by Steve Levy
Since 2005, Google had been the most forceful voice in corporate America for the concept of net neutrality. When Google had begun this argument, net neutrality was closely aligned with the company’s self-interest: an outsider to the establishment, it depended on the free access the Internet offered. But as Google became one of the biggest players in the establishment, it was clear that even if online companies had to pay Internet service providers for access, the search giant could afford those fees. It was thus in a position to bar the door to future innovators but chose not to use that power. That fact gave credibility to Google’s argument that it was pushing for an open Internet not just for itself but for the next Google, the next YouTube—for innovation itself.
But now Google was saying that it had recalibrated its views on net neutrality. Working with one of the putative villains of the net neutrality battle—the huge telecom Verizon—Google proposed a new framework that would grant neutrality to land-based Internet service but not include the fast-growing area of wireless communications. Even worse, one of its blood rivals, AT&T, hinted that the Google statement was a positive step. Critics instantly pounced on the betrayal, and for a relatively wonky issue, the flip-flop drew broad coverage. Google’s positioning meshed with other issues of the day—privacy, the Apple competition—to create an underlying narrative that the company was no longer a source of goodwill but just another corporate bully.
Jon Stewart devoted a segment of The Daily Show to Google’s “sellout.” A Taiwanese website specializing in current-events animations made a short video on the decline of Google’s values with a memorable shot of a cartoon Eric Schmidt clinking his wineglass with the Verizon representative, who was in devil guise. After the deal was sealed, Schmidt grew his own devil horns and bellowed a bawdy world-domination laugh. With the Street View Wi-Fi scandal still generating outcries, there was suddenly a critical mass of Google disillusionment. Even random developments, such as the news that Google had ordered a sophisticated surveillance-capable autonomous drone, poured rocket fuel on the conflagration. (Actually, the drone was a private purchase by Android honcho Andy Rubin, ever the robotics enthusiast.)
On August 13—a Friday—protesters took to the Googleplex. The scene was more a geek version of Yippie theater than an angry riot; the highlight was a musical tribute to Google’s perfidy by a singing group called the Raging Grannies. Yet the groups behind the event—including MoveOn, Free Press, and the Progressive Change Campaign Committee—represented true disenchantment by Google’s former allies. And they carried a petition of displeasure with 300,000 signatures. Their signs read google, don’t be evil.
Back when the company was young, Larry Page and Sergey Brin could startle people with their conviction that Google was destined to be a huge company that would change the world. In subsequent years, the prescience of those assertions would be recalled with a sense of awe. But Page and Brin’s clairvoyance never extended to a day when they would mobilize their company not to create the next revolution but to fight a rearguard action against a competitor’s revolution. At the same time, as it was developing its new catch-up product, Google was shutting out the noise of protesters objecting to a sellout of the company’s principles, begging it not to be evil.
Google had not turned evil. It still pursued social innovation regardless of profits. Its corporate culture remained uniquely geared to the most literate and brainy products of the Internet era, and its leaders still believed in a future guided by benevolent algorithms of loving grace. But by chasing Facebook’s taillights, Google was behaving very much like the kind of corporation that Larry Page once promised it would not be: conventional.
Yet in other quarters, the company was still launching moon shots. For instance, in late 2010 came news of its most audacious projects yet. Back in 2007, Larry Page had convinced Sebastian Thrun, the head of Stanford’s Artificial Intelligence lab and the leader of the team that built the autonomous robot car named Stanley, to take an academic leave to work at Google. Thrun had initially worked on Street View technology, but in early 2009, Page commissioned him to develop self-driving Google cars that would ride on actual roads and set the stage for the technology to reach the mainstream. Thrun gathered an all-star team of roboticists and A.I. specialists and, in effect, created a follow-up to the 2005 contest where Stanley had prevailed. This time, the goal was to have autonomous Toyota Priuses negotiate a complicated 1000-mile course around California, including a cruise down the Pacific Coast Highway, a run through Beverly Hills, and a virtual obstacle course in the Bay Area, which included the twisty streets of San Francisco and (hardest of all) a narrow unpaved road in Tiburon in Marin County, where oncoming cars forced drivers to reverse into driveways of the nearest home in order to let them pass. (Google employees rode in the driver’s seats, ready to take charge in case of computer failure.) After over 140,000 miles of test driving, Google’s cars passed the test. The only accident occurred when one of the Google cars was rear-ended at a red light by a human driver.
Critics charged that the project was a sign of Google’s lack of focus—why was an Internet search company working on cars that drive themselves? Actually, the project was well within Google’s wheelhouse. Since its earliest days, Brin and Page have been consistent in framing Google as an artificial intelligence company—one that gathers massive amounts of data and processes that information with learning algorithms to create a machinelike intelligence that augments the collective brain of humanity. Google’s autonomous cars are information-collectors, scanning their environment with lasers and sensors, and augmenting their knowledge with Street View data. (Unlike human drivers, they always know what’s around the corner.) “This is all information,” says Thrun. “And it will make our physical world more accessible.”
What will Google’s explorations in artificial intelligence eventually yield? Will we routinely cruise in autonomous cars powered by Google—undoubtedly capable of pointing out sightseeing highlights and culinary opportunities as they whisk us to destinations? Will the brain “implant” that Larry Page referred to in 2004 become a Google product at some point? (In late 2010, introducing the Google Instant search product—once referred to internally as “psychic search”—Sergey Brin had repeated the sentiment: “We want Google to be the third half of your brain.”) Google, after all, was founded on the premise that the best path to success is doing what the conventional wisdom says you cannot do. In an era of unprecedented technology leaps, that has turned out to be an excellent premise. “It’s quite amazing how the horizon of impossibility is drifting these days,” says Thrun.
The revelation of the autonomous vehicle program at the end of 2010 had all the earmarks of a Larry Page project—scary ambition, groundbreaking AI, massive processing of information in real time, and rigidly enforced stealth. (Only when a reporter learned of the project did Google agree to talk about it.) The glimpse it provided of Page’s priorities turned out to be more significant than expected when an apparently predestined change in Google’s leaders occurred sooner than observers had expected.
On January 20, 2011, Google began its quarterly earnings call (trumpeting yet another record high in revenue—$8 billion for the quarter, making the 2010 total almost $30 billion) by announcing that in April Eric Schmidt would step down as CEO. He would assume a new title, Executive Chairman. His replacement would be Larry Page.
“I believe Larry is ready,” said Schmidt. In addition to advising Page and Brin (whose new title, “Co-Founder,” was conveniently vague), he would focus on presenting Google’s case to regulators and critics, he announced. The troika explained that they had been discussing the change for months, but had accelerated the talks during the end-of-year holidays. Some observers wondered whether Schmidt’s departure was a consequence of being outvoted in the intense debates over the China problem during the previous holiday season—for the past year Schmidt had often been away from Mountain View appearing at numerous conferences, sometimes engaging in desultory speculation
about the technological future. But an assessment of Larry Page’s consistently ardent possessiveness over the company he cofounded (still blessing or rejecting the hiring of every single employee in a workforce that now approached 24,000) indicates that all during the Schmidt era, Page had been the once and future leader of Google.
Less than a year earlier, at the end of a long interview, I had asked Page whether he would become CEO when Schmidt stepped down. He ducked the question. He wasn’t even comfortable saying whether working at Google would be a lifelong situation for him. “I think it’s hard to predict what happens in your life and the changing conditions, but I’m very committed to the company and I really enjoy what I do,” he said. “And I think I’m able to positively affect a lot of things, which makes me feel really good, and I don’t see any likely change in that.” That was the closing note of the conversation. But a few minutes later he returned, wanting to say more. He wanted to reiterate some earlier points he made about ambition.
“I just feel like people aren’t working enough on impactful things,” he said. “People are really afraid of failure on things, and so it’s hard for them to do ambitious stuff. And also, they don’t realize the power of technological solutions to things, especially computers.” He went on to rhapsodize about big goals like driving down the price of electricity to three cents a watt—it really wouldn’t take all that much in resources to launch a project to do that, he opined. In general, society wasn’t taking on enough big projects, according to Page. At Google, he said, when his engineers undertook a daunting, cutting-edge project, there were huge benefits, even if the stated goal of the project wasn’t accomplished. He implied that even at Google there wasn’t enough of that ambition. “We’re in the really early stages of all of this,” he said. “And we’re not yet doing a good job getting the kinds of things we’re trying to do to happen quickly and at scale.”
Now Larry Page would be running Google, and he would get his chance to fulfill unbounded ambition. But he would also have new responsibilities that present considerable challenges to a Montessori kid who hates meetings, doesn’t want administrative assistants, and has little patience for schmoozing and politicking.
It had been almost exactly ten years since Page and Brin had hired Schmidt, backing down from their insistence that they could run the company by themselves. Schmidt’s comment posted on Twitter looked back to that day when Google was quite a different company and Larry Page was a twenty-eight-year-old unschooled in management.
“Day-to-day adult supervision no longer required,” Schmidt tweeted.
The veracity of that statement remains to be seen. But one thing seems indisputable: Larry Page would not be a conventional CEO. Google’s future would continue to court the unexpected. And maybe the impossible.
ACKNOWLEDGMENTS
Google is known for its willingness to take risks. But it took an uncharacteristic and brave one in allowing a journalist to spend hundreds of hours with its employees, look over engineers’ shoulders as they developed products, and sit in on TGIFs, GPSs, and other councils and confabs. I don’t know how deep a breath Eric Schmidt, Larry Page, and Sergey Brin took before approving the project, but Elliot Schrage, then the head of Google’s global policy, clearly exhaled a sigh of relief at the sign-off, as did his colleagues David Krane, Gabriel Stricker, and Karen Wickre. All were champions of the scheme I presented to them.
The Googlers themselves could not have been more generous with their time and assistance. Though the list could literally go into the hundreds, I will dare to single out a few who took extraordinary pains to help me understand Google: Paul Buchheit, Matt Cutts, David Drummond, Urs Hölzle, Bradley Horowitz, Kai-Fu Lee, Salar Kamangar, Joe Kraus, Andrew McLaughlin, Marissa Mayer, Sundar Pichai, Andy Rubin, Amit Singhal, Hal Varian, and Susan Wojcicki. (Apologies in advance to others worthy of explicit mention.) I also benefited from the friendship and insights from my shadow network of compatriots from the APM trip that inspired the book. (On a foray to Baghdad, I forged similar bonds with Kannan Pashupathy and Hunter Walk.)
Google’s communications team grasped what I was trying to do and worked hard to make sure I had the access and information to do it. Krane and Stricker (and later, Jill Hazelbaker and Rachel Whetstone) were especially helpful in strategizing ways to get onto various schedules, particularly the founders’. Megan Quinn was a tolerant minder on the APM trip and a good friend thereafter. John Pinette orchestrated a very deep dive into Google China. Diana Adair and Nate Tyler took on the challenge of helping me peer inside the respective black boxes (though they both hate that term) of ads and search.
A very special thanks goes to Karen Wickre, my designated PR “shepherd” and the best guide to Google that anyone could imagine. She masterfully threaded a difficult needle as a consistent advocate for my project and a loyal representative of her employer. And she was wonderful company throughout.
My editors at Wired understood how the book project of their new employee would be not a distraction but a benefit to the magazine. So thanks to Chris Anderson, Bob Cohn, Thomas Goetz, Mark Horowitz, Jason Tanz, and Mark McCluskey. I am also grateful to my previous employer Newsweek and my editors there (particularly Mark Whitaker, George Hackett, and David Jefferson) for providing me with a platform to research Google’s early days in real time. I also appreciate Kathy Deveny’s sign-off on the expense statement for the APM trip.
A virtual participant in all the interviews was my master transcriber Victoria Wright, who must now be the most knowledgeable person about Google who never set foot on the campus. During the spring of 2009 I had research assistance from Andrew Marantz, under the auspices of New York University’s Literary Reporting mentorship program. Zach Gottlieb helped with research on Google.org. My friend Lynnea Johnson proved a lifesaver when she offered the Palo Alto cottage she co-owns with Carolyn Rose as my base camp for the project. The actual writing of the book accelerated because of a fantastic uncluttering of my office by Erin Rooney Doland. My fact-checking team included Deborah Branscum, Victoria Wright, Stacy Horn, Teresa Carpenter, and Andrew Levy. (Though, as always, the buck stops with the author.) I got wisdom and advice along the way from John Markoff, Kevin Kelly, and Brad Stone. My first and most enthusiastic reader, of course, was my wife, Teresa Carpenter. (Having a Pulitzer Prize winner in the house is pretty useful.)
As always, my agent Flip Brophy was invaluable at every stage of the perilous publishing process. At Simon & Schuster, Bob Bender was again my sharp-eyed editor, with Johanna Li assisting. The meticulous copyeditor at Simon & Schuster was Nancy Inglis. David Rosenthal believed from the beginning.
Every author relies most of all on the sacrifices and support of loved ones, and I’m no exception. My deepest love and gratitude to Teresa and Andrew.
Finally, this book—as with almost every piece of nonfiction journalism written in the twenty-first century—would have been immeasurably more difficult to produce without the Google search engine. Thanks to Larry, Sergey, and all of the engineers who produced and improved this technological and cultural marvel.
SOURCES
This book is based largely on a series of more than two hundred interviews with past and present Googlers, as well as a number of people who have interacted with Google and could shed light on its operations and practices. I have also drawn on my previous reporting on Google since 1999 as well as my reporting on the technology industry in general. I have had numerous interviews with Google principals, including several lengthy sessions with Sergey Brin, Larry Page, and Eric Schmidt that revealed their thinking at the time. I have drawn on my notes from those sessions when writing about major developments or controversies involving the company and its products. During my time researching the book (beginning in June 2008), I attended many meetings and events; most of the quotes in this book come from that reporting. Exceptions are cited in the notes below.
I also drew on the accounts of the company provided by other journalists, notably John Battelle, T
he Search (Portfolio, 2005), David Vise and Mark Malseed, The Google Story (Delacorte, 2005), Randall Stross, Planet Google (Free Press, 2008), Richard Brandt, Inside Larry and Sergey’s Brain (Portfolio, 2009), and Ken Auletta, Googled (Penguin, 2009). I also consulted the hundreds of articles in magazines, newspapers, and online sources.
Prologue
1 “Have you heard of Google?” I wrote about the APM trip in “Google Goes Globe-Trotting,” Newsweek, November 3, 2007.
2 “Google, the Net’s hottest search engine” Steven Levy, “Free PCs … for a Price,” Newsweek, February 22, 1999. It was an article about Bill Gross, contrasting his GoTo search engine’s prowess unfavorably to Google’s.
4 “We envision a world” The description is reprinted in a blog item by Dan Siroker, “What would you say you do here?” Siroker Brothers (blog), May 11, 2006.
Part One: The World According to Google
9 “There is just too much” Transcript of The Authors Guild, Inc., et al., v. Google Inc., 05 Civ. 8136, United States District Court, Southern District of New York, February 18, 2010.
10 “A major threat” Yasuhiro Saito, Japan’s PEN.
10 “An unjustified monopoly” Michael Guzman, representing AT&T.