NEXT: A Market Where Data Creation is Paid Labor
I recently completed a six-month program with my insurance company where I implanted a device in my car that measured my driving habits. In exchange, based on the quality of my driving as measured by their metrics, I got a discount on my rate. (But I could have just as easily received no discount.) The data I generated in the act of driving is now the property of my insurance company, the value of that data now of benefit to them. User-generated data of the kind we create when we go about our daily lives is indeed becoming more valuable, which may be catalytic for a new kind of economic system.
Practically any time we engage in an act of (digital) consumption, we emit data: data is a by-product, the exhaust of our consumption. Most of the time we are either unaware that we are so emitting data, or that companies are collecting that data, or that our data has any value. When we thoughtlessly click on “agree” for some Terms of Service contract, we are typically allowing the app to collect and take ownership of our data. Indeed, for some companies collecting user-generated data is their business. In an Internet of Things world, we will be surrounded by more and more devices that will capture even more of our data.
The value of our data is increasing, especially with regards to the development of artificial intelligence. Machine learning algorithms are becoming better and smarter because they are being fed loads of data from which to learn. The more data an algorithm can learn from, the more intelligent it grows, and so this by-product becomes a vital input for improved AI. It is ironic: artificial intelligence is improving because of the actions of consumers, the same people whose jobs are threatened by advancing artificial intelligence.
A group of economists and technologists, including virtual reality pioneer Jaron Lanier, have proposed a radical new market, one where users—the generators of the data that tech firms are using to develop AI—retain the ownership of that data, and thus retain the value of that data for themselves. The group refers to this condition as “data as labor,” as opposed to our current condition where “data is capital.”
As artificial intelligence is poised to take away more and more human jobs, one response is the establishment of a universal basic income. But another response to the social and economic dislocation that AI might unleash is the creation of a labor market for data, where we are compensated for our online activities, compensated for the data we emit as a by-product of those activities. In a sense, in such a world with a data-as-labor market, my “job” becomes data production.
Humans have been generating data for a while now. Statistics—literally “data about the state”—was developed in the eighteenth century, and was the information governments collected about the activities within the borders of the state. Inhabitants would generate data by virtue of their being a citizen. Voting yields data as do polling and surveys, data that has value and is used for a variety of purposes. In many of these cases, users have been generating data that is then given away for free.
Whenever I swipe my grocery store loyalty card at checkout, I am letting Kroger know what I’ve purchased, giving that data away to them for nothing, except a few coupons and gas points. The collective actions of Kroger shoppers, however, yield data which has value to both Kroger and to marketers. In the past, the value of the data we generate via voting or shopping has been marginal, such that today many users freely give up their data in exchange for “free” services such as Facebook and Google. But with AI, especially, the value of data is rising quickly. Are you willing to give away such a valuable commodity so freely?
How might such a data-as-labor market be created? Lanier et. al. point to three potential sources: first, companies might begin to pay for data as they would for any other commodity. There is a start-up called Meeco that has created such a service that manages personal data, which would help facilitate any such exchange.
The Meeco Manifesto reads:
Up until now the power to capture, analyse and profit from personal data has resided with business, government and social networks. What if you and I had the same power? Imagine the power if that capability could be transferred to you, to help you capture a life view of your habits, spending, preferences and your journey through life.
With the technology to so capture and retain personal data in place, governments might also enact regulations that legalize the ability of users to retain and profit from the value of their data. The European General Data Protection Regulations go into effect in May of 2018, and seek to ensure that ownership of data resides with the users who generate it. Given the regulatory roll-backs from the Trump administration and the Republican-controlled Congress, it seems unlikely at least in the short run that they would agree to new regulations that would be of benefit to users.
But, that is not to say that future administrations might not consider enacting laws that guarantee that those who create data are the legal owners of that data. Finally, Lanier and his colleagues imagine a revival of labor unions, but a labor union of data-generators, a way for users to wrest control of their data from large companies, a mechanism for collective bargaining around data.
From the exhaust of consumption, a new economic system is emerging.
David Staley is interim director of the Humanities Institute and a professor at The Ohio State University. He is president of Columbus Futurists and host of CreativeMornings Columbus.
The next Columbus Futurists monthly forum will be Thursday, February 22 at 6:30 p.m. at the Panera Bread community room (875 Bethel Rd.) Our topic for the evening will be “The Future Without Net Neutrality.”
The next CreativeMornings Columbus will be Friday, February 16 at 8:30 a.m. at the Amelita Mirolo Barn. Cindy Meyers Foley will speak on the theme “Curiosity.”