Time to Break Vicious Circle of Technology Backlashes

Over the past two weeks, the criticism of Facebook has reached a boiling point, having simmered since the Cambridge Analytica data scandal was uncovered in 2018. Several global brands, including Unilever, Coca Cola, and Starbucks, have announced they are boycotting Facebook, accusing Mark Zuckerberg of turning a blind eye to hate speech. 

Rival platforms Redditt, Youtube, and Twitter fare little better in public perception but have at least removed some content considered hateful, often wielding blunt instruments. Reddit, for example, banned a feminist sub reddit but left pages that promote misogyny and pornography.

Kevin Roose, the tech columnist for the New York Times, thinks we have reached the inevitable end of the romantic, thrilling digital realm that he called the Wild Wild Web. 

This is not the first such backlash, and unless we make changes at the governance level, it won’t be the last. The problems at Facebook and Youtube are not service or product level, these are at the governance level. 

We need to shift the discussion from internet freedom toward accountability, which Roose points out should be in the context of technology development and governance. This may be the only way to avoid these continual backlashes in the last 50 years. Parallels can be drawn with the automotive industry.  

 

Car Trouble

Lee Siegel, the writer, and cultural critic discussed how people came to realize the downside of the convenience of automobiles in his book “Against the Machine”: 

“By the early 1960s, fifty thousand people were dying in car accidents every year… Yet the public did not complain. The rhetoric surrounding the automobile had made it impervious to skepticism. Cars were not just a marvel of convenience, people were told, they were a miracle of social and personal transformation. 

Advertising identified the car’s power and mobility with the promise of American life itself… The “make” of your car was the very definition of your social relevance as its owner. 

Such an illusion gave auto manufacturers pretext for their neglect. They could hide their cost-conscious refusal to make cars safer behind their claim that nothing could be done-the trade-off in human life was inevitable and inexorable. That was the nature of cars…The increasingly affordable automobile seemed like the ultimate proof and fulfillment of democracy. If people were dying on the road in greater numbers, it was because greater numbers of people were enjoying the freedom, choice, and access provided by the new machines roaring along the open road. Criticize the car and you were criticizing democracy. Anyway, that’s just the way things were.

Until 1965. That was the year Ralph Nader publishes Unsafe at Any Speed, his classic expose of the automobile industry’s criminal neglect. The public was horrified. It seemed that auto industry executives had known all along what the problems were. Engineers had pressed them to make changes that would have saved tens of thousands, but the bosses silenced the criticism for the sake of cutting costs and protecting shareholders and their own jobs.

Not only the public was horrified; it was shocking. What it had accepted as an inevitable condition turned out to be wholly arbitrary. Things should have been very different from the way they were. And gradually, by means of public pressure, the “permanent” condition of the necessarily dangerous car did yield to the new condition of the safety-conscious auto industry. People stopped dying on the road in staggering numbers. Things changed.”

What the automotive industry went through is not that different from what we are going through now. In the 1960s, despite annual profits running into the billions, less than 0.1 percent of that money was allocated to safety research. What percentage has today’s internet giants allocated to user security? Could we even know?

As we can see, the path of automotive technology was shaped not by car companies but by the people, exactly the same as the case of Facebook.

 

What pushes companies to this point?

All these companies are private entities. In a growth economy model  – where there is also degrowth – shareholder pressure on tech executives demands revenue increases no matter what. Since there are no policies to measure the tech companies’ negative impact on society – unless there is serious criticism coming from the public – the argument that they bring convenience can be a pretext for their negligence toward their social impact. 

“In the early years of new technologies, people embraced technology too readily, seeking only the benefits, and ignoring the many downfalls,” said Siegel.

It has taken years for people to understand the price of internet convenience.

The question now is whether we are going to repeat the same mistakes with emerging technologies such as AI or blockchain, which are already embraced dearly.

We certainly need a new and different lens through which to look at the increasingly complex problems of the 21st century, as well as our technology development model.

 

What could be different?

Currently, I am part of a technology project funded by the European Commission. These projects have a unique governance model, in our case involving 20 organizations from 9 countries, which all have to follow the EC’s framework. Even the project’s scope is outlined by the EC. 

The commission enforces stringent regulations, from ethics to data protection and working under these regulations is certainly slower and in some cases not as efficient as working methods in the private sector, but efficiency is fairly an easy problem to solve. The most important part is to have these frameworks. 

In the past 20 years, I have tried to center technology development on humans and the ecosystem, depending on the context. I can easily say that regulations enforced by the “funding body” model have worked best.

 

Is regulation the right way?

I cannot get out of my mind a simple question. When there are heavy regulations and procedures controlling the development of drugs, which can take more than a decade and still result in no approval, how are we able to develop and have an app on the AppStore in a week? An app’s negative impact on society can be more scalable and longer-term than the harm that a drug can have. Just look at Facebook and Youtube.

But as we all know, these regulations work quite well in the short run but in the long run, it creates its own dynamics, brings lobbying companies to the picture, and creates a new set of huge problems as in the pharmaceutical markets. Another negative example could be from the automotive industry: There can still be cheating on diesel emissions by all these credible car brands despite the heavy regulations. 

 

Governance level Innovation

Silicon Valley companies of the Wild Wild Web mainly focused on service, product, and business level innovation, and enjoyed technology’s scalability. This brought power and economic attractiveness and neglected the governance-level problems, as there is no such requirement. It was inevitable that we have come to where we are now. What were we expecting?

Now it is time for governance-level innovation to fix the Wild Wild Web. However, for the upcoming waves of technology, instead of a”shoot-first, fix-later” approach, we need more fundamental changes. They include how these companies are valued in the market. 

The only long-term governance-level trial that I know of from Silicon Valley is LTSE by Eric Ries, who is trying to set up a stock exchange where companies can focus on their long-term goals an take care of topics like sustainability and inclusivity.
Shoot-first, fix-later makes the public pay the price while companies profit. To change this, we need to focus on the foundations of economic models in the larger context and consider even new governance and economic models, where profitability is not companies’ only focus, and perhaps we can even move away from growth economic models.

 

@gulayozkan, July 2020 Karaagac

Category : Blog, Uncategorized
Tags :