Companies such as Google/YouTube, Facebook, Airbnb and Uber benefit from strong network effects - the phenomenon through which their services become increasingly attractive as more people use them, to the extent that it becomes near impossible for any other company to compete - or for governments to challenge them.
Joanna Bryson, in a fascinating blog about AI, notes that the consequential "winner take all" nature of Internet commerce is also powered by the relatively low cost of transport of the product - i.e. the electronic data. We cannot travel far to buy our groceries, however excellent the shops may be in another country. But we can buy entertainment, social media, search and so on from anywhere in the world. The BBC, for instance, are therefore very concerned about competition from the west coast of America - Netflix, YouTube and the rest
But there is a real danger that these newly dominant companies will come under less and less pressure to remain efficient, or to innovate, or to be good corporate citizens.
The following paragraphs summarise the principal criticisms of the behaviour and culture of the Technology Giants. Particular examples of their alleged failings may be found towards the end of this web page.
They are said to export the American 'see you in court' approach to regulation. Regulation, they believe, is for 'the guidance of wise men and the observance of fools'. Uber, for instance, is said by the New York Times to have 'tended to barrel into new markets by flouting local laws, part of a combative approach to expand globally'. And according to the same newspaper, Uber developed an app which ensured that city officials were not able to book a car and so scrutinise the service. The Times called them the 'Predators of the internet who do as they please'. Facebook certainly seemed unperturbed when it was fined €110m (£94m) by the EU for providing misleading information about its 2014 takeover of WhatsApp.
There are growing concerns that the Internet Giants - and Google in particular - are exploiting and abusing their dominant position by squashing potential competitors before they get a chance to establish themselves.
A commentator called Kassandra commented that few if any had appreciated that data can be weaponized. 'The internet is a fantastic resource. Social media is not. It was dreamed up by naive American college kids who had no sense that personal information cannot only be used for advertising. It can also be weaponized. Had they bothered to read or shown any interest in totalitarianism, they would have known. But they came of age in the early 2000's when the march of liberal democracy and globalization seemed unstoppable. Or maybe they just didn't care about politics. Maybe they were simply oblivious about the world around them.'
Evan Williams — a Twitter founder and co-creator of Blogger — was reported as follows in 2017:-
“I think the internet is broken ... And it’s a lot more obvious to a lot of people that it’s broken.” People are using Facebook to showcase suicides, beatings and murder, in real time. Twitter is a hive of trolling and abuse that it seems unable to stop. Fake news, whether created for ideology or profit, runs rampant. Four out of 10 adult internet users said in a Pew survey that they had been harassed online. “I thought once everybody could speak freely and exchange information and ideas, the world is automatically going to be a better place,” Mr. Williams says. “I was wrong about that.”
John Lanchester argued (Sunday Times October 2017 )that Alphabet/Google/YouTube and Facebook/Instagram were both walking the fine line between ethical growth and making even more money by 'doing evil'. Quoting an industry insider:-
“YouTube knows they have lots of dirty things going on and are keen to try and do some good to alleviate it, ... “Terrorist and extremist content, stolen content, copyright violations. That kind of thing. But Google, in my experience, knows that there are ambiguities, moral doubts, around some of what they do, and at least they try to think about it. Facebook just doesn’t care. When you’re in a room with them you can tell. They’re” — he took a moment to find the right word — “scuzzy.”
Some commentators have drawn attention to the tendency of some technology to encourage 'extremism', with YouTube in particular being accused of driving people from reasonable videos to ever more extreme content on those subjects because it captures their attention. “Truth” doesn’t enter into it. Google’s mission statement is to “organise the world’s information and make it accessible”. It doesn’t include anything about separating accuracy from untruth.
It must certainly be increasingly irritating (to use a mild word) for the established media to watch Facebook and others publish material which would lead to others being brought low. Imagine what would happen if the BBC decided that they would allow the public to broadcast murders and suicides live on one of their channels.
Nick Srnicek has described data as the modern equivalent of oil - essential to the modern economy and maybe needing something like the 1911 anti-trust break-up of Rockefeller's Standard Oil. And Facebook's rather cavalier sharing of data with Cambridge Analytica blew up in its face in March 2018 - see further below.
Joanna Bryson has raised an interesting tax point in her AI blog. "Every interaction with Google or Facebook is a barter of information. With no money changing hands, there is no tax revenue ... [We may need] to turn instead to taxing wealth ... which is becoming harder [to hide] because of the information age."
She also believes that we are in danger, as in the 19th/early 20th Centuries, of allowing technology to increase inequality and hence political polarisation.
Google/YouTube, Facebook, Twitter etc. all claim to be mere platforms, passively hosting content that they are unwilling to assess. They assert that they "enforce our policies rigorously and when a violation is brought to our attention we take swift action". But they do not proactively look for violations. Their algorithms to some extent choose what their readers get to see, and the companies are financed by advertising, much like traditional media companies. They distribute fake news and other attention-grabbing content, regardless of its quality, veracity or decency, including material which encourages terrorism. It is well established that detailed guides showing how to make nail bombs, and ricin poison are freely available on Facebook and YouTube.
There is also the danger that algorithmic news poses a risk to democracy as 1.2 billion daily Facebook users, for instance, mainly listen to louder echoes of their own voices - the so-called filter bubble. But Facebook’s relatively modest efforts to curb misinformation have been met with fury on the right, with Breitbart and The Daily Caller fuming that Facebook had teamed up with liberal hacks motivated by partisanship. If Facebook were to take more significant action, like hiring human editors, creating a reputational system or paying journalists, the company would instantly become something it has long resisted: a media company rather than a neutral tech platform. Facebook’s personalisation of its news feeds would perhaps be less of an issue if it were not crowding out every other source – but as a result it is beginning to appear that Facebook must be responsible for finding solutions to its problems.I was going to give the last word, in this section, to Facebook's apology - see box on the right. But then I read Zeynep Tufekci's excoriating piece in Wired in which she drew attention to the huge number of times that Mark Zuckerberg and his colleagues have previously apologised for their actions - but only after being challenged - including the company's 2011 admission to the Federal Trade Commission that it had deceptively promised privacy to its users and then repeatedly broken that promise. It very much looks as though Facebook's unacceptable behaviour is hard-wired into its culture, and nothing will change until politicians decide that its apologies are insufficient and that a firm response is needed. So ...
How Might Governments Respond - with Regulation?
Governments hesitate to regulate these companies because:
a) the companies are immensely powerful, and their founders are politically powerful,
b) politicians would be accused of limiting freedom of speech, and
c) internet-based applications and communications are a real and very valuable force for good, both economically and politically (in the way in which they facilitate political and social engagement).
On the other hand, Facebook etc.'s data handling rules are far too complex for most of us to understand. Governments may therefore need to establish and enforce 'technology safety codes', just as they enforce food safety and building regulations, recognising that most of us to know, when we enter a restaurant, whether we are in danger of getting food poisoning or having the ceiling fall down on us.
In more detail ...
The companies' huge size and international reach certainly make it very difficult for any individual regulator to tackle them with any prospect of success, partly because their businesses are so complex and partly because their resources enable them to out-gun all but the most persistent and well-funded regulators.
It is at first sight hard to see how competition law could be brought to bear. The companies' services are generally provided free of charge for the general public - although we do in return give them lots of our collectively valuable personal data. And they compete hard to attract their paying customers, the advertisers. They would certainly run into trouble if they started to abuse their dominant position by making it difficult for new entrants to compete with them, but there is as yet no sign that they need to do so, given their advantages mentioned in the first paras above. The main exception seems to be Google which has been accused by the European Commission of abusing its dominant position in 'search'.
On the other hand, the German competition regulator - which tends to be a bit tougher than some others - has pointed out that, if Facebook has a dominant position, then the consent that the user gives for his/her data to be exploited is no longer voluntary. (S)he has to use Facebook if (s)he wants to use a social network.
Building on this, maybe the industries themselves need to be analysed in a different way? Rather than worrying about concentration in particular markets, such as search, maybe regulators should be concerned about customer lock-in - the unattractive obverse of strong network effects. And interoperability? Should we be worried, for instance, that customers are increasingly committed to one or other of the Apple and Android domains.
The companies also claim that it would be too technically complex to tackle their content problems effectively, arguing that they cannot be expected to vet/censor everything that it posted on their sites. Here are some thoughts on this tricky issue:
- They have shown themselves to be adept at addressing copyright violation when it suits them.
- It is of course true that the companies cannot rely on their human editors, even though Google/YouTube intend to employ 10,000 'reviewers' (not 'editors') by the end of 2018. As of late 2017, more than 400 hours of footage is uploaded to YouTube's main site every minute.
- As of 2018, around 6 million YouTube videos are uploaded every day. If a tool could be developed that identified objectionable material with amazing 99.995% accuracy, it would still unnecessarily delete 300 videos a day. Is this a reasonable price to pay in order to censor the occasional terrorist video?
- But JP Morgan Chase announced in January 2018 that they had created an algorithm with 17 layers to identify and filter racist and terrorist YouTube clips to stop them appearing alongside the bank's online advertisements.
- Another approach might be to monitor the comments posted underneath a video for signs of outrage.
- Artificial intelligence cannot at present distinguish between the the video of the of murder of Robert Godwin (which should not be shown - see above) and that of a fatal shooting by a police officer - which surely should be made available to the public.
The overriding reason, of course, is that they fear opening themselves up to endless and expensive litigation if they admit to being publishers. There really is no obvious reason reason why the giant companies should be exempt from the 'free speech' limitations that apply to the rest of us, and to other media companies. But active (as distinct from responsive) moderation of their content would likely open Facebook, for instance, to potentially massive legal liabilities.
Optimists have suggested that the companies' consumers may become a sort of regulator if they begin to desert the platforms in protest against their content or behaviour. But there is little sign of this happening, and many users are in effect locked into the products as a result of the strong network effects mentioned above. It can only add to commentators' concerns that it is well known in Silicon Valley that 'If you are not paying for it then you are the product'.
But pressure is now being applied by the companies that fund the Giants via advertisements etc. There were two particularly interesting developments in the Summer of 2017. First, Google and Facebook admitted that they could if necessary remove unpleasant content, particularly in order to ensure that it did not appear next to advertisements from its blue-chip advertisers. Second, they discovered that they could, after all, remove illegal streaming of football matches, as well as 30,000 video clips, when required to do so by a court order sought by the Premier League. So 'the Giants' can censor when necessary. It remains to be seen whether the UK or any other government will have the temerity to make similar demands to those made by large commercial interests.
Maybe the EU will collectively be strong enough to take the companies on, encouraged by Germany which has a tradition of strong form-based regulation? A start has been made by the EU's Competition Commissioner who has fined Google for abusing its dominant position in 'search'. But competition authorities are usually well resourced and not scared of companies like Google. It is a lot harder for smaller regulators to tackle bad behaviour which is not anti-competitive.
The European Commission nevertheless fired a shot across Facebook's and Twitter's bows in September 2017 when it issued a proclamation that the companies must do more to remove 'illegal content inciting hatred, violence and terrorism', and threatening additional measures as necessary. But Bird & Bird's Graham Smith pointed out that the EU-preferred systems relied upon 'trusted flaggers' of illegal etc. content, but did not include provisions to ensure that the trusted flaggers were making good decisions and/or should be trusted with such censorship power.
(In his speech 'Antitrust in the Tech Sector' Senator Orrin Hatch argues that it is a mistake to ask competition authorities to take on extra regulatory duties which have little to do with the effectiveness of competition. See also item 14 here. )
The Home Affairs Select Committee published a report on Hate Crime in early 2017. The committee strongly criticised social media companies for failing to take down and take sufficiently seriously illegal content – saying they were "shamefully far" from taking sufficient action to tackle hate and dangerous content on their sites. The Committee recommended that the Government should assess whether failure to remove illegal material is in itself a crime and, if not, how the law should be strengthened. They recommended that the Government should also consult on a system of escalating sanctions to include meaningful fines for social media companies which fail to remove illegal content within a strict time frame.
Maybe an international convention is the way forward, as suggested by The Good European:
The notion that the internet is 'beyond jurisdictions' has passed its sell by date. It was always a nonsense and by giving it credence a quarter of a century's worth of legislators, especially in America, have done the potential of the world wide web a great disservice. That's twenty five wasted years and twenty five years of damage to what should have been one of the great inventions, now too much a tainted and lawless badlands instead of being the wholesome resource of promise. What is needed is for every jurisdiction across the planet to sign up to and jointly police a basic convention. To those who say that would be impossible, I offer exhibit A: International Maritime Law. Its been around in one form or another since at least Roman times and is how we regulate both inshore waters and the open oceans. These days it's overseen by the International Maritime Organization, a sub division of the United Nations and headquartered in London UK. Every jurisdiction has a seat at the table.
The FT carried this interesting report in in April 2017:
Facebook blamed human error for its failure to remove dozens of images and videos depicting child pornography after they were flagged to the company. The Times newspaper reported that it had alerted the social media giant, using a dummy Facebook profile, to potentially illegal content that was posted on to its website by users, including images of an allegedly violent sexual assault on a child and cartoons of child abuse. The British newspaper accused Facebook of failing to remove many of the images but the company said this was because of human oversight and that its reviewers should have spotted that the content should be removed. ...“We are always looking for other ways to use automation to make our work easy, but ultimately content review is manual,” said Monika Bickert, head of global policy management at Facebook in a past interview with the Financial Times.
This is a clear admission of failure of quality control. I have no doubt that most media organisations would have run into severe trouble - and probably been prosecuted - if they had behaved like Facebook. But I am not aware that any formal action has yet been taken against the company.
Later that month, a man killed his daughter and then himself whilst live-streaming his actions on Facebook Live. The company had previously streamed a video which showed that a man chose 74 year old Robert Godwin Snr at random and then shot him. Again, the company's attitude seemed to be that its customers' wish to view live video outweighed any moral or other pressure that the content should be pre-approved. Presumably nothing will change unless and until someone prominent is killed by someone seeking Facebook fame - though its COO was reported to have had a moment of doubt - see box on right.
The Times reported in November 2017 that a YouTube channel Toy Freaks - where a parent posted hundreds of videos of two girls in distress - was not removed until it had been reported to the channel multiple times over more than a year.
And there were numerous complaints of inappropriate material appearing on YouTube Kids.
Uber admitted in November 2017 that it had concealed a data breach in October 2016 in which hackers accessed the personal details of 57 million customers worldwide. Instead of reporting the breach to the authorities, the company paid the hackers $100,000 to keep silent and deleted the data, although of course no-one, least of all Uber, can be sure that they did so.
Facebook was in the firing line again in March 2018 when UK, EU and US regulators and politicians began to investigate reports that Cambridge Analytica, a data analysis firm employed by Donald Trump’s presidential campaign, mined the personal data of 50m Facebook users to create profiles to target them in elections. It was alleged that the company broke Facebook’s rules by using data collected solely for research, but it seems astonishing that Facebook handed over the data so willingly and without subsequently auditing its use. Facebook's Head of Security's comment is on the right. The Guardian's Alex Hern's response was devastating:
If I walk into a hospital and tell them I’m the butt inspector and they should give me pictures of all their patients’ butts and they do, that’s a data breach, regardless of whether the hospital obtained the butt pics with patient consent.
The difficulty for Facebook was that its business depends on sharing information about its users with others, including advertisers and the many companies whose apps are accessed via, and interact with, Facebook. As Hugo Rifkind pointed out: "the idea of a) harvesting data and then, b) selling a service based on having it is not some nefarious perversion of utopian innocence ... Rather, it is, and always has been, what Facebook is all about. It is what the whole damn website is for."
Facebook's lawyers will have been acutely aware of the danger of crossing some invisible line and so breaching data protection legislation. But if they are shown to have drawn the line in the wrong place, or failed to guard the boundary effectively enough, then Facebook could be in serious trouble.
Stig Abell (former Director of the Press Complaints Commission and also former Managing Editor of the Sun newspaper) complained to Twitter in 2018 that a man had started posting about how he knew where Mr Abell lived and - in detail - how he wanted to rape Mr Abell's wife. Twitter said that it was not a breach of its rules, though - under pressure - it later changed its mind. Hugo Rifkind pointed out that Twitter had told a Parliamentary committee that it investigates c.6.4 million dodgy or abusive accounts each week - with a staff of only 3,800!
As of around 2017:
- The 'Big Four' technology giants - Amazon, Facebook, Alphabet (i.e. Google & YouTube) and Netflix - were valued on Wall Street at around $1.5 trillion in May 2017.
- Facebook has over 2 billion users, approaching one-third of the World's population.
- Google has 87% of the search market in the USA, and 83% in the UK.
- Amazon has 43% of the online retail market in the USA and 16% in the UK.
William Perrin kindly drew my attention to Christopher Sirrs' interesting 2015 commentary on the 1972 Robens Report on health and safety which described a fractured and ineffective regulatory landscape very much like that today on social media.