Online Safety Bill

Baroness Bull Excerpts

Committee stage
Tuesday 9th May 2023 Lords Chamber

My second question is similar. We see now around the world—it is not available in the UK—that Meta has a verified subscription, for which you can pay around $15 per month. It is being piloted in the US as we speak. Again, I ask whether that satisfies the duty in terms of it being affordable to the average UK user. I am concerned that most UK social media users will not be able to afford £180 per social media account for verification. If that ends up being Meta’s UK offering, many users would not be given a proper, meaningful chance to be verified. What powers are there in the Bill for Ofcom to send Meta back and offer something else? So my questions really are about what “verified” means in terms of the Bill.

[Baroness Bull (CB)]

My Lords, I rise to speak to Amendment 141 in the names of the noble Lords, Lord Stevenson and Lord Clement-Jones. Once again, I register the support of my noble friend Lady Campbell of Surbiton, who feels very strongly about this issue.

Of course, there is value in transparency online, but anonymity can be vital for certain groups of people, such as those suffering domestic abuse, those seeking help or advice on matters they wish to remain confidential, or those who face significant levels of hatred or prejudice because of who they are, how they live or what they believe in. Striking the right balance is essential, but it is equally important that everyone who wishes to verify their identity and access the additional protections that this affords can do so easily and effectively, and that this opportunity is open to all.

Clause 57 requires providers of category 1 services to offer users the option to verify their identity, but it is up to providers to decide what form of verification to offer. Under subsection (2) it can be “of any kind”, and it need not require any documentation. Under subsection (3), the terms of service must include a “clear and accessible” explanation of how the process works and what form of verification is available. However, this phrase in itself is open to interpretation: clear and accessible for one group may be unclear and inaccessible to another. Charities including Mencap are concerned that groups, such as people with a learning disability, could be locked out of using these tools.

It is also relevant that people with a learning disability are less likely to own forms of photographic ID such as passports or driving licences. Should a platform require this type of ID, large numbers of people with a learning disability would be denied access. In addition, providing an email or phone number and verifying this through an authentication process could be extremely challenging for those people who do not have the support in place to help them navigate this process. This further disadvantages groups of people who already suffer some of the most extensive restrictions in living their everyday lives.

Clause 58 places a duty on Ofcom to provide guidance to help providers comply with their duty, but this guidance is optional. Amendment 141 aims to strengthen Clause 58 by requiring Ofcom to set baseline principles and standards for the guidance. It would ensure, for example, that the guidance considers accessibility for disabled as well as vulnerable adults and aligns with relevant guidance on related matters such as age verification; it would ensure that verification processes are effective; and it would ensure that the interests of disabled users are covered in Ofcom’s pre-guidance consultation.

Online can be a lifeline for disabled and vulnerable adults, providing access to support, advice and communities of interest, and this is particularly important as services in the real world are diminishing, so we need to ensure that user-verification processes do not act as a further barrier to inclusion for people with protected characteristics, especially those with learning disabilities.

[Baroness Fox of Buckley (Non-Afl)]

My Lords, the speech of the noble Baroness, Lady Buscombe, raised so many of the challenges that people face online, and I am sure that the masses who are watching parliamentlive as we speak, even if they are not in here, will recognise what she was talking about. Certainly, some of the animal rights activists can be a scourge, but I would not want to confine this to them, because I think trashing reputations online and false allegations have become the activists’ chosen weapon these days. One way that I describe cancel culture, as distinct from no-platforming, is that it takes the form of some terrible things being said about people online, a lot of trolling, things going viral and using the online world to lobby employers to get people sacked, and so on. It is a familiar story, and it can be incredibly unpleasant. The noble Baroness and those she described have my sympathy, but I disagree with her remedy.

An interesting thing is that a lot of those activities are not carried out by those who are anonymous. It is striking that a huge number of people with large accounts, well-known public figures with hundreds of thousands of followers—sometimes with more than a million—are prepared to do exactly what I described in plain sight, often to me. I have thought long and hard about this, because I really wanted to use this opportunity to read out a list and name and shame them, but I have decided that, when they go low, I will try to go at least a little higher. But subtweeting and twitchhunts are an issue, and one reason why we think we need an online harms Bill. As I said, I know that sometimes it can feel that if people are anonymous, they will say things that they would not say to your face or if you knew who they were, but I think it is more the distance of being online: even when you know who they are, they will say it to you or about you online, and then when you see them at the drinks reception, they scuttle away.

My main objection, however, to the amendment of the noble Baroness, Lady Buscombe, and the whole question of anonymity in general is that it treats anonymity as though it is inherently unsafe. There is a worry, more broadly on verification, about creating two tiers of users: those who are willing to be verified and those who are not, and those who are not somehow having a cloud of suspicion over them. There is a danger that undermining online anonymity in the UK could set a terrible precedent, likely to be emulated by authoritarian Governments in other jurisdictions, and that is something we must bear in mind.

On evidence, I was interested in Big Brother Watch’s report on some analysis by the New Statesman, which showed that there is little evidence to suggest that anonymity itself makes online discourse more febrile. It did an assessment involving tweets sent to parliamentarians since January 2021, and said there was

“little discernible difference in the nature or tone of the tweets that MPs received from anonymous or non-anonymous accounts. While 32 per cent of tweets from anonymous accounts were classed as angry according to the metric used by the New Statesman, so too were 30 per cent of tweets from accounts with full names attached.18 Similarly, 5.6 per cent of tweets from anonymous accounts included swear words, only slightly higher than the figure of 5.3 per cent for named accounts.”

It went through various metrics, but it said, “slightly higher, not much of a difference”. That is to be borne in mind: the evidence is not there.

In this whole debate, I have wanted to emphasise freedom as at least equal to, if not of greater value than, the safetyism of this Bill, but in this instance, I will say that, as the noble Baroness, Lady Bull, said, for some people anonymity is an important safety mechanism. It is a tool in the armoury of those who want to fight the powerful. It can be anyone: for young people experimenting with their sexuality and not out, it gives them the freedom to explore that. It can be, as was mentioned, survivors of sexual violence or domestic abuse. It is certainly crucial to the work of journalists, civil liberties activists and whistleblowers in the UK and around the world. Many of the Iranian women’s accounts are anonymous: they are not using their correct names. The same is true of Hong Kong activists; I could go on.

Anyway, in our concerns about the Bill, compulsory identity verification means being forced to share personal data, so there is a privacy issue for everyone, not just the heroic civil liberties people. In a way, it is your own business why you are anonymous—that is the point I am trying to make.

There are so many toxic issues at the moment that a lot of people cannot just come out. I know I often mention the gender-critical issue, but it is true that in many professions, you cannot give your real name or you will not just be socially ostracised but potentially jeopardise your career. I wrote an article during the 2016-17 days called Meet the Secret Brexiteers. It was true that many teachers and professors I knew who voted to leave had to be anonymous online or they would not have survived the cull.

Finally, I do not think that online anonymity or pseudonymity is a barrier to tracking down and prosecuting those who commit the kind of criminal activity on the internet described, creating some of the issues we are facing. Police reports show that between 2017-18, 96% of attempts by public authorities to identify anonymous users of social media accounts, their email addresses and telephone numbers, resulted in successful identification of the suspect in the investigation; in other words, the police already have a range of intrusive powers to track down individuals, should there be a criminal problem, and the Investigatory Powers Act 2016 allows the police to acquire communications data—for example, email addresses or the location of a device—from which alleged illegal anonymous activity is conducted and use it as evidence in court.

If it is not illegal but just unpleasant, I am afraid that is the world we live in. I would argue that what we require in febrile times such as these is not bans or setting the police on people but to set the example of civil discourse, have more speech and show that free speech is a way of conducting disagreement and argument without trashing reputations.

--- Later in debate ---

[Lord Stevenson of Balmacara (Lab)]

My Lords, I have slightly abused my position because, as the noble Baroness has just said, this is a rather oddly constructed group. My amendments, which carve great chunks out of the Bill—or would do if I get away with it—do not quite point in the same direction as the very good speech the noble Baroness made, representing of course the view of the committee that she chairs so brilliantly. She also picked out one or two points of her own, which we also want to debate. It therefore might be easier if I just explain what I was trying to do in my amendments; then I will sit down and let the debate go, and maybe come back to pick up one or two threads at the end.

In previous Bills—and I have seen a lot of them—people who stand up and move clause stand part debates usually have a deeper and more worrying purpose behind the proposition. Either they have not read the Bill and are just trying to wing it, or they have a plan that is so complex and deep that it would probably need another whole Bill to sort it out. This is neither of those approaches; it is done because I want to represent the views mainly of the Joint Committee. We had quite a lot of debate in that committee about this area, beginning with the question about why the Bill—or the White Paper or draft Bill, at that stage—used the term “democratic importance” when many people would have used the parallel term “public interest” to try to reflect the need to ensure that matters which are of public good take place as a result of publication, or discussion and debate, or on online platforms. I am very grateful that the noble Lord, Lord Black, is able to be with us today. I am sure he will recall those debates, and hopefully he will make a comment on some of the work—and other members of the committee are also present.

To be clear, the question of whether Clauses 13, 14, 15 and 18 should stand part of the Bills is meant to release space for a new clause in Amendment 48. It is basically designed to try to focus the actions that are going to be taken by the Bill, and subsequently by the regulator, to ensure that the social media companies that are affected by, or in scope of, the Bill use, as a focus, some of the issues mainly related to “not taking down” and providing an appeal mechanism for journalistic material, whether that is provided by recognised news publishers or some other form of words that we can use, or it is done by recognised journalists. “Contentious” is an overused word, but all these terms are difficult to square away and be happy with, and therefore we should have the debate and perhaps reflect on that later when we come back to it.

The committee spent quite a lot of time on this, and there are two things that exercised our minds when we were working on this area. First, if one uses “content of democratic importance”, although it is in many ways quite a clever use of words to reflect a sensibility that you want to have an open and well-founded debate about matters which affect the health of our democracy, it can be read as being quite limiting. It is very hard to express—I am arguing against myself here—in the words of a piece of legislation what it is we are trying to get down to, but, during the committee’s recommendations, we received evidence that the definition of content of democratic importance was wider, or more capable of being interpreted as wider, than the scope the Government seem to have indicated. So there is both a good side and a bad side to this. If we are talking about content which is, or appears to be, specifically intended to contribute to the democratic political debate of the United Kingdom, or a part or area of the United Kingdom, we have got to ask the Minister to put on the record that this also inclusive of matters which perhaps initially do not appear necessarily to be part of it, but include public health, crime, justice, the environment, professional malpractice, the activities of large corporations and the hypocrisy of public figures when that occurs. I am not suggesting this is what we should be doing all the time, but these are things we often read about in our papers, and much the better off we are for it. However, if these things are not inclusive and not well rooted in the phrase “content of democratic importance”, it is up to the Government to come forward with a better way of expressing that, or perhaps in debate we can find it together.

I have some narrow questions. Are we agreed that what is currently in the Bill is intended specifically to contribute to democratic political debate, and is anything more needed to be said or done in order to make sure that happens? Secondly, the breadth of democratic political debate is obviously important; are there any issues here that are going to trip us up later when the Government come back and say, “Well, that wasn’t what we meant at all, and that doesn’t get covered, and therefore that stuff can be taken down, and that stuff there doesn’t have to be subject to repeal”? Are there contexts and subjects which we need to talk about? This is a long way into the question of content of democratic importance being similar or limited to matters that one recognises as relating to public interest. I think there is a case to be argued for the replacement of what is currently in the Bill with a way of trying to get closer to what we now recognise as being the standard form of debate and discussion when matters, which either the Government of the day or people individually do not like, get taken up and made the subject of legal discussion, because we do have discussions about whether or not it is in the public interest.

We probably do not know what that means. Therefore, a third part of my argument is that perhaps this is the point at which we try to define this, even though that might cause a lot of reaction from those currently in the press. In a sense, it is a question that needs to be resolved. Maybe this is or is not the right time to do that. Are the Government on the same page as the Joint Committee on this? Do they have an alternative and is this what they are trying to get across in the Bill?

Can we have a debate and discussion in relation to those things, making it clear that we want something in the Bill ensuring that vibrant political debate—the sort of things the noble Baroness was talking about on freedom of expression, but in a broader sense covering all the things that matter to the body politic, the people of this country—is not excluded by the Bill? That was the reason for putting down a raft of rather aggressive amendments. I hope it has been made clear that that was the case. I have other things that I would like to come back to, but I will probably do that towards the end of the debate. I hope that has been helpful.

[Baroness Bull (CB)]

My Lords, I will speak to the amendments in the name of the noble Baroness, Lady Stowell, to which I have added my name. As we heard, the amendments originally sat in a different group, on the treatment of legal content accessed by adults. Noble Lords will be aware from my previous comments that my primary focus for the Bill has been on the absence of adequate provisions for the protection of adults, particularly those who are most vulnerable. These concerns underpin the brief remarks I will make.

The fundamental challenge at the heart of the Bill is the need to balance protection with the right to freedom of expression. The challenge, of course, is how. The noble Baroness’s amendments seek to find that balance. They go beyond the requirements on transparency reporting in Clause 68 in several ways. Amendment 46 would provide a duty for category 1 services to maintain an up-to-date document for users of the service, ensuring that users understand the risks they face and how, for instance, user empowerment tools can be used to help mitigate these risks. It also provides a duty for category 1 services to update their risk assessments before making any “significant change” to the design or operation of their service. This would force category 1 services to consider the impact of changes on users’ safety and make users aware of changes before they happen, so that they can take any steps necessary to protect themselves and prepare for them. Amendment 47 provides additional transparency by providing a duty for category 1 services to release a public statement of the findings of the most recent risk assessment, which includes any impact on freedom of expression.

The grouping of these amendments is an indication, if any of us were in doubt, of the complexity of balancing the rights of one group against the rights of another. Regardless of the groupings, I hope that the Minister takes note of the breadth and depth of concerns, as well as the willingness across all sides of the Committee to work together on a solution to this important issue.

[Viscount Colville of Culross (CB)]

My Lords, I put my name to Amendment 51, which is also in the name of the noble Lords, Lord Stevenson and Lord McNally. I have done so because I think Clause 15 is too broad and too vague. I declare an interest, having been a journalist for my entire career. I am currently a series producer of a series of programmes on Ukraine.

This clause allows journalism on the internet to be defined simply as the dissemination of information, which surely covers all posts on the internet. Anyone can claim that they are a journalist if that is the definition. My concern is that it will make a nonsense of the Bill if all content is covered as journalism.

I support the aims behind the clause to protect journalism in line with Article 10. However, I am also aware of the second part of Article 10, which warns that freedom of speech must be balanced by duties and responsibilities in a democratic society. This amendment aims to hone the definition of journalism to that which is in the public interest. In doing so, I hope it will respond to the demands of the second part of Article 10.

It has never been more important to create this definition of journalism in the public interest. We are seeing legacy journalism of newspapers and linear television being supplanted by digital journalism. Both legacy and new journalism need to be protected. This can be a single citizen journalist, or an organisation like Bellingcat, which draws on millions of digital datapoints to create astonishing digital journalism to prove things such as that Russian separatist fighters shot down flight MH17 over Ukraine.

The Government’s view is that the definition of “in the public interest” is too vague to be useful to tech platforms when they are systematically filtering through possible journalistic content that needs to be protected. I do not agree. The term “public interest” is well known to the courts from the Defamation Act 2013. The law covers the motivation of a journalist, but does not go on to define the content of journalism to prove that it is in the public interest.