Civil-military gap, knowledge gap, training gap, memory gap …. and the national security imperative of closing them.
Closing the gap between technology leaders and policy makers will require a radically different approach from the defense establishment.
DEC 13, 2018
Amy Zegart, Contributing editor at The Atlantic
Kevin Childs, Lieutenant Colonel in the United States Air Force
A silent divide is weakening America’s national security, and it has nothing to do with President Donald Trump or party polarization. It’s the growing gulf between the tech community in Silicon Valley and the policy-making community in Washington.
Beyond all the acrimonious headlines, Democrats and Republicans share a growing alarm over the return of great-power conflict. China and Russia are challenging American interests, alliances, and values—through territorial aggression; strong-arm tactics and unfair practices in global trade; cyber theft and information warfare; and massive military buildups in new weapons systems such as Russia’s “Satan 2” nuclear long-range missile, China’s autonomous weapons, and satellite-killing capabilities to destroy our communications and imagery systems in space. Since Trump took office, huge bipartisan majorities in Congress have passed tough sanctions against Russia, sweeping reforms to scrutinize and block Chinese investments in sensitive American technology industries, and record defense-budget increases. You know something’s big when senators like the liberal Ron Wyden and the conservative John Cornyn start agreeing.
In Washington, alarm bells are ringing. Here in Silicon Valley, not so much. “Ask people to finish the sentence, ‘China is a ____ of the United States,’” said the former National Economic Council chairman Keith Hennessey. “Policy makers from both parties are likely to answer with ‘competitor,’ ‘strategic rival,’ or even ‘adversary,’ while Silicon Valley leaders will probably tell you China is a ‘supplier,’ ‘investor,’ and especially ‘potential market.’”
In the past year, Google executives, citing ethical concerns, have canceled an artificial-intelligence project with the Pentagon and refused to even bid on the Defense Department’s Project jedi, a desperately needed $10 billion IT-improvement program. While stiff-arming Washington, Google has been embracing Beijing, helping the Chinese government develop a more effective censored search engine despite outcries from human-rights groups, American politicians, and, more recently, its own employees. Since the 2016 presidential election, Facebook executives have been apologizing to Congress in public while waging a campaign to deny, delay, and deflect regulation and stifle critics in private.
Former Secretary of Defense Ash Carter, Google’s Eric Schmidt, Amazon’s Jeff Bezos, LinkedIn’s Reid Hoffman, Code for America’s Jen Pahlka, and others have been working hard to bridge the divide, bringing technology innovation to Washington and a sense of national service to the tech industry. But their efforts are nowhere near enough. The rift is real, deep, and a long time coming, because it’s really three divides converging into one.
There is a yawning civil-military relations gap between the protectors and the protected. When World War II ended, veterans could be found in seven out of 10 homes on a typical neighborhood street. Today it’s two. Less than half a percent of the U.S. population serves on active duty. A senior executive from a major Silicon Valley firm recently told us that none of the company’s engineers had ever seen anyone from the military.
It should come as no surprise that when people live and work in separate universes, they tend to develop separate views. The civil-military gap helps explain why many in tech companies harbor deep ethical concerns about helping warfighters kill people and win wars, while many in the defense community harbor deep ethical concerns about what they view as the erosion of patriotism and national service in the tech industry. Each side is left wondering, How can anyone possibly think that way? Asked last week what he would tell engineers at companies like Google and Amazon, Chairman of the Joint Chiefs of Staff General Joseph Dunford said, “Hey, we’re the good guys … It’s inexplicable to me that we wouldn’t have a cooperative relationship with the private sector.”
There’s a training gap between leaders in Washington, who are mostly lawyers struggling to understand recent technological advances, and leaders in Silicon Valley, who are mostly engineers struggling to understand the age-old dynamics of international power politics. Congress has 222 lawyers but just eight engineers. On the Senate Armed Services Committee, it’s even more stark. Of its 25 members, 17 are lawyers and just one is an engineer. (He’s actually the only engineer in the entire Senate.) In the past, policy makers didn’t have to work that hard to understand the essence of breakthrough technologies like the telegraph, the automobile, and nuclear fission. Sure, technology moved faster than policy, but the lag was more manageable. Digital technologies are different, spreading quickly and widely on the internet, with societal effects that are hard to imagine and nearly impossible to contain. Understanding these technologies is far more challenging, and understanding them fast is essential to countering Russia and China.
At the same time, today’s brightest young engineers barely remember 9/11, view the Cold War as ancient history rather than lived experience, and can get computer-science degrees at elite institutions without ever taking a course about cybersecurity or thinking about what is in the national interest. For technologists, technology holds the promise of a brighter future, not the peril of dark possibilities. Their overriding challenge is getting a breakthrough to work, not imagining how it could be used by bad actors in nefarious ways.
The congressional hearings with the Facebook CEO Mark Zuckerberg on April 10 and 11 brought the two perspectives—and the chasm between them—into full view. For the tech community, it was a jaw-dropping moment that revealed just how little members of Congress know about the products and companies that are transforming global politics, commerce, and civil society. Senator Orrin Hatch appeared surprised to learn that Facebook earned the majority of its revenue through ad sales. “How do you sustain a business model in which users don’t pay for your service?” Hatch asked quizzically. “Senator, we run ads,” replied Zuckerberg, his aides grinning behind him. Senator Lindsey Graham asked whether Twitter was the same thing as Facebook. Even Senator Brian Schatz, considered one of Congress’s tech aficionados, didn’t seem to know the difference between social media, email, and encrypted text messaging. As Ash Carter wrote, “All I can say is that I wish members [of Congress] had been as poorly prepared to question me on war and peace in the scores of testimonies I gave as they were when asking Facebook about the public duties of tech companies.”
For the policy-making community, the hearings were a jaw-dropping moment showing just how much naïveté and profits were driving Facebook’s decisions, and just how little Zuckerberg and his team ever considered the possibility that all sorts of bad actors could use their platform in all sorts of very bad ways. In his opening statement, Zuckerberg admitted, “Facebook is an idealistic and optimistic company. For most of our existence, we focused on all of the good that connecting people can do.” Zuckerberg added, “But it’s clear now that we didn’t do enough to prevent these tools from being used for harm.”
The third divide is generational. In Washington, power runs vertically and rests in the hands of gray eminences. In Silicon Valley, power runs horizontally and rests in the hands of wunderkinds and their friends. Steve Jobs was 21 years old when he started Apple with his buddy Steve Wozniak. Bill Gates quit college his junior year to start Microsoft. Zuckerberg launched Facebook in his sophomore dorm room. Larry Page and Sergey Brin were old men, starting Google at the age of 25. In the policy world, 30 years of experience usually makes you powerful. In the technical world, 30 years of experience usually makes you obsolete. Policy makers who think college engineering students should be grateful for the opportunity to shadow them and photocopy during college summers have it all wrong. Interns on Capitol Hill answer phones. Interns at SpaceX launch rockets into orbit. For gray eminences silently lamenting in their Washington corner offices, “Who needs these whiny young Millennials?” the answer is: America does.
It’s hard to overstate just how foreign the worlds of Washington and Silicon Valley have become to each other. At the exact moment that great-power conflict is making a comeback and harnessing technology is the key to victory, Silicon Valley and Washington are experiencing a “policy makers are from Mars, tech leaders are from Venus” moment, with both sides unable to trust or understand each other. Even the dress codes are vexing and perplexing. In the tech industry, adults dress like college kids. Inside the Beltway, college kids dress like adults.
Closing this divide is a national-security imperative. And it requires thinking differently, generating inspiration rather than just regulation, and targeting the leaders of tomorrow, not just the leaders of today.
For starters, the Pentagon needs a messaging overhaul. Stop telling engineering students at top universities, “If you want to make money, go into industry, but if you want a mission bigger than yourself, work for me.” When Admiral Mike Rogers, who led the U.S. Cyber Command and the National Security Agency, gave this standard recruiting pitch to Stanford undergraduates a few years ago, it fell flat. It still does. We recently held a focus group of Stanford computer-science majors. When we tested the message on them, heads started shaking in a Wow, you just don’t get it kind of way. “One of the main reasons people pick companies is they want to do social good,” said Anna Mitchell, a senior. “People would laugh if the government said the only way to be impactful is to work in government.”
For these students and their peers, the desire for impact is real and deep. They believe that they can achieve large-scale change faster and better outside the government than within it. “A message suggesting a dichotomy of working in companies versus helping your country alienates a good portion of people on the fence,” Michael Karr, a Stanford junior, told us. “If you’re working on autonomous vehicles, you could be saving lives by making cars safer.” So what message does work? Giving them opportunities for impact at scale that don’t take a lifetime of moving up the ladder. Deploying the best young engineers against the toughest challenges, early. Telling them what Kevin tells potential recruits: If you do cyber operations for anyone else, you’ll get arrested. If you do them for me in the Air Force, you’ll get a medal.
The Pentagon also needs to create ambassadors, not lifers. More than getting technical experts into government for their entire careers, we need to get more national-security-minded engineers into tech companies. Winning hearts and minds in the tech world starts early, with new college graduates who are more open to new experiences that can last a lifetime. Imagine a Technology Fellows Program like the White House Fellows program, only younger. It would select the 50 most talented American engineering students graduating from college for a prestigious, one-year, high-impact stint in government service, working directly for senior leaders like the Air Force chief of staff, the secretary of defense, or the commander of U.S. forces in the Middle East.
Tech fellows would work on the most important projects and participate in special programs for their cohort to bond and form a lifelong network. “People really care about their cohort,” said Andrew Milich, a Stanford senior specializing in artificial intelligence. Tech fellows could defer company jobs or take a leave of absence, knowing that all the other fellows would be the best in the world who would also be heading back to industry. The goal isn’t for them to stay in government. The goal is for their government experience to stay with them. As one of our students told us, “Everyone has a friend at Google.” Imagine the ripple effects if these friend networks across the tech industry included tech-fellow alumni.
Doing it right won’t be easy. The Tech Fellows Program would have to be high on prestige and low on bureaucracy. Fellows would need flexibility to select projects that align with their values, not just their expertise. As the sophomore Gleb Shevchuk told us, “There has to be a transparent discussion of ethics. The program has to come off as a program that understands the concerns of people who dislike certain things the government is doing.” Google engineers may object to helping the Pentagon improve its targeting algorithms, but they might jump at the chance to defend U.S. satellites from attacks in space.
In addition, the program would have to reduce logistical pain points dramatically. Tech companies compete aggressively on quality-of-life dimensions for their workforce—locating in cities where top talent wants to live, providing free housing and transportation, and offering exciting programs outside of the job. The Tech Fellows Program would need to do the same. The National Security Agency has cutting-edge technological programs that would be a natural fit for tech fellows, but that’s a hard sell. The hot cities for attracting top engineers include Austin, Seattle, San Francisco, New York, and Denver—but not Fort Meade.
In the longer term, the Pentagon needs a radically new civilian talent model. Programs like the Air Force’s Kessel Run and the Defense Digital Service are breaking new ground to bring technology and tech talent into the Pentagon, but these programs are green shoots surrounded by red tape. Will Roper, the assistant secretary of the Air Force for acquisition, technology, and logistics, and someone who is no stranger to innovating inside the Defense Department, would like to see a much more fluid pathway in and out of industry and government. “I would invest to make the term revolving door superlative instead of pejorative,” he told a Georgetown class. “The people that we want are going to be people in industry that will want to come in and help us, and be able to go back out and come back in and help us, [so] that we’re continually refreshing the ideas, the creative thought … and right now we make it damn difficult to get in and out of government.”
These challenges are substantial, but small steps could have big impact over time. Congress could start by holding hearings with the goal of writing the best proposals into the National Defense Authorization Act this year. And if Congress doesn’t take action, then the Pentagon should, creating a Rapid Capabilities Office dedicated to developing new civilian talent programs, just like it has for developing new technologies.
In 1957, the launch of Sputnik spawned a fear that an underfunded education system had allowed the U.S. to lose its technological advantage to the Soviets. A year after the launch, Congress passed the National Defense Education Act, increasing funding for science, mathematics, and foreign-language education at all levels and allowing for substantially more low-cost student loans. Within a decade, the number of college students in the U.S. had more than doubled, supercharging U.S. breakthroughs in the space race. What our national leaders realized in 1957 is still true today: What people know and how they think are just as important to the nation’s defense as the weapon systems we deploy.