Uber's Greyball: The Dark Side of Tech Innovation

TL;DR

Uber's 'Greyball' was a secret system to deceive regulators, sparking a debate: is breaking the rules a necessary evil for disruptive innovation? #VentureStep #Uber #TechEthics

INTRODUCTION

Uber's rapid global dominance wasn't just the result of a revolutionary app; it was fueled by an aggressive, win-at-all-costs mentality. This approach often involved operating in legal and ethical gray areas, but no tactic was more sophisticated or controversial than a secret program known as "Greyball." This tool was specifically designed to identify and deceive government officials, allowing Uber to launch in cities where it was not yet legally approved.

In this episode of VentureStep, host Dalton Anderson dives deep into the infamous Greyball case. Drawing on his background in programming and data science, Dalton unpacks the intricate mechanics of this system, exploring how Uber waged a high-tech cat-and-mouse game against regulators. The conversation moves beyond the technical details to confront the difficult questions this strategy raises for founders and tech leaders.

From geofencing government buildings to digital fingerprinting and scraping social media, the episode reveals the escalating complexity of the Greyball program. Dalton examines the fallout, including the public backlash and the eventual ousting of CEO Travis Kalanick, and frames the debate within the larger "move fast and break things" ethos. Is this level of deception a necessary evil for innovation, or is it a line that should never be crossed?

KEY TAKEAWAYS

• Greyball was a complex, multi-layered system designed to identify and deceive regulators by showing them a fake version of the Uber app with no available cars.
• Uber waged a sophisticated "whack-a-mole" game with officials, escalating from simple geofencing to tracking credit card data, digital fingerprinting devices, and even banning cheap "burner" phones.
• Uber's public defense was that the tool targeted "undesirables" who violated their terms of service, a legal ambiguity that helped them navigate the repercussions.
• The strategy raises a critical founder's dilemma: if companies like Uber and Airbnb had followed every rule, they might never have succeeded, yet their methods challenge fundamental ethical boundaries.
• The Greyball scandal, combined with a toxic work culture, contributed significantly to the public backlash that forced CEO Travis Kalanick to resign.

FULL CONVERSATION

Dalton: Welcome to VentureStep podcast where we discuss entrepreneurship, industry trends, and the occasional book review. Uber's ride-sharing service and dominance has a dark side. Today we're going to be discussing one of the most sophisticated acts for avoiding regulatory and legal action in what is coined the Greyball case.

Dalton: On today's agenda, we'll be discussing what Greyball is and how it worked. What were Uber's motivations for using such a deceptive technology? What are the legal and ethical implications of Greyball? And broader questions about corporate or, in this case, regulatory espionage.

It's in between two worlds of corporate espionage, regulatory arbitrage, and just general deception.

Dalton: It's kind of a gray area and that's why it was termed Greyball. Corporate espionage would be stealing from another company. And so it's not really doing that, but it is stealing protected areas where taxi companies have their license with the state and the city. Then Uber comes in and the taxi drivers are complaining, "Uber is taking our jobs and revenue." And then the city officials try to use the app and they're like, "The app doesn't work for me. I could see the cars but no one's picking me up."

Dalton: But before we dive in, I'm your host Dalton Anderson. My background is a bit of a mix of programming, data science, and insurance. Offline, you could find me running my side business or lost in a good book. You can listen to this podcast in video and audio format on YouTube. And if audio is more your thing, you can find the podcast on Apple Podcasts, Spotify, or wherever else you listen.

What Was Uber's 'Greyball' Program?

Dalton: Okay. So the first thing that we're going to dive into is what is Greyball. Greyball was a sophisticated act to detect "undesirables" that would potentially violate the terms and service of Uber. Really what that means is it was a way for Uber to detect people that might cause the company harm. The people that they deemed would cause the most harm were individuals that could regulate or take legal action on the company.

The 'Two Apps in One' Deception

Dalton: So they created two apps. There's the app that you and I would use. And then there's a different app that is on the same app store. You download one app, but the app is actually two apps, and one works and the other doesn't. The app that works and the app that you and I use, it would show the cars, you could book a ride, and all this other stuff. The app that the regulators were seeing was an app where you would log in and you would see everyone getting picked up, but you couldn't get a ride.

Dalton: I read a story where one of the regulators was getting really frustrated that they couldn't get a ride, but someone right next to them could. You can't get a ride, but you could see all the car activity and people driving around. And then when you'd go to book a ride, it would say "no rides available" or it would just take forever to book and then say "no rides available." So the regulators really couldn't get any evidence on Uber because they were trying to get firsthand accounts of interacting with the drivers and how the app works. They started this around early 2014.

When Is Breaking the Rules Justified?

Dalton: I recently recalled this because I had a conversation with some founders at the InsureTech Conference in Las Vegas. I was talking to this Australian founder and, in a general sense, what she was saying her company would do was illegal. And I called that out, basically framing it like, "Hey, in America, what you're saying you want the company to do is illegal." And she's like, "Yeah, I know."

And she's like, "Well, think Uber, think Airbnb. If they followed every single rule from the governments that they try to operate in, they would have died."

Dalton: She argued that the value Uber and Airbnb created for society was worth the ethical questions they faced. Overall, it was a net positive. I think that's a fair perspective, but then there are other groups that might take that in a different direction.

Corporate Scandals: Where Uber Drew The Line vs. Others

Dalton: There are some examples where people did get hurt. Volkswagen lying about emissions, where they programmed their diesel engines to have a reduction in emissions. Then there's Purdue Pharma, who really pushed for Oxycontin and said it wasn't addictive. Then we have Theranos by Elizabeth Holmes, which was a complete fraud. And then we have Lumber Liquidators who sold laminate from China that had carcinogens. Those are a couple of examples where they're pushing the legal barrier in a bad way, where people are getting hurt. This Uber example and the Airbnb example, not so much. People might get upset, but no one's really dying or getting cancer or being defrauded out of billions of dollars.

The High-Tech Cat-And-Mouse Game With Regulators

Dalton: So how did the Greyball system work? It got started because a lot of the taxi companies were complaining that some kind of app was stealing their revenue. How did they actually do it? They had a couple of metrics to detect if people were from the government.

This is almost a company in itself. Like the amount of stuff that they're doing to detect if they're a government official was nuts.

Dalton: First off, they were geofencing all of the government buildings—federal, state, and city buildings. Within the geofence, they called it the "blackout zone," so there were no rides taken in that area or dropped off in the area. So the regulators started walking outside of this geofence and then calling rides there to build a case. Uber understood that and just started banning those accounts.

Dalton: The next thing the regulators did was sign up as a fake person, but they would use their normal credit card. Uber understood that and knew this one account that was banned is using similar bank account information. So let's ban that one.

From Geofencing to Digital Fingerprinting

Dalton: So then they were like, "Well, I think it's my phone." And that was true. What Uber was doing was a digital fingerprint, which is definitely against the terms of service of Apple. When you delete an app, you've got to wipe all their data. What Uber was doing was keeping some sort of key. So when they re-downloaded the app, the key would match up and it would detect that this is the same person.

Dalton: The next thing the regulators started to do is they started buying off-the-shelf burner phones. And what Uber did was they just started banning all low-quality phones. They would also scour social media like LinkedIn or Instagram to build out these digital profiles of the regulatory individuals. And they also had this thing called "eyeballing," which was detecting suspicious activities and then just banning you. If you were making too many accounts or downloading and then deleting, they would just ban you.

The Fallout: Whistleblowers and Public Backlash

Dalton: So the regulators knew something was up, but couldn't necessarily put their finger on it. And Uber had just built out what started as a simple means to reduce their likelihood of getting regulatory action.

What started out as kind of childish became something way too big... this sophisticated dark monster that they created, which is called Greyball.

Dalton: Once you start linking credit card information and device verification and scraping social media to create these digital profiles, it becomes very sophisticated. It's almost a fraud detection system they built in-house just to avoid regulation. Internally, people felt uncomfortable with it. Eventually, there was a whistleblower and then there was the New York Times exposé. And Uber was like, "Hey, you got me." It became a big deal. At the time, 250,000 people deleted the app. But during the three or four years they were using Greyball, they were able to launch in so many countries and major cities, and the advantage they got on their competitors was massive.

Dalton: The issues that happened were, okay, they lost 250,000 customers. They got in big trouble. They had a huge brand reputational issue on top of the bad work culture at Uber—the sexual harassment cases, the racism reports, the issues with aggression in the workplace. This, on top of all that, definitely put the nail in the coffin for Travis Kalanick. He was asked to resign. I think the five most major investors in Uber insisted he resign. He chose to resign and was out.

The Legal Defense: Targeting 'Undesirables'

Dalton: On the legal scrutiny, they faced legal action and potentially a justice probe from the US government. The way they were able to get out of it was that they said they weren't targeting regulators. They were targeting people that were against their terms and service. Their quote-unquote "undesirables" were people who used low-end phones, people who worked in government areas, people who constantly made new accounts and deleted their old accounts. That was their explanation and they were able to legally get out of it. It seems pretty clear to me that they were targeting regulators, but whoever their lawyer was did a great job.

The $90 Million Price Tag for Public Opinion

Dalton: My understanding is that it wasn't a surprise that it was eventually going to come out. They knew what they were doing was way too sophisticated for it not to be released. They probably already had a budget for lobbying and other things. They spent close to a hundred million dollars—I'm pretty sure it was $90 million—on just lobbying, legal changes, and influencing politicians and public opinion to allow Uber to operate and improve their public perception.

The Founder's Dilemma: Innovation vs. Ethics

Dalton: So where does Uber stand in this ethical dilemma? Would you consider what Uber did wrong? I think so, yeah. But is it necessarily hurting people? And where do you draw the line? At the same time, if they didn't do it, there would be no Uber. There would be no ability for people to have a way to supplement their income. I think there's a middle ground they could have approached.

At what point are you just avoiding your problems instead of facing them head on?

Dalton: I think once they got to a certain scale, they should have been like, "Hey, we are doing these things that we shouldn't be doing and I would like to make these things right." We needed to grow and we couldn't wait. But keeping this whole Greyball thing going for four years and getting to that level of sophistication, I don't agree with that. But on the other side of the coin, if they just said, "Okay, I'll do whatever you say, let's go through all the proper rules," it would have never worked.

Is This Just a Modern-Day Trolley Problem?

Dalton: Where do you draw the line on a startup driving into the unknown and changing public perception? LinkedIn wasn't an original idea. Vine was a video-sharing platform very similar to TikTok. Hindsight's 2020, but it might not be the right time for that product. Maybe you force the times, but at what point do you call it where it's too far?

At what point are you, and I'm talking to you, the listener, gonna call it where you're like, this is too far?

Dalton: An easy win would be if people are dying or getting hurt financially or their health is deteriorating. But in this Uber example, you were hurting a small group of individuals—the taxi license owners. You were hurting them and the rest of the people were benefiting. Is this similar to the trolley problem where you sacrifice one to save the many? Because the trolley problem to me is pretty easy. If you're sacrificing one to save thousands, is it that big of a deal? I think that's the essence of management and being a leader—making those tough decisions.

Dalton: I find the topic quite interesting and I hope you do. If you have any suggestions or commentary regarding the topic, then please go ahead. I'd love to hear your opinion. But once again, this is where we say our goodbyes. Wherever you are in this world, good night, good evening, good morning. I hope you have a great day and I hope that you tune in next week. See everyone. Goodbye.

RESOURCES MENTIONED

• Companies: Uber, Airbnb, Volkswagen, Purdue Pharma, Theranos, Lumber Liquidators, Apple, Google, Waymo, Lyft, LinkedIn, Vine, TikTok, Facebook, Instagram
• Events: InsureTech Conference (ITC) Las Vegas
• Universities: Florida Atlantic University (FAU), Florida State University (FSU)
• Publications: The New York Times

INDEX OF CONCEPTS

Dalton Anderson, Elizabeth Holmes, Travis Kalanick, Uber, Greyball, Airbnb, Volkswagen, Purdue Pharma, Oxycontin, Theranos, Lumber Liquidators, Apple, Google, Waymo, Lyft, LinkedIn, Vine, TikTok, Facebook, Instagram, InsureTech Conference, ITC Las Vegas, Florida Atlantic University, FAU, Florida State University, FSU, The New York Times, Boca Raton, Palm Beach County, Tallahassee, corporate espionage, regulatory arbitrage, geofencing, digital fingerprinting, terms of service, whistleblower, trolley problem