Internet of Things: Regulation

Debate between Jon Cruddas and Chi Onwurah

Thursday 3rd October 2019 Westminster Hall

[Jon Cruddas (Dagenham and Rainham) (Lab)]

I, too, look forward to hearing the Minister’s response to all those questions in a few minutes’ time. I congratulate my hon. Friend the Member for Newcastle upon Tyne Central (Chi Onwurah) on securing the debate, which covers some of the most challenging issues that society— indeed, humanity—will face over the coming years, many of which are rarely discussed in Parliament. Her speech was quite brilliant.

The internet of things is such a vast subject that it is difficult to know where to start, but I will restrict myself to the ethical questions that underlie the regulation issues that my hon. Friend spoke about, given the epochal technological challenges. In a general sense, many challenges that the country faces appear inversely related to our capacity as politicians to properly discuss them, let alone resolve them. Increasingly, liberal democracies appear unable to navigate the complexities of the modern world. One obvious example is the escalating authoritarianism across Europe and the globe—where is the political diagnosis and response to it, and where is the defence of liberal democracy? To give another example, do we really talk, post referendum, about the issues and feelings that ushered in the referendum, or are we preoccupied instead with the technical aspects of Brexit?

Maybe politics has lost its ethical grip and become too technocratic, and maybe today’s populism is a backlash against that managerialism. Maybe we require a different conversation that addresses moral and ethical questions about the lives that people wish to live. I realise that that point appears unrelated to questions of robotics, the internet of things and artificial intelligence, but I would argue that it is imperative to embed our discussion of those technological changes in a deeper conversation. I welcome this debate because maybe we can start that conversation—arguably the most profound conversation that confronts us as politicians and public policy makers in this country and across the planet.

Whether the forecasts are apocalyptic or utopian, no one doubts the significance of artificial intelligence and the internet of things. They have the potential to affect all aspects of policy, from education to the labour market, and from policing to health and social care. However, much of the current political thinking about artificial intelligence is reactive and geared simply towards ensuring that Britain is at the forefront of technological change—we might describe that as the utilitarian approach. Maybe we should begin instead by discussing what role technology should and should not play in our societies, our workplaces and our personal lives. That departure point would be different from the one that tends to dominate the utilitarian approach: instead of focusing simply on utility or economic benefit to Britain plc, it would focus on justice and how society should be organised.

Shrinking the political debate down to technical rather than ethical terms is especially dangerous in this area of technological change, owing to our lack of expertise in it—notwithstanding some notable exceptions, some of whom have just spoken. For example, being unable to evaluate the claims of developers or independently discern the likely outcomes and risks of their products means that politicians and the public are prone to being swayed by either apocalyptic or utopian technological narratives. Many technologists have bought into what has been termed techno-solutionism: the idea that all problems that humanity faces can be solved using technology—even those that technology has caused.

[Chi Onwurah]

I thank my hon. Friend for his excellent remarks, which cover the ethical debate about technology that we too rarely have about the internet of things. One example of the approach he describes—the idea that technology can solve all our problems—is the proposals for alternative arrangements on the island of Ireland, which I understand are being driven by blockchain and other technologies that the Government are not fully familiar with. That libertarian idea that technology is the answer to everything has driven our regulatory approach for too long, so he is right to say that we need experts on technology who can stand up for and consider its future applications from the point of view of society and citizens.

[Jon Cruddas]

That is bang on. For many in silicon valley, that confidence in the potential of technology goes hand in hand with a widespread libertarianism: as the role of technology and profit margin expands, so the role of the state should contract.

My hon. Friend did not mention those who come at the issues from a transhumanist approach. Modern transhumanism asserts that technological change creates the opportunity to transcend the human condition and become transhuman, and that that is to be celebrated, while resistance is deemed nostalgic or parochial. Politicians now and in the future will have to defend a discernible human condition in these debates, which will be a huge challenge.

For example, what happens when transhumanist thinking informs the technologists? Nick Bostrom is the director both of Humanity+, an international transhumanist organisation, and the Future of Humanity Institute at Oxford University, which regularly produces policy recommendations for Government. The point is that politicians and policy makers need to avoid being captivated by the promise of technological progress without an appreciation of the philosophical assumptions that inform the thinking behind the policies being advocated by those with agendas. Consequently, philosophers such as Jürgen Habermas have argued that politicians and policy makers should maintain a “species ethic” when navigating this terrain. These are deep waters, yet such questions are not really addressed in modern political debate.

On a slightly more practical level, the potential risks of mismanaging artificial intelligence are phenomenal. The most obvious example is mass unemployment. It is not possible to pick up a newspaper without reading about the march of the robots and the end of work. Estimates of the proportion of jobs in the UK that could, over the next two decades, be replaced by artificial intelligence and related technologies range from some 22% to between 40% and 45%. There are a wide range of estimates—some of them quite dodgy—of future structural unemployment, and they point to a range of conflicting policy options, such as universal basic income versus full employment. That suggests a wider range of policy remedies, but we are not spending enough time scrutinising the assumptions and empirical data that underscore those policy debates. Maybe we should.

To give a further example, we have already seen data analytics being used malignly in targeted political campaigns, and that practice will become ever more sophisticated, at the expense of our democratic process. As has been mentioned, in the corporate world facial recognition software is now being trialled for the purpose of marketing, to detect the efficacy of an advert on the viewer by judging their facial expressions. Businesses now have the potential to reach into people’s lives in the way Orwell’s “1984” imagined for totalitarian regimes.

Similarly, we have seen the social media filter bubble effect on civic and social life. It feeds us information that aligns with our preconceived notions of the world, closing us off from any contradictory information. Perhaps in the future our children will ask why we as parents allowed them to be so unprotected against such technological power. Left unchallenged, future public debate will suffer from the ease with which fake news could be produced on an industrial scale, given that AI makes the processing and manipulating of all forms of digital data substantially easier and cheaper.

Our very knowledge of the world around us and notions of truth are at stake. That may seem melodramatic, but I do not think it is. The greatest threat to the established political parties, however, could come from the powerlessness and exclusion felt by many as they feel that decisions about them—from hiring, to policing, to insurance—are made by machines. In its evidence to the Lords inquiry into AI, Future Intelligence said that

“the most challenging point relating to AI and democracy is the lack of choice that is offered to the population at large about the adoption of technology. It is, to say the least, undemocratic”.

As wealth becomes increasingly concentrated in the hands of businesses that employ fewer and fewer humans, our society will be riven by inequality on a scale perhaps never before seen. Brexit pales by comparison.

[Chi Onwurah]

My hon. Friend is making excellent points. Although my remarks on Brexit and technology were limited, I want to emphasise his point. If we agree that part of the Brexit vote was based on people’s sense of disconnect from Brussels and the corridors of power, how much greater will that sense of disconnect be when all decisions are made through technology that monitors but is not under the control of the people?

[Jon Cruddas]

Exactly. These are essential issues for the democratic character of western market democracies. That takes us back to the question my hon. Friend asked the Minister about the Government’s proposed remedies and policies. As it stands, policy proposals to meet these challenges are phenomenally weak. For instance, they include developers undergoing training in ethics as part of their computer science degrees, companies ensuring that their workplaces are diverse, and individuals who are made redundant by AI, perhaps repeatedly, being able to train for a new career. As I mentioned earlier, universal basic income is one proposal floated to ensure that those who lose their jobs are not made destitute, but that would mean the state taking on a phenomenal welfare burden just at the time when fewer people were able to pay income tax. To make up the deficit, people such as Bill Gates have suggested a robot tax, but would we tax algorithms as well as robots? Trying to define a robot is a legal and regulatory nightmare.

Returning to the question of regulation, before we make good policy, perhaps we need to return to first principles, asking questions about the values we place on work, freedom, privacy, community and justice—in short, what we want our society to look like. From there, we can then discern the role that we wish to allocate to technology, rather than being seduced by the hype of novelty and processing power. We decide the ethical environment and responsibilities of technologists and their platforms, not vice versa. If we do not build policy on a well-defined vision of human flourishing, policy makers run the risk of slipping into techno- solutionism, thereby putting technological and economic progress above people, leaving them to become citizens of those corporations.

Alternatively, we could endorse a somewhat softer technological determinism and use policy only to manage what we euphemistically call “risk”, when what is really at stake is huge social issues: rising inequality, the accumulation of power in the hands of private companies and human dignity itself. Deeper political conversations are required about what constitutes a good life and a good society. That should inform our approach to regulation. We literally need to rethink human rights in a different way, in terms of the preservation of the species. Thanks to my hon. Friend the Member for Newcastle upon Tyne Central, we can start that conversation.