Advancements in cyber physical systems and artificial intelligence (AI) are reshaping our society, promising a world where machines can do almost any task that a human can do and, more often then not, do it better.
This creates a problem, however. If AI enabled machines take all the jobs then unemployment will rise to unprecedented levels, leaving huge parts of the population with no means of income. Production costs may fall but who will have money to buy the products that AI machines produce, and what will this new order mean for our society?
An open discussion on the topic is required; these important questions must be answered, and failure to do so could have dire consequences for our civilisation… This week Memoori speaks with Thomas Kochan, the George Maverick Bunker Professor of Management, a Professor of Work and Employment Research, and the CoDirector of the MIT Sloan Institute for Work and Employment Research. Prof. Kochan is a prominent voice in the discussion on technology advancements and their impact on employment.
The discussion on the impact of AI on jobs tends to go one of two ways; Either, AI will replace the majority of workers leaving large part of society in disarray and with no means of income | Or, AI will mean better jobs for humans, in some sort of star trek style utopia of abundance, where almost everyone works in R&D. Where do you lie?
Firstly, both are true. AI will eliminate some jobs, there’s no question, but we’re also going to create some new ones. So the first question is about balance; are we going to create more jobs than we eliminate or not? And this is still unknown.
The second and much more fundamental question is; how are we going to get on top of the issue so we can use AI to address the big problems in society and create new jobs or career opportunities for the future? That’s the big question, because if we’re not going to be proactive the worst-case scenario will likely dominate.
However, if we think about it strategically and if we engage more people in the design of AI systems, then I think we’ll be able to solve problems and create more jobs to better make this adjustment. It requires a very proactive strategy and it requires that decisions aren’t made by a very narrow band of technology designers, business leaders and would-be-entrepreneurs.
As long as shareholder value is the primary goal in the business world, won’t AI just keep replacing jobs until there are very few left?
Only if we let it. There’s no iron law of economics or market forces that says shareholder value has to be the primary focus of economic activity. That developed in the last three decades because society let shareholder value dominate. We can change that, we can be proactive and specific about saying we want the investments in AI to be widely distributed.
Business schools adopted this shareholder value model, hook, line and sinker, since the 1980s. It’s being challenged however, by impact investors and even by our students right now, who are thinking about a more inclusive economy for the future. So we’ve got to be part of a movement that changes the way in which we approach technology decisions. Make sure that technology and social systems are better integrated as we go forward.
The likes of Stephen Hawking, Elon Musk and Ginni Rometty have been highlighting the dangers of AI to society for years. In contrast, the current US administration says it is not even on their radar. What does this mean for the progress of this discussion?
As I said publically, after the treasury secretary made that silly statement; the current US administration have their head in the sand if they think these issues aren’t going to get serious for another 50-100 years. That’s clearly just hogwash.
I don’t believe we can avoid dealing with these issues, nor will we avoid dealing with them. The American public and people all over the world have shown what the consequences of not being proactive. Global trade has been both politically and economically affected and we can’t make the same mistake about the future developments of technology. They’re here today, they’re happening, we need to address them and we can address them, if we’re smart about it.
It’ll require political leadership, but not the kind of leadership we’re getting out of Washington or London today. It’s going to require people standing up and saying, “this is our economy, it’s our kids future, and we want to make sure that we create a future where we all share the benefits of these wonderful opportunities.”
Once the discussion is on the agenda; what shape do you see the strategy taking? What options do we have to balance the adoption of AI against the protection of society?
Well, we have to start by rebuilding the voice of workers. I don’t just mean bring back trade unions in their mirror image. We’ve got to invent new ways in which workers have a voice and consumers have a voice in how we use these new technologies.
Suppose we think about ride service businesses, Uber for example. There’s a reaction to Uber’s narrow use of AI, its big data system and GPS technologies; they are ultimately the internet of things (IoT) in ride services. They brought all of these technologies together but they’ve done so only for the benefit of owners and investors who are driving up the valuation of Uber in capital markets.
Yes they serve consumers, they have improved the opportunity for people to access more efficient transportation, but they’ve done so on the back of the drivers. You see more and more of a revolt by drivers and cities around the world that say, “yes you disrupted the industry but you’re doing so at the expense of some key stakeholders, we’re going to rebalance that.”
So we can rethink our ride services and delivery in ways that benefit the consumer and provide a return on investment, but also provide better jobs and career opportunities. We have seen new ride service companies that follow this line of thinking and we can put more pressure on companies to follow suit.
JustCapital is an organisation in the US that uses a variety of metrics to rate a company based on its contribution to society; we have to do more of that, make it more public and hold these companies accountable. That’s a small development but I think there’s a growing thirst, in societies around the world, for that kind of corporation.
Developing your Uber example; when AI technology gets to the stage of driverless cars and there are no human drivers to protect, what will that mean for this debate?
That’s a very good example, which illustrates a basic point we need to address. Driverless cars is not the goal, it is a tool. We need to think about what transportation systems and infrastructure we need to make driving safer, more efficient and more accessible to larger numbers of people in our society. We need to think about the use of AI and associated technologies for transportation, and then we’ll shift away from this silly race to try to produce “driverless cars”.
We’ll ask, how can we introduce these technologies incrementally and in the long run to improve our transportation system, then we’ll get a whole range of options. We’ll think about how to integrate the vehicles with the sensors and the community investments that are needed, in our roads and structures, to support it. We’ll think about how we can extend transportation out into areas that are underserved today or better serve parts of our population like the elderly or children.
We need to reframe that debate, not just look for driverless cars as the goal but as a tool in a larger picture. The original automobile only became a part of our economy when we built our highway system. It only became part of a major transformation as it expanded into tractors and efficiency of our farming communities. It only became a driver of economic progress as we thought about how an internal combustion engine could assist in everything from the generation of electricity, to replacing outmoded technology and as electronics developed to support it.
This is the kind of way that we should think about it. Not just “driverless cars,” but how we can use this wonderful technology to improve transportation, create a return for private investors, and for the public who has to make the big public investment that make driverless vehicles viable.
Recently we have seen a backlash against AI being used for employee monitoring. Do you think this example suggests the co-existence of AI and human workers may cause a resistance to the technology?
I think you’re on to a very important point there. Not trying to single out Uber, but a recent story has emerged of the company using behavioural economics and decision theory advances to urge drivers to stay out there longer when they need more drivers. They are using information to not only monitor drivers as they go offline but also to control or even manipulate them, against the driver’s economic or social interests.
There is tremendous opportunity for abuse of information in monitoring our activities, control our thinking and to know more about us than we’d like others to know. The only way to deal with this is for people to say, “I want control of my information and I want to have a voice.” If we’re going to monitor people on the job then employees must have a voice on how their data is used.
That’s why I’m such a strong advocate of reimagining, rekindling and re-empowering worker voice in our society and in our workplaces. These are the modern issues; the full spectrum of workers need to have a voice in how information and big data are being used. I think we need to open up corporations and institutions to this kind of thinking. If we gave more people access to information you would see all kinds of new applications of big data.
What will happen if we don’t have the discussion on AI?
If we don’t have this discussion or create more action in this direction we’re going to have a revolt. I don’t mean a revolution in the traditional sense but we are going to have more unrest. We’re going to have more of a division between the winners and the losers in the race with technology. We’re going to take what we experienced with Brexit or Trump and multiply it.
That’s the alternative future, and yes some people will be winners, gigantic winners; meaning greater inequality, more angry and disenchanted people, and more members of the next generation left behind. It will not be a picture that will strengthen our societies or make us proud citizens. I worry enormously about the consequences of that trajectory.
We don’t have to go down that road, but if we are passive and let practices prevail the way they are now, that trajectory may well dominate our future. I think we need to do everything we can to avoid it.
Prof. Kochan offers a free online course entitled ‘Shaping the Future of Work,’ which explores and develops plans of action for improving the job and career opportunities for today and tomorrow’s workforce. You can enrol or share from here.
[contact-form-7 id="3204" title="memoori-newsletter"]