Revenge of the AI Nerds
How Blockchain Will Save Our Lives and Pay You for Your Data
“I think it will take 15 years for a big chunk of jobs to be displaced. ” — Kai-Fu Lee on the automation threat
In the sci-fi movie Her set in a near-future Los Angeles, the protagonist Theodore purchases an operating system with an AI-powered virtual assistant, Samantha voiced by Scarlett Johansson. The two hit it off and end up getting romantic, despite Samantha being a computer-generated voice with no body. The movie came out in 2013 and seemed eerie at the time. Having rewatched it recently, I found myself thinking, “this will probably happen.”
In order to determine such trends of the future, we should look at what’s going on today at the fringes of society, says thinker and futurist Kevin Kelly. Computer nerds were tinkering with computers before they were adopted on a mass scale. The cypherpunks were enthralled with cryptographically encoded money years before the bitcoin and the blockchain emerged.
And now an idea formerly on the fringes, Universal Basic Income (UBI), is quickly coming to light with the help of data computing and the blockchain. UBI is a type of social security system whereby every citizen would receive a regular, periodic payment. It’s a proposed solution to the threat of AI and automation taking our jobs in the coming decades.
This fear of job displacement has been voiced for a long time with little to show. In the early 1900’s horse-driven carriages disappeared, but new jobs were created as the automobile industry emerged. Even Uber, while highly disruptive to the taxi industry, created new opportunities for anyone to turn their car into an income-generating machine.
However, as technologies continue to permeate every aspect of our lives, the trend is going in the opposite direction, to more jobs disrupted and fewer jobs created. Massive industries like finance, manufacturing, and retail are no longer seeing job growth. McDonalds in Arizona is already being fully run by robots, and dozens of fast food chains will automate their restaurants. Huge media publishers like The Washington Post, Yahoo! and Forbes are using AI to generate articles; unlike a human writer, it’ll never complain or take sick leave. And despite Uber’s growth, their ultimate goal is to have self-driving cars and eliminate the need for drivers all together.
These new technologies are going to destroy more jobs than they will create, and a McKinsey report predicts that robots will replace up to 800 million jobs by the year 2030, thanks to advances in robotics, machine learning, AI and automation. (for more on this check out the great book, Rise of the Robots)
In response to this threat, presidential candidate Andrew Yang has put it at the forefront of his campaign, vowing to give Americans a $1,000 monthly stipend, no questions asked.
He says, “…the smartest people in the world now predict that a third of all working Americans will lose their job to automation in the next 12 years. Our current policies are not equipped to handle this crisis. Even our most forward-thinking politicians are unprepared.”
In the US, support for UBI has risen from 12% to 48% (Northeastern University-Gallup survey). Most models pay for the system by providing everyone the same payout, irrespective of work input. The money to enable UBI could be funded from cutting military spending (unlikely), phasing out or even creating a new tax. While there are interesting proposals, the recently failed UBI experiments in Finland and Canada show us that distributing money in an intelligent way is not as easy as it seems.
Most importantly, we need to decide whose wallet this is going to come from. The answer might be literally right under our fingertips — harnessing and selling our data to create universal basic income. This may sound like a futuristic idea at first, but today with the advent of new blockchain technology, it’s looking more and more feasible.
In order to understand how this feat would be possible, we first need to understand how we got to our present situation.
The New Data Overlords
“Donald Trump warned voters that the Mexicans and Chinese would take their jobs, and that they should therefore build a wall on the Mexican border. He never warned voters that algorithms would take their jobs, nor did he suggest building a firewall on the border with California.” — Yuval Harari, 21 Lessons for the 21st Century
Enlightenment thinker John Locke classified our three unalienable rights as life, liberty and estate. He defined estate as the idea that “everyone is entitled to own all they create or gain through gift or trade so long as it doesn’t conflict with the first two rights.” Locke said this over 300 years ago — 100 years before trains, 200 years before air conditioning, and 280 years before the internet.
If Locke were alive today to update his view, he might argue that data falls under this third right. After all, data is something that ‘we create through gift or trade,’ isn’t it? Each time you like a Facebook entry, post and share an Instagram photo or choose a movie on Netflix, you are generating valuable data. Everything you do on Google Maps, on Gmail, on Google Docs, and pretty much anything you do “in the cloud” is mined for data. Google has over 14 exabytes of data that it is processing, or over 16 billion gigabytes. That’s a buttload of data.
The more data that Google is able to amass and the longer you use Google, the more valuable your data becomes. When technology is able to perform tasks cheaper and faster than humans, then we are no longer necessary. In other words, in carrying out unpaid mundane digital tasks on Google and Facebook, you are not only creating valuable data for these marketing monoliths but also for the machine learning robots that will one day replace your job.
This may not be intuitive, but it is explicitly what Google cofounder Larry Page envisions.“The goal of disruptive technology companies, in the statistical sense, is to reduce GDP,” Page says. “To wipe out transaction costs, which are being measured, and to replace them with convenience, which is not being measured. So the economy is shrinking, but everyone is getting a better deal. Lots of what tech is doing is destroying what wasn’t needed. The end result is you’re going to have less of an economy, but higher welfare.” (source: David Pilling’s in the book The Growth Delusion).
Unfortunately, Page doesn’t explain exactly how we’re going to achieve this ‘higher welfare’ after technology proceeds to shrink our economy. From the rising income inequality and lowered economic mobility, it seems more likely that this technology-driven ‘welfare’ will go to the hands of a few.
But I like taking selfies!
While you may be alarmed, annoyed, or perhaps a bit of both, you might also argue that you’re fairly remunerated by access to using these great services, discounts and freebies like free messaging, search and email tools that make daily life easier. “I like taking pictures every time I get excited,” you might add.
There’s no denying that tech companies have provided us with a new standard of living and services that we do love. I use these services too. The point is not to condemn them or try to return to a world without them, which is neither feasible nor desirable. Rather, it’s to think about the future and question the underlying agreement that we have in our current digital society.
The problem is that despite your contribution, you don’t actually own your data. The largest shareholders of Google, Amazon, Alibaba, and Facebook own your data. These internet giants then make money selling this data to advertisers. They can do whatever they want with it.
How did this happen? These companies were able to “blitz-scale” and dominate because there were no property rights rules governing the world of data and privacy. These data overlords raised boatloads of money because they figured out a way to drive lots of traffic to their sites, and lots of traffic creates lots of data.
Where things get dark
“I am now convinced that Google searches are the most important dataset ever collected on the human psyche.” — Google data scientist Seth Davidowitz
Google can already predict flu epidemics before they happen based on people’s searches for symptoms. This data could actually be quite helpful to society if used wisely. More sinisterly, if someone is searching for “how to kill someone” and “body disposal techniques,” Google knows about it too.
The question is, then, if the person making those searches was your neighbor, would you want to know about it? Do you have a right to know about? Maybe they’re a student doing a harmless research paper for a criminology class. Or maybe they’re planning on killing you.
The moral and ethical implications are interesting to debate. But here’s the point — it doesn’t matter what you think about the above example. If Google controls your data, then you will have absolutely zero say in it either way. They can use your search data any way they want; for or against you.
Therein lies the problem. It’s that the plot of Minority Report where people are arrested for crimes before they are committed is not only realistic, it’s possible today with the technology we have. It’s a much deeper problem than technology just taking our jobs away.
We are like turkeys, happily enjoying our days strolling in the gardens of our owner until we’re met with the reality of their shocking change of heart on Thanksgiving Day. But unlike a turkey who is oblivious to the butcher sharpening the knife, we should be smart enough to take a hint.
Amazon knows when you eat, shit and sleep
By 2020, Amazon, Google, and Facebook will make up 63% of all US advertising spending. One has to ask, how are they increasing value for consumers? Google has added small features to Google Maps, but generally, it hasn’t changed much. However, the revenue per user for Google Maps has increased. Where is the extra value is being derived from? Yep, you guessed it. Your data.
This data is about to increase substantially in value. The Internet of Things (IoT) is connecting all aspects of our lives and creating rich data flows throughout your day. Google already knows your deepest secrets — the ones you don’t share with your closest friends or family, but those you type into the search bar at 3 a.m. when no one is looking (except that someone is watching).
Alexa, whom you have so kindly invited into your home, is getting rich off of you — not only from your music choices and grocery orders but virtually everything that has any digital input. If your Amazon echo or Google home is connected to your Philips light bulb, then Amazon knows when you go to sleep. The cascade of connected devices — from toilets to stoves and everything beyond — means that Amazon, to put it bluntly, will soon know when you eat, shit and sleep.
In a world like this, don’t you at least want some control over your data?
Data as Universal Basic Income
What if, instead of a government-led solution, everyone was paid for their data? If individuals were paid for their data, this data revenue could then be used to create a universal basic income. Without a doubt, the spectre of more taxation made UBI a no go for many. But data-driven UBI could draw the social tax ‘burden’ from the passive income and dramatically reduce/slash our tax bill. Perhaps data is the missing piece that enables a functioning system for UBI.
We already make money from leveraging technology. I can take my iPhone out and list my apartment on Airbnb, become an Uber driver, or become a freelancer on Upwork. Underneath all of my interactions is a Matrix-like stream of data that is connected to my personal profile over thousands of interactions. What if I were able to leverage this, too?
Imagine getting paid for providing personalized and relevant data to marketers. Rather than being creeped out by irrelevant shopping ads that retarget you based on your past history, you would know how that data was being used, and in turn perhaps you’d be more willing to share it. The data would be more accurate and rich, which would be more useful to companies, and good for you because you could get ultra-personalized recommendations that you’d actually want. It would be a win-win.
How much is my data worth?
It’s hard to come to an exact figure to determine how much your data is worth, but we know that revenues from the ad industry (how Facebook and Google make 80% of their money) were over $130 billion in the U.S. alone. Facebook’s average yearly revenue per user is $108, which may not seem like a lot of money, but when you combine all of the online/mobile services you use, this number is likely much higher.
In an interesting experiment, Federico Zannier put up all his data for sale, including all his web browser history, PC activity, and photos including timestamps and geolocation. He was able to rake in $2,733.
Keep in mind that some of the most valuable types of data are financial, insurance, medical and other personal data you normally wouldn’t share online. That said, they are starting to be shared online — take for example 23andme, which is already selling your genetic data to third parties. There are also companies working on app stores for your genetic data. There are ways in which our data will be shared, parsed and aggregated that we cannot begin to fathom.
The value of your data may not be enough to make UBI effective as a monthly stipend. Maybe it’s just a few hundred dollars a month. However, with the rise of automation and the previously mentioned multi-billion dollar IoT market that connects our cars, houses, hardware, software and pretty much everything — the value of our data will continue to increase.
Okay, screw Facebook and Google. Can’t I just keep and sell my own data?
Unfortunately, it’s not that simple (yet). Imagine Uber trying to convert every car on the street into personal drivers without the widespread network of smartphones — it’s virtually impossible. Similarly, while there are mountains of data in your possession, the average person does not have the capability to simply sell their own data. It’s too difficult to identify who wants your data, package your own data, and then try to sell it. There needs to be an underlying infrastructure that you can easily use. Currently, there’s simply too much friction and technical expertise required for the average Joe.
Revenge of the AI Nerds
What if the top AI, machine learning and data security minds formerly working for Google, Facebook, and Microsoft came together, and instead of making more tools for the Internet monoliths, provided them to the public to use instead?
Enter the Harmony blockchain network. The former tech executives of Amazon, Google and Microsoft that run Harmony have raised $18 million in funding, but still work out of a garage (usually a good sign). They are fusing their knowledge to hand this AI-powered data control over to individuals.
Harmony is a public blockchain project that’s aimed at creating technology that will allow you to own and monetize your data. They’re using blockchain to solve this problem on a technical level with a two-pronged approach — deep sharding and zero-knowledge proofs.
The first piece, deep-sharding, solves the problem I described previously about creating an infrastructure that actually allows the average person to monetize. There needs to be a fast enough network that can hold massive amounts of data, quickly confirm and validate the use of data and disburse incentives accordingly. Deep-sharding is a new network layer that allows for that speed and scalability.
The second piece, zero-knowledge proof, solves the privacy concerns. It allows people to validate information without revealing all their data. In this way, you would be able to share your income (a very personal question) without necessarily revealing your identity. [crypto nerds can read more about the tech here].
Harmony’s solution isn’t just about monetizing your data but marks an ideological shift from data centralization to data sovereignty.
And they aren’t the only ones that are working on it.
Two-thirds of Korean mobile phone owners are already making money across their apps with an application called AirBloc. Another app, Wibson, pays you in tokens for your social media data. UK startup digi.me allows you to upload and store data in a single app where you have control over it.
A shift from data owned by companies to data owned by you may be the most powerful wave of democratization since Athenian democratic rule in the sixth century BC Greece. The data-driven universal basic income model shifts from government distribution to a bottoms-up personal data ownership and monetization model — and the underlying blockchain tech is now making this a possibility.
Owning the Fuel That Feeds the Machines
In the industrial world, the oil barons owned the fuel of the economy. In the digital economy, the fuel of the learning machines is data. Robots may indeed take our jobs packaging the books, groceries and other products we buy on Amazon or Alibaba. If it makes our lives easier, then so be it. But as we order these products, share and recommend them, we will be making the money on these digitized daily tasks, not the Internet monopsonies.
As fear-mongering politicians create a false reality of robots running the world, Harmony and others are creating a different future with a different ending. Like the robber barons of the gilded age who controlled gold and oil in the 19th century, the internet monoliths have commoditized and controlled our data. With the help of the AI Nerds building the new decentralized data infrastructure — the railroads of the digital age — consumers are gaining control of their data through connected apps. Data is still a commodity, but we are gaining ownership of the commodity. Our personal data will be our most valuable economic input on both an economic and social level.