not take any action themselves to check. Are you saying that you think that those two companies should be doing some kind of verification process?

Claire Perry: I think a way around squaring the circle would be for them to indicate to other users those users that have chosen to stay anonymous. I have no issue, as I am sure many users of Facebook and Twitter do, with my identity being public because I am obviously there as a public person and I think most people would feel the same way. If you are aware that so and so has chosen to remain anonymous and sit behind their funny little avatar picture then it is your choice as to whether you engage with them or indeed accept communications from them. Perhaps the companies could give us the choice of who we engage with, but certainly flag up who is choosing to remain anonymous. My sense is that might start to drive down some of this abhorrent behaviour across all the social communities.

Chair: I think it was John Carr who suggested to us that Facebook should be required to verify the identity. I am not sure that is practical, but is this an area where the Government feels more could be done?

Mr Vaizey: Again, this is an area where you can work with the organisations. I certainly think there is mileage in sitting down with the social media companies.

You made the point earlier, Chairman, about the different approach of Twitter and Facebook in terms of verifying identities, and I think there are a whole range of issues that the layman would take as important in a straightforward way, "How do I report abuse? How quickly will they respond to me about the abuse? How can I challenge a decision?" If they decide that the neighbour’s Facebook page saying that you are a such and such and a so and so is fair comment rather than abuse, "How do I appeal that?" Also, do the same principles apply across different social media sites? Obviously individual social media sites should have the freedom to have their own approaches, but I think there is a way of joining things up. Neelie Kroes, the Digital Commissioner, put together some safer social networking principles a few years ago. She asked some of the CEOs that work with her on those to look at them again and update them and my understanding is that they are going to come out with a statement in early 2014. It will be interesting to see that. I certainly do think there is an opportunity for the Government to broker a dialogue between different social media sites and also to examine how effective their protocols are.

Q219 Chair: As I understand it, in terms of identifying criminality the police would have to first of all obtain an order to obtain an IP address off Facebook or Twitter, you then have to go and probably get a second order in order to get the ISP to reveal whose the IP address is. Are you content that that system is sufficient?

Damian Green: It does work.

Chair: It does work?

Damian Green: As Claire said, that there were at the height of the particularly horrible trolling that is going on over the summer, people were visited and the police found out who they were and went and knocked on their door. I just observe as a fact that it is possible to identify individuals, even when they are trying to be anonymous and saying repellent things using things like Twitter.

Q220 Mr Bradshaw: Mr Vaizey, in your evidence you singled out for praise the South West Grid for Learning, part of the UK Safer Internet Centre that happens to be based in Exeter, for the work it does in mediating reports of abuse and threatening behaviour and getting stuff taken down. I do not know whether you are aware that they run on a shoestring and the only money they get is from the European Union. I just wondered whether you thought there might be grounds for doing the same kind of thing with abusive and threatening behaviour as already happens with child abuse and there needs to be some sort of industry scheme to fund an organisation like that, and perhaps some national recognition from Government?

Mr Vaizey: Yes, I have met with the South West Grid for Learning and I am aware that they run as a lean machine and the vast majority of their funding comes from Europe. I would certainly happily look at that. That is a very good suggestion of the wider principle that the ISPs and social media sites support the work that is being done to mitigate some of the harms. The mantra is there is a huge amount of good that comes from the internet, but there is a significant degree of harm. In terms of supporting the Internet Watch Foundation and also working with CEOP, particularly into supplying skilled engineers and technicians, I will certainly look at that on your behalf.

Q221 Conor Burns: I would just like to go back to what we were talking about earlier, about the filtering. Claire, you say that people will have in the course of the next year, possibly their browsing session interrupted and you conveyed that as though that was a very good thing. In pursuing the aim of clamping down on illegal abhorrent content on line that is a good thing, but would you agree that we have to be slightly cautious that we do not step over the line into seeming to be moralising people who do wish to continue to access material that is perfectly legal whatever one’s personal moral judgment about it may be?

Claire Perry: I completely agree. A big question for the ISPs was how do you get to people who most need the filters. To be clear, we know that only four out of 10 families currently employ any sort of filtering device at home. This is Ofcom data. It may have gone up slightly in the last six months, but we have a real problem in that the companies have worked quite hard to generate filters and we know that the take up is very poor. Browser interrupts is one of several ways, other companies who have a very substantial amount of online billing say that they can do it in the "my account" space. But, of course, one question is if there is a kid surfing at that time when the interrupt comes up how does that satisfy the requirement, the adults are the ones who should be putting on the filter. This is for the companies to deliver.

One of the questions, and we did test this quite heavily in focus groups, was do people mind being interrupted or being asked about filters and is it too much to ask them to unclick the "Yes, please" box, which is effectively what you are asking. The response was actually no. The analogy given was that it is a bit like slowing down for a lollypop lady, that you may not have school age children and you might be in a hurry, but because of the general question of child safety you are prepared to slow down and potentially even stop for that school crocodile to pass. Again, there has been a little bit of push back since the announcement was made, but not as much as one might think. I think people think it is reasonable and unobtrusive. One of the challenges was people then said, "Oh, but people will know if I have said I don’t want the filters", which is sort of mad, because in a way they know now. The ISPs know now because you do not request to download them. There has been this sort of air of are you invading people’s privacy. Well, of course Google knows every single keystroke and every single search that everybody is conducting, so it is slightly bizarre that people are prepared to give up their anonymity to private companies who sell it and are concerned that somehow they will end up on some sort of list. There is no list, there is no Government mandate of this, this is the companies basically interacting with their customers and saying, "Would you like to download the filters? Yes, please or no, thanks". I think it is a very valid point.

By the way, the tone of this has always been about among consenting adults not an issue, this is absolutely fine. Although of course with closing the rape pornography loophole there are some rather unpleasant areas of currently legal material that the Government rightly has decided should be illegal. Again, it does not seem to be problem so far, and I think we need to see with implementation and Ofcom’s feedback if people do think it is unreasonable, because it may be companies have to revisit how they market their products.

Q222 Chair: Can I just pursue this point, which may be best for Damian, of material that is just about legal for adults, so it is pretty hardcore but it is legal. It was pointed out to us that a successful prosecution had been brought on the basis that the material could be viewed by children. They did not have to prove that it had been, just the fact that children could view it had allowed a successful prosecution. Is that a sort of route you think could be used again?

Damian Green: As you say, it is extremely unusual and Claire has made the point about pornography depicting rape, which I think one can regard in the same way as hate speech if you like, speech that will incite violence or something like that, because that might incite criminal behaviour. The issue that I know DCMS have been dealing with is pornographic content that is entirely legal and that nobody would be disturbed, or most people would not be disturbed with adults viewing. But it is age inappropriate and the ability to stop normally young boys repeatedly and incessantly watching that kind of material is both a technical issue and a policy issue of what you should stop and where you draw the lines. It is not criminal material so it does not fall under the criminal law, but it is clearly age inappropriate for normally young boys.

Chair: I think that was the point that was put to us, which was that it could be deemed criminal under the Obscene Publications Act if insufficient safeguards have not been put in place to prevent young people accessing it.

Mr Vaizey: It is an interesting issue. Obviously these are the points put forward by ATVOD, the Association for Television on Demand, and I think they have made some significant progress on this. As was indicated earlier by Claire, we are moving to outlaw R18 material. At the moment ATVOD can step in where sites are offering R18 material, but we want to remove any ambiguity because there is some discussion about whether one has to show that there has been harm, and obviously rape porn as well.

The question arises about the Obscene Publications Act and whether or not a website that effectively doesn’t have any mechanism for identifying the age of the user and no kind of pay wall that you have to get behind to access the material breaches the Obscene Publications Act. I am not sure it is for me to say in terms of whether sufficient prosecutions have been conducted, because that would obviously be something for the CPS. I do know that ATVOD, in a rather imaginative way, is looking at working with financial services providers and to say to them-because obviously a lot of these websites offer you free access and then hope to make you pay later-that they should not provide payment facilities for sites that do not offer robust age verification. That is issue number one.

Obviously nothing should stop us doing the right thing, in terms of prosecutions or clarifying the law. Nevertheless we do have to be aware that a lot of these sites do not provide suitable identification as to who owns them, and again ATVOD is suggesting that we clamp down on those sites by denying them their financial support. That is something we need to continue to work closely with ATVOD on, they do have a serious point. We are also in UKCCIS-the United Kingdom Council for Child Internet Safety-looking at robust age verification measures, technical measures, whether it is the unique pupil identifier that is used in schools to enable children to get access to the web, or the digital wallet that I think is a line pursued in Demark to ensure that you can identify the age of someone by effectively giving them a digital wallet to use on the internet. I also think, and ATVOD say it in their evidence submitted to you, that there is a role for the European Union-sorry to keep coming back to Europe-of working across all these territories, because again one of the issues here is what is acceptable in this country or not acceptable in this country, in terms of culture, if you like, is acceptable in other countries across the EU. It is something we need to continue to look at and I think it is an important issue.

Chair: Thank you. I think we have exhausted the questions, thank you very much.

Prepared 25th November 2013