The future of data, AI, work and democracy: an interview

This is an interview for the Cypriot newspaper Fileleftheros. The Greek translation can be found in this link. What follows is the original text in English. The interviewer was Xenia Trouki.

Let’s start by understanding what a futurist does, including how you go about forecasting.

A futurist aims to predict how current trends may result in significant outcomes in the near, medium and long term. I use a number of approaches, including Foresight methodology, and essentially try to identify and understand how technological, economic, social, and geopolitical change drivers interact, and what this interaction may evolve and under what circumstances.

What is your starting point to think about how will things change going forward? Can you use the past to help predict what’s coming?

I start by looking into current change drivers and tensions. For example, an important change driver is the accelerated virtualization of the workplace that we have witnessed because of the pandemic. An example of a geopolitical tension is how fragmented the internet is becoming because of new antagonisms between the West and China. Taking into account past events is very important, as well as understanding the context of geopolitics and technology in terms of history and culture. Of course, as with every prediction about the future, one cannot possibly forecast so called “black swan” events. For example, the pandemic was not unpredictable but it was a “black swan event” in the sense that it happened when it did.

If you can think in terms of 15-20 years from now, how do you think an average person’s life will change?

We live in an era of exponential change, this means that 5 years from now we will be able to do more than twice of the things that we were able to do 5 years ago, and so on. To take an example, the confluence of quantum computing, artificial intelligence and life sciences could provide us with new therapies for curing major chronic and rare diseases by the end of the decade; and quite possibly the ability to design new materials that can do amazing things, such as convert light into energy just like the leaves of plants, and thus revolutionize how we power our homes and cars. There will be major advances in space exploration, and space travel will become more accessible for more people. The average person in 2040 will have many more opportunities for a better and healthier life. There will, however, be significant challenges in how nation-states with democratic governments survive the crisis of national identity and loyalty. Many people in mid-century may be more engaged and loyal to communities, and maybe “countries”, that are wholly digital and have no physical manifestation.

There’s no doubt that the coronavirus pandemic has changed our daily life. Which changes will last? How our lives will be when it’s all over?

There have been two major changes that will have long repercussions in the years to come. The first was the breakdown between public and private space. Working from home and doing Zoom calls all day meant that we invited everyone in our living rooms and bedrooms. This experience will impact how we understand privacy in the digital economy. We will be more “open” to our privacy becoming compromised, which will enforce the debate on whether we should hide ourselves in cyberspace or we should develop new forms of governance where we the citizens are directly in control of our data and digital avatars, rather than private companies and governments as is the case today.

The second change has been the virtualization of the workplace. Having most workers working remote has many advantages in terms of life-work balance. But it also introduces the opportunity for new ways to labor arbitrage. Many companies will shift towards a hybrid way of working by outsourcing much of the work to workers living in low-cost areas and pay them less salary.

Is this crisis an opportunity for change at the organizational level? And if yes, do you think that will take it and use it?

Clearly the virtualization of the workplace has accelerated the digital transformation of companies. In combination with intelligent automation this trend will drive much efficiency into business processes and help most companies recover quickly from the economic damage of the pandemic. I do however worry about human productivity. Companies may not invest enough in technologies that augment productivity and focus solely into reducing operational cost. If that happens, then we will see a significant negative impact on jobs and worker income. I am however hopeful, that in the medium to long term labor markets will recover as companies realize that in order to remain competitive they need to also be innovative, not just cost-efficient.

Did coronavirus place any challenges for democracies? And if yes, which are they?

Democracies faced terrible dilemmas during the pandemic. For example, how far should a democratic government reduce civil liberties in the name of public health? And whose welfare should the government prioritize? The vulnerable elderly who are not currently productive but have worked all their lives, or the young who are productive and need to work and learn in order to ensure a viable and prosperous future for themselves and society? And, finally, with so many internal social conflicts, how can we ensure broad social consensus? I think most democratic governments have failed in getting the balance right, and this will have an adverse impact when the dust settles and we slowly return to normal life. To give you an example, authoritarian China is arguing that their political system was better at dealing with the pandemic. They certainly have a point, if authoritarianism is the only approach to resolving a public heath crisis. But if they are right what is the relative advantage of democracy? To the latter question I am afraid no democratic government has a good answer to give after the way they handled the pandemic.

You are arguing that if we want to preserve our democratic freedoms and liberties, we must reinvent democracy by enabling more direct participation of citizens in policy making. How can we achieve this?

The pandemic has demonstrated how difficult is for democratic governments to resolve social dilemmas without becoming authoritarian. If we want to preserve civil liberties and democracy we must introduce much more direct participation of citizens in decision-making. Citizen Assemblies are a good and tested approach that democracies can adopt. These are groups of citizens elected by sortition, who are then given the task to deliberate on a specific issue and advise a government or a parliament on legislation. Citizen Assemblies are increasingly used by many nations to depolarize society. For example, Ireland used one to debate abortion rights ahead of its 2018 referendum that amended its constitution. In a world where citizens are losing trust in their system of government it will be only through direct citizen engagement and Citizen Assemblies that we can rejuvenate our democracies, as I argue in my book.

Is AI a threat for the democracy?

AI is by its very nature authoritarian. This is because at the core of AI systems is the idea of autonomy, in other words that AI systems should be able to take decisions independently and on their own. AI autonomy is causing a multitude of ethical problems, especially when they embed in government systems, for instance when they decide how welfare should be distributed, are used by the police or the justice system etc. AI algorithms are also responsible for much or the rage and polarization that takes place in cyberspace and spills over in the physical world as well. This is because online companies that use data to understand people have pioneered Ai systems that push advertising content. These AI content “recommendation engines” have had the adverse effect of personalizing content on social media so much, that people constantly confirm their biases and only rarely are exposed to opposing views or engage in cordial debate. The result is rage, which – as Aristotle has argued – is what ultimately brings all democracies down.

Some AI applications will augment human productivity and will thus create demand for new services and therefore jobs. AI applications that simply automate existing tasks will displace workers but will deliver economic efficiency.

When people approach you to ask about whether their job will get replaced by either AI or robotics, what do you reply to them?

Probably. Most economists agree that more than half of today’s jobs will be eliminated in the next ten years. But new jobs will also be created. These two effects – the job displacement and job compensation effect – are always present during massive technological disruption, and that’s exactly what we are facing due to intelligent automation and work virtualization. But these effects will not impact all countries equally. Those countries that have the best social infrastructures to absorb the impact of automation will be the most resilient and will bounce back first. It will be a big struggle for countries that are not investing on people, skills and retraining.

How can we determine the value of skills relevant to the future-of work-marketplace, and how can we increase the value of human labor in the 21st century?

Industries where efficiency is key, such as logistics, transportation, and manufacturing will be fully automated by mid-century or earlier. Industries that rely on human creativity, empathy and care will flourish, for example healthcare, entertainment, art, hospitality, science, engineering and architecture, design, etc. The value of skills is always relative to demand, which fluctuates and is hard to predict, and one must also take into account labor arbitrage due to work virtualization – this means that some skills may be on high demand but companies may find them in low-cost countries rather than source them locally. So policies for skills should align with industrial policy. Governments, especially in small countries, should focus on key competencies and industries where they have – or should have – clear global advantages, and reskill people on the basis of that.

What new systems of education and training could help workers reap gains from technology and automation?

Artificial Intelligence is a key technology to use in personalizing training and education. We can easily imagine AI systems as personal “training coaches” that understand who we are and what we want to do, predict what skills and capabilities will be in demand that would fit our profile, and on the basis of that help us shape reskilling pathways into the future. I expect that the whole education systems, at all levels, will change dramatically in the next decade. The pandemic showed that the current model is inadequate and can be replaced with a combination of online and on situ education. Teachers and professors will need to be retrained to become mentors and coaches, rather than disciplinarian “machines” that churn out facts that can be more easily acquired online.

How would we prepare our children of today for the future?

Give them self-confidence. Teach them how to collaborate in teams and support each other. Make them love and care for the natural world. Tell them that they live in the best possible world and that, despite its many shortcomings and challenges, it will be in their power to make it a better place.

Do you agree that a significant portion of future’s jobs won’t exist yet or this is just a myth?

New jobs will be created due to the compensation effect, as described above. This is because of human innovation. Innovation is combining things into novel configurations to solve problems in better ways. The digital economy is currently undergoing innovation at an exponential rate. I can imagine new jobs that will be completely taking place in the virtual worlds of video games. Healthcare and life sciences are also coming into a tipping point where the long-awaited promise of personalized medicine will become a reality. In combination with AI and data healthcare will be completely transformed, and new jobs will be created in the space between nurses and doctors, but with many technical and digital skills.