Progress? It's All About Access to Data
MAKING THE HUGE AMOUNT OF INFORMATION HELD BY THE BIG PLATFORMS MORE ACCESSIBLE, ALSO RETHINKING THE RULES ON GDPR: THIS IS THE ONLY WAY TO TRULY ENABLE INNOVATION BY OTHER COMPANIES AND BREAK THE MONOPOLY OF BIG TECH. VIASARFATTI25 DISCUSSES IT WITH VIKTOR MAYER SCHOENBERGER, AUTHOR OF ACCESS RULESby Jennifer Clark
With law degrees from the University of Salzburg and Harvard Law School, plus deep hands-on knowledge as a computer program developer, Viktor Mayer-Schoenberger is uniquely placed to shed light on the relationship between information and power in our networked age. His 2010 book “Delete: The Virtue of Forgetting in the Digital Age” (Princeton, 2009) won two awards and influenced EU thinking about privacy policy. He followed it up by co-authoring “Big Data: A Revolution That Will Transform How We Live, Work, and Think” (HMH, 2013), which was shortlisted by the Financial Times as one the year’s best business books. He is currently Professor of Internet Governance and Regulation at the Oxford Internet Institute, University of Oxford. viaSarfatti25 talked to him about his book “Access Rules. Freeing Information to Stop Big Tech, Revive Information, and Empower Society,” published in Italian by EGEA in 2021 as Fuori i dati! Rompere i monopoli sulle informazioni per rilanciare il progresso.
What was your goal in writing “Access Rules”?
Starting In the 1970s, data protection laws have focused on not only empowering the individual but also about rebalancing the power of information access between the haves and the have-nots. And we have forgotten about that second part. If we look at Google, Amazon, Facebook and Apple, or GAFA, and say “we need to destroy them,” we’ve got it backwards. Humans have always learned from information. The access to data needs to be more widely spread, and not in the hands of just a few monopolistic companies in Silicon Valley. The insights need to be available to the economy and society at large. That was the starting point.
What is problematic about a data monopoly?
In an economic context, today data more and more is crucial for innovation, because companies can learn from data, either through statistical analysis on people’s preferences, or through artificial intelligence and machine learning. This is a huge problem for Europe in particular, because its economy is full of small- and medium-sized enterprises. Lots of smart people with great ideas. But in order to take the idea and make it in a real product, they often need data, and they don’t have it. So they will never be able to compete with Amazon or Google, even if they have a better idea for a search engine or an online store.
The G7 recently reached an agreement on more rigorous taxation of Big Tech. What are governments “doing wrong” in terms of curbing the power of what you call data monopolies?
The problem is that governments are still thinking in the 20th century. They think that what is really powerful is money, is capital. And to an extent that’s true. But if you are a startup company in Europe and have all the money in the world, and you go to Google and say “I would like to get your data for self-driving cars” Google can simply tell you to go away. Taking a little bit of money out of Jeff Bezos’s pocket is not going to change much. I am not against taxing large platforms. But that will not undo the information imbalance.
Would it be enough for governments to just treat them like a monopoly, and break them up?
If you slice Google into multiple small companies, and you don’t undo the underlying economic mechanism that innovation is fed by access to data, you are basically cutting off the heads of the hydra. They will grow back again, like AT&T did after it was cut into many “Baby Bells” in the 1980s.
How do we “fix” Big Tech to make it fairer?
Very easily. By forcing the large platform to let other companies to have access its data. Not all of it, 5% or 10% would be sufficient for small companies to have enough training data to produce great products. That rekindles competition, it invigorates innovation, and it diminishes the information power concentration of the large monopolies.
You say we need “rules of access.” Can you give us an example of how this would work?
I can give you an example of how it already has worked. About 10 years ago, Google bought a large company that was doing back-office work for airline ticket processing. And Google got all the data. The U.S. Department of Justice stepped in and ordered Google to provide open access to data to anyone who asked for it, even competitors, for a period of five years.
But you also explain that Google shut off access on the very day that the five years was up.
Unfortunately, yes. By the way, we use a similar access mechanism in patents. Patent law is designed for companies to be able to recoup their development costs and make a profit for a limited period of time, after which everybody else can benefit from access to the knowledge because patent applications need to make knowledge transparent. That’s what we need do with data, too. It’s not an expropriation, it’s just making data more liquid, more accessible.
Are you saying that governments already have tools to deal with this?
We have done it before. All we need is to do it again, and not do it on a case-by-case basis. We need a general rule that says if you are above a certain threshold of size as a company and have lots of data, you need to open up that access to that data to anybody else who wants it.
Why do you say that European General Data Protection Regulation (GDPR) backfires? Is the focus on “data privacy” misguided in your view?
From the very beginning, as I mentioned, data protection laws had two elements. One was to empower the individual. And the other was to undo information concentrations. We have forgotten about the second, unfortunately, and we have concentrated on the first. This gives us the appearance of choice. But if we want to shop with Amazon, we can mostly just accept their onerous rules. That is not the informational self-determination that GDPR intended. This is just going through the formal hoops of extra clicks. We have convinced ourselves in Europe that we have strong data protection, but in reality, we don’t.
Why has the question of the power of who controls the access to information not been addressed in Europe?
Because we forgot about it. Over time, you know, we focused more and more on empowering the individual. But the individual doesn’t want to spend three hours a day exercising their rights on the internet. I often compare it to going shopping in the supermarket. You don’t bring a chemistry lab to test what is on sale there. You assume that somebody tested it, and there is an agency that assures what is sold on the shelves is safe. In data protection, we have put all the responsibility on the shoulders of the individuals.
Could you give us an example of the advantages of open data?
One of the key examples about open data is GPS. It was originally designed by the American military with satellites to send signals on earth to find out locations. It was only for the military. Then in the 1980s a Korean commercial flight was shot down by the Russians accidentally, because it veered off course. U.S. President Ronald Reagan decided to make GPS available to all. Now we have GPS in our cars, in airplanes, in delivery vans, our smartphones. This service rests entirely on data that is open and accessible.
What has Covid taught us about the usefulness of data and access to data?
At the very beginning of the pandemic about a year and a half ago, we didn’t know anything about the virus. What happened was quite remarkable. Researchers at National Institutes of Health in the United States were able to access the virus’ genetic footprint that was shared by Chinese and Australian researchers online in January 2020. Using this data the NIH researchers were able to develop the key element of the vaccine extremely quickly. After that it simply became an issue of finding the right delivery platform and enough companies to produce it. The real “eureka” moment of finding the vaccine was done in a very short time because the genetic footprint was shared online.
Has your view evolved since writing “Big Data: The Revolution That Will Transform How We Live, Work and Think”?
The fundamental idea in “Big Data” was that access to data and the analysis of data provides us with insights, so we can make better decisions. The problem that we have today is that ten years after Big Data started becoming available, fewer and fewer companies can make sense of the data. Most of the data collected in Europe, 85%, is never even used once. Google and Amazon would never do that. The problem that we have today is not that we have too little data. We have data concentrated in too few hands. Only a very small number of organizations and companies can create value out of data and don’t let others have access to it. That is extremely dangerous because it destroys competitiveness and innovation in our economy, and because it undoes the democratic and decentralized fabric of our democracy.
What are your thoughts on this undoing of the democratic fabric?
The fundamental problem with Facebook, for example, is a structural one. If something goes wrong with an algorithm, then the data Facebook has can become weaponized against democracy. Why is this problematic? Because the people at Facebook have all the data and nobody can control them or compete with them. Democracy lives because there is open robust debate. But that requires that information be accessible to all of us, so we can build our own opinion. If we don’t undo this concentration of power, we unfortunately cement this two-tier system where some know everything and others know nothing.
