Yves here. This is a wide-ranging, lively, sobering talk about the implications of tech feudalism and what we can do about it.
This is the text version of a talk I gave on May 10, 2017, at the re:publica conference in Berlin.
The good part about naming a talk in 2017 ‘Notes from an Emergency’ is that there are so many directions to take it.
The emergency I want to talk about is the rise of a vigorous ethnic nationalism in Europe and America. This nationalism makes skillful use of online tools, tools that we believed inherently promoted freedom, to advance an authoritarian agenda.
Depending on where you live, the rise of this new right wing might be nothing new. In the United States, our moment of shock came last November, with the election of Donald Trump.
The final outcome of that election was:
65.8 million for Clinton
63.0 million for Trump
This was the second time in sixteen years that the candidate with fewer votes won the American Presidency. There is a bug in the operating system of our democracy, one of the many ways that slavery still casts its shadow over American politics.
But however tenuously elected, Trump is in the White House, and our crisis has become your crisis. Not just because America is a superpower, or because the forces that brought Trump to power are gaining ground in Europe, but because the Internet is an American Internet.
Facebook is the dominant social network in Europe, with 349 million monthly active users. Google has something like 94% of market share for search in Germany. The servers of Europe are littered with the bodies of dead and dying social media sites. The few holdouts that still exist, like Xing, are being crushed by their American rivals.
In their online life, Europeans have become completely dependent on companies headquartered in the United States.
And so Trump is in charge in America, and America has all your data. This leaves you in a very exposed position. US residents enjoy some measure of legal protection against the American government. Even if you think our intelligence agencies are evil, they’re a lawful evil. They have to follow laws and procedures, and the people in those agencies take them seriously.
But there are no such protections for non-Americans outside the United States. The NSA would have to go to court to spy on me; they can spy on you anytime they feel like it.
This is an astonishing state of affairs. I can’t imagine a world where Europe would let itself become reliant on American cheese, or where Germans could only drink Coors Light.
In the past, Europe has shown that it’s capable of identifying a vital interest and moving to protect it. When American aerospace companies were on the point of driving foreign rivals out of business, European governments formed the Airbus consortium, which now successfully competes with Boeing.
A giant part of the EU budget goes to subsidize farming, not because farming is the best use of resources in a first-world economy, but because farms are important to national security, to the landscape, to national identity, social stability, and a shared sense of who we are.
But when it comes to the Internet, Europe doesn’t put up a fight. It has ceded the ground entirely to American corporations. And now those corporations have to deal with Trump. How hard do you think they’ll work to defend European interests?
The Feudal Internet
The status quo in May 2017 looks like this:
There are five Internet companies—Apple, Google, Microsoft, Amazon and Facebook. Together they have a market capitalization just under 3 trillion dollars.
Bruce Schneier has called this arrangement the feudal Internet. Part of this concentration is due to network effects, but a lot of it is driven by the problem of security. If you want to work online with any measure of convenience and safety, you must choose a feudal lord who is big enough to protect you.
These five companies compete and coexist in complex ways.
Apple and Google have a duopoly in smartphone operating systems. Android has 82% of the handset market, iOS has 18%.
Google and Facebook are on their way to a duopoly in online advertising. Over half of the revenue in that lucrative ($70B+) industry goes to them, and the two companies between them are capturing all of the growth (16% a year).
Apple and Microsoft have a duopoly in desktop operating systems. The balance is something like nine to one in favor of Windows, not counting the three or four people who use Linux on the desktop, all of whom are probably at this conference.
Three companies, Amazon, Microsoft and Google, dominate cloud computing. AWS has 57% adoption, Azure has 34%. Google has 15%.
Outside of China and Russia, Facebook and LinkedIn are the only social networks at scale. LinkedIn has been able to survive by selling itself to Microsoft.
And outside of Russia and China, Google is the world’s search engine.
That is the state of the feudal Internet, leaving aside the court jester, Twitter, who plays an important but ancillary role as a kind of worldwide chat room.
Google in particular has come close to realizing our nightmare scenario from 1998, a vertically integrated Internet controlled by a single monopoly player. Google runs its own physical network, builds phone handsets, develops a laptop and phone operating system, makes the world’s most widely-used browser, runs a private DNS system, PKI certificate authority, has photographed nearly all the public spaces in the world, and stores much of the world’s email.
But because it is run by more sympathetic founders than Bill Gates, because it builds better software than early Microsoft did, and because it built up a lot of social capital during its early “don’t be evil” period, we’ve given it a pass.
It’s not clear that anyone can secure large data collections over time. The asymmetry between offense and defense may be too great. If defense at scale is possible, the only way to do it is by pouring millions of dollars into hiring the best people to defend it. Data breaches at the highest levels have shown us that the threats are real and ongoing. And for every breach we know about, there are many silent ones that we won’t learn about for years.
A successful defense, however, just increases the risk. Pile up enough treasure behind the castle walls and you’ll eventually attract someone who can climb them. The feudal system makes the Internet more brittle, ensuring that when a breach finally comes, it will be disastrous.
Each of the big five companies, with the important exception of Apple, has made aggressive user surveillance central to its business model. This is a dilemma of the feudal internet. We seek protection from these companies because they can offer us security. But their business model is to make us more vulnerable, by getting us to surrender more of the details of our lives to their servers, and to put more faith in the algorithms they train on our observed behavior.
These algorithms work well, and despite attempts to convince us otherwise, it’s clear they work just as well in politics as in commerce. So in our eagerness to find safety online, we’ve given this feudal Internet the power to change our offline world in unanticipated and scary ways.
These big five companies operate on a global scale, and partly because they created the industries they now dominate, they enjoy a very lax regulatory regime. Everywhere outside the United States and EU, they are immune to government oversight, and within the United Statesl the last two administrations have played them with a light touch. The only meaningful attempt to regulate surveillance capitalism has come out of the European Union.
Thanks to their size and reach, the companies have become adept at stonewalling governments and evading attempts at regulation or oversight. In many cases, this evasion is noble. You don’t want Bahrain or Poland to be able to subpoena Facebook and get the names of people organizing a protest rally. In other cases, it’s purely self-serving. Uber has made a sport of evading all authority, foreign and domestic, in order to grow.
Good or bad, the lesson these companies have drawn is the same: they need only be accountable to themselves.
But their software and algorithms affect the lives of billions of people. Decisions about how this software works are not under any kind of democratic control. In the best case, they are being made by idealistic young people in California with imperfect knowledge of life in a faraway place like Germany. In the worst case, they are simply being read out of a black-box algorithm trained on God knows what data.
This is a very colonial mentality! In fact, it’s what we fought our American War of Independence over, a sense of grievance that decisions that affected us were being made by strangers across the ocean.
Today we’re returning the favor to all of Europe.
Facebook, for example, has only one manager in Germany to deal with every publisher in the country. One! The company that is dismantling the news industry in Germany doesn’t even care enough to send a proper team to manage the demolition.
Denmark has gone so far as to appoint an ambassador to the giant tech companies, an unsettling but pragmatic acknowledgement of the power relationship that exists between the countries of Europe and Silicon Valley.
So one question (speaking now as an EU citizen): how did we let this happen? We used to matter! We used to be the ones doing the colonizing! We used to be a contender!
How is it that some dopey kid in Palo Alto gets to decide the political future of the European Union based on what they learned at big data boot camp? Did we lose a war?
The lack of accountability isn’t just troubling from a philosophical perspective. It’s dangerous in a political climate where people are pushing back at the very idea of globalization. There’s no industry more globalized than tech, and no industry more vulnerable to a potential backlash.
China and Russia show us that the Internet need not be a world-wide web, that it can be subverted and appropriated by the state. By creating a political toolkit for authoritarian movements, the American tech giants may be putting their own future at risk.
Given this scary state of the world, with ecological collapse just over the horizon, and a population sharpening its pitchforks, an important question is how this globalized, unaccountable tech industry sees its goals. What does it want? What will all the profits be invested in?
What is the plan?
The honest answer is: rocket ships and immortality.
I wish I was kidding.
The best minds in Silicon Valley are preoccupied with a science fiction future they consider it their manifest destiny to build. Jeff Bezos and Elon Musk are racing each other to Mars. Musk gets most of the press, but Bezos now sells $1B in Amazon stock a year to fund Blue Origin. Investors have put over $8 billion into space companies over the past five years, as part of a push to export our problems here on Earth into the rest of the Solar System.
As happy as I am to see Elon Musk and Jeff Bezos fired into space, this does not seem to be worth the collapse of representative government.
Our cohort of tech founders is feeling the chill breath of mortality as they drift into middle age. And so part of what is driving this push into space is a more general preoccupation with ‘existential risk’.
Musk is persuaded that we’re living in a simulation, and he or a fellow true believer has hired programmers to try to hack it.
Peter Thiel, our most unfortunate German import, has built a survival retreat for himself in New Zealand.
Sam Altman hoards gold in Big Sur.
OpenAI, a religious cult thinly disguised as a research institution, has received $1B in funding to forestall the robot rebellion.
The biggest existential risk, of course, is death, so a lot of money is going to make sure that our big idea men don’t expire before the world has been received the full measure of their genius.
Google Ventures founded the very secretive life extension startup Calico, with $1.5B dollars in funding. Google loses $4B a year on its various “moon shots”, which include life extension. They employ Ray Kurzweil, who believes we’re still on track for immortality by 2045. Larry Ellison has put $370M to anti-aging research, as anybody would want to live in a world with an immortal Larry Ellison. Our plutocrats are eager to make death an opt-out experience.
Now, I’m no fan of death. I don’t like the time commitment, or the permanence. A number of people I love are dead and it has strained our relationship.
But at the same time, I’m not convinced that a civilization that is struggling to cure male-pattern baldness is ready to take on the Grim Reaper. If we’re going to worry about existential risk, I would rather we start by addressing the two existential risks that are indisputably real—nuclear war and global climate change—and working our way up from there.
But real problems are messy. Tech culture prefers to solve harder, more abstract problems that haven’t been sullied by contact with reality. So they worry about how to give Mars an earth-like climate, rather than how to give Earth an earth-like climate. They debate how to make a morally benevolent God-like AI, rather than figuring out how to put ethical guard rails around the more pedestrian AI they are introducing into every area of people’s lives.
The tech industry enjoys tearing down flawed institutions, but refuses to put work into mending them. Their runaway apparatus of surveillance and manipulation earns them a fortune while damaging everything it touches. And all they can think about is the cool toys they’ll get to spend the profits on.
The message that’s not getting through to Silicon Valley is one that your mother taught you when you were two: you don’t get to play with the new toys until you clean up the mess you made.
The circumstances that have given the tech industry all this power will not last long. There is a limited time in which our small caste of tech nerds will have the power to make decisions that shape the world. By wasting the talents and the energies of our brightest people on fantasy role play, we are ceding the future to a more practical group of successors, some truly scary people who will take our tools and use them to advance a very different agenda.
To recap: the Internet has centralized into a very few hands. We have an extremely lucrative apparatus of social control, and it’s being run by chuckleheads.
The American government is also being run by chuckleheads.
The question everybody worries about is, what happens when these two groups of chuckleheads join forces?
For many Americans, the election was a moment of profound shock. It wasn’t just Trump’s policies that scared us. It was the fact that this unserious, cruel, vacant human being had been handed the power of the American presidency.
Scariest to me was how little changed. No one in the press or in social media had the courage to say “we fucked up.” Pundits who were stunned by the election result still made confident predictions about what would happen next, as if they had any claim to predictive power.
After the election both Facebook and Google looked at the mountains of data they had collected on everyone, looked at the threats the Trump Administration was making—to deport 11 million people, to ban Muslims from entering the country—and said to themselves, “we got this.”
The people who did worry were tech workers. For a moment, we saw some political daylight appear between the hundreds of thousands of people who work in the tech sector, and the small clique of billionaires who run it. While the latter filed in to a famously awkward meeting with Trump and his children at the top of his golden tower, the former began organizing in opposition, including signing a simple but powerful pledge to resign rather than help Trump fulfill one of his key campaign promises: barring Muslims from the United States.
This pledge was a small gesture, but it represented the first collective action by tech workers around a political agenda that went beyond technology policy, and the first time I had ever seen tech workers come out in open defiance of management.
A forest of new organizations sprung up. I started one, too, called Tech Solidarity, and started traveling around the country and holding meetings with tech workers in big cities. I had no idea what I was doing, other than trying to use a small window of time to organize and mobilize our sleepy industry.
That feeling of momentum continued through when Trump took office. The Women’s Marchin January brought five million people out onto the streets. America is not used to mass protests. To see the streets of our major cities fill with families, immigrants, in many cases moms and daughters and grandmothers marching together, that was a sight to take your breath away.
Hard on the heels of it came the travel ban, an executive order astonishing not just in its cruelty—families were split at airports; in one case a mom was not allowed to breastfeed her baby—but in its ineptitude. For a week or two lawyers were camped out at airports, working frantically, sleeping little, with spontaneous efforts to bring them supplies, get them funding, to do anything to help. We held a rally in San Francisco that raised thirty thousand dollars from a room of a hundred people. Some of the organizations we were helping couldn’t even attend, they were too busy at the airport. It didn’t matter.
The tech companies did all they could to not get involved. Facebook has a special ‘safety check‘ feature for exactly this kind of situation, but never thought of turning it on at airports. Public statements out of Silicon Valley were so insipid as to be comical.
Employees, however, were electrified. It looked like not only visitors but permanent residents would be barred from the United States. Google employees staged a walkoutwith the support of their management; Facebook (not wishing to be left behind) had its own internal protest a couple of days later, but kept it a secret. Every time the employees pushed, management relented. Suddenly top executives were going on the record against the travel ban.
People briefly even got mad at Elon Musk, normally a darling of the tech industry, for his failure to resign from the President’s advisory council. The silent majority of tech employees had begun to mobilize.
And then… nothing happened. This tech workforce, which had gotten a taste of its own power, whose smallest efforts at collective action had produced immediate results, who had seen just how much sway they held, went back to work. The worst of Trump’s travel ban was blocked by the court, and we moved on. With the initial shock of Trump in office gone, we now move from crisis to crisis, but without a plan or a shared positive goal.
The American discomfort with prolonged, open disagreement has set in.
When I started trying to organize people in November, my theory was that tech workers were the only group that had leverage over the tech giants.
My reasoning went like this: being monopolies or near-monopolies, these companies are impervious to public pressure. Boycotts won’t work, since opting out of a site like Google means opting out of much of modern life.
Several of these companies are structured (unusually for American corporations) in such a way that the board can’t control the majority of votes. At Google and Facebook, for example, the ultimate say goes to the founders. And since Google and Facebook are the major online publishing outlets, it’s unlikely that the press would ever criticize them, even if journalists were capable of that kind of sustained attention.
So that leaves just one point of leverage: employees. Tech workers are hard to find, expensive to hire, take a long time to train, and can have their pick of jobs. Tech companies are small compared to other industries, relying heavily on automation. If even a few dozen workers on an ops team acted in concert, they would have the power to shut down a tech giant like Google. All they had to do was organize around a shared agenda.
Workers seemed receptive to the argument, but confused about how they could make collective action a reality. Trade unions in the United States have been under attack for decades. There is almost no union culture in technology. Our tech workers are passive and fatalistic.
So here I am in Europe, wondering, what on Earth can we do?
And I keep coming back to this idea of connecting the tech industry to reality. Bringing its benefits to more people, and bringing the power to make decisions to more people.
Closing the Loop
After Communism collapsed in Poland, I started visiting the country every eight months or so. Even in the darkest period of the 1990’s, it was striking to see people’s material standard of living improve. Suddenly people had cars, phones, appliances. These gains were uneven but broad. Even farmers and retirees, though they were the hardest hit, had access to consumer goods that weren’t available before. You could see the change in homes and in public spaces. It was no longer necessary for office workers in Kraków to change their shirts at lunchtime because of soot in the air. The tap water in Warsaw went from light brown to a pleasant pale yellow.
For all the looting, corruption, and inefficiency of privatization, enough of the new wealth got through that the overall standard of living went up. Living standards in Poland in 2010 had more than doubled from 1990.
In the same time period, in the United States, I’ve seen a whole lot of nothing. Despite fabulous technical progress, practically all of it pioneered in our country, there’s been a singular failure to connect our fabulous prosperity with the average person.
A study just out shows that for the median male worker in the United States, the highest lifetime wages came if you entered the workforce in 1967. That is astonishing. People born in 1942 had better lifetime earnings prospects than people entering the workforce today.
You can see this failure to connect with your own eyes even in a rich place like Silicon Valley. There are homeless encampments across the street from Facebook headquarters. California has a larger GDP than France, and at the same time has the highest poverty rate in America, adjusted for cost of living. Not only did the tech sector fail to build up the communities around it, but it’s left people worse off than before, by pricing them out of the places they grew up.
Walk the length of Market Street (watch your step!) in San Francisco and count the shuttered store fronts. Take Caltrain down to San Jose, and see if you can believe that it is the richest city in the United States, per capita. The massive increase in wealth has not connected with a meaningful way with average people’s lives even in the heart of tech country, let alone in the forgotten corners of the country.
The people who run Silicon Valley identify with progressive values. They’re not bad people. They worry about these problems just like we do; they want to help.
So why the failure to do anything?
Like T.S. Eliot wrote:
Between the idea
And the reality
Between the motion
And the act
Falls the Shadow
As I said earlier, the tech industry hates messy problems. We’d rather dream up new problems we can solve from scratch.
One reason nothing happens is a culture of tax evasion. There’s a folk belief in American business that if you pay full taxes, you’re not doing your fiduciary duty, and your board will fire you.
Apple now has a quarter trillion dollars offshore that it refuses to put into direct productive use in the United States. Apple boasts that its products are designed in California—they will sell you a $300 book called Designed By Apple In California. But they do their damndest to make sure that California never sees a penny of their overseas profits.
You in the EU are all too familiar with this brand of tax evasion. Google, Amazon, Apple, Microsoft have all been under investigation or in court on charges of evading European taxes.
Another reason good intentions don’t translate is that capitalism, especially venture capital, doesn’t work very well when there is vast wealth inequality.
The richest 20 people in tech control a fortune of half a trillion dollars in personal wealth, more than the GDP of Sweden.
This small subculture of wealthy technophiles promotes investment into luxury goods for rich people, or into “mom as a service” types of companies that cater to spoiled workaholics in the tech industry. And so we end up with things like a $120M juice squeezer, or three startups competing to deliver organic baby food.
Silicon Valley brings us the worst of two economic systems: the inefficiency of a command economy coupled with the remorselessness of laissez-faire liberalism.
One reason it’s been difficult to organize workers in the tech industry is that people have a hard time separating good intentions from results. But we have to be cold-blooded about this.
Tech companies are run by a feckless leadership accountable to no one, creating a toolkit for authoritarianism while hypnotized by science-fiction fantasy.
There are two things we have to do immediately. The first is to stop the accelerating process of tracking and surveillance before it can do any more harm to our institutions.
The danger facing us is not Orwell, but Huxley. The combo of data collection and machine learning is too good at catering to human nature, seducing us and appealing to our worst instincts. We have to put controls on it. The algorithms are amoral; to make them behave morally will require active intervention.
The second thing we need is accountability. I don’t mean that I want Mark Zuckerberg’s head on a pike, though I certainly wouldn’t throw it out of my hotel room if I found it there. I mean some mechanism for people whose lives are being brought online to have a say in that process, and an honest debate about its tradeoffs.
I’m here today because I believe the best chance to do this is in Europe. The American government is not functional right now, and the process of regulatory capture is too far gone to expect any regulations limiting the tech giants from either party. American tech workers have the power to change things, but not the desire.
Only Europe has the clout and the independence to regulate these companies. You can already point to regulatory successes, like forcing Facebook to implement hard delete on user accounts. That feature was added with a lot of grumbling, but because of the way Facebook organizes its data, they had to make it work the same for all users. So a European regulation led to a victory for privacy worldwide.
We can do this again.
Here are some specific regulations I would like to see the EU impose:
- A strict 30 day time limit on storing behavioral data.
- The right to opt out of data collection while continuing to use services.
- A ban on the sale or transfer of behavioral data, including to third-party ad networks.
- A requirement that advertising be targeted strictly to content, not users.
With these rules in place, we would still have Google and Facebook, and they would still make a little bit of money. But we would gain some breathing room. These reforms would knock the legs out from underground political ad campaigns like we saw in Brexit, and in voter suppression efforts in the US election. They would give publishers relief in an advertising market that is currently siphoning all their earnings to Facebook and Google. And they would remove some of the incentive for consumer surveillance.
The other thing I hope to see in Europe is a unionized workforce at every major tech company. Unionized workers could demand features like ephemeral group messaging at Facebook, a travel mode for social media, a truly secure Android phone, or the re-imposition of the wall between Gmail and DoubleClick data. They could demand human oversight over machine learning algorithms. They could demand non-cooperation with Trump.
And I will say selfishly, if you can unionize here, it will help us unionize over there.
If nothing else, we need your help and we need you to keep the pressure on the tech companies, the Trump Administration, and your own politicians and journalists, so that the disaster that happened in the United States doesn’t repeat itself in Germany.
You have elections coming soon. Please learn from what happened to us. Please stay safe.
And please regulate, regulate, regulate this industry, while you can.