Join us this new year
for new conversations at the Commonwealth Club. Hello and good evening. My name is Quentin Hardy and I'm pleased
to be back at the Commonwealth Club to be the moderator for tonight's Commonwealth
Club program focused on Amy Zagat's. New book Spies, Lies and Algorithms. Amy is the senior fellow at the Freeman
supposedly Institute for International Studies at Stanford
University and the Maurice Arnold and not a Jean Cox
senior fellow at the Hoover Institution. Also at Stanford, her longtime
specialty is American Intelligence. And her book explores the growth
and the many challenges facing America's intelligence infrastructure today, particularly
because of changes in technology tonight. As you can see,
we continue in our virtual format. But the Commonwealth Club
is returning to more in-person programs at its San Francisco headquarters. You can learn more about upcoming programs
at the club's website at WW W Dot Commonwealth Club. One word dot org. Two quick notes before we jump in. Today's program. If you have a question for Professor
Zegart or me, please put it in the YouTube chat box. That way your questions will be forwarded
to me during the program. And additionally, though,
I work at Google, I am here independent of my job,
and nothing we cover tonight should be construed as either supported
or opposed by my employer. Alphabet or any related organizations. I believe Professor
Zegart has a similar relationship with her opinions and Stanford University,
Hoover Institution, etc., etc. And with that on tonight's program today, the surveillance world
has a severe problem and its abundance. There are hundreds of government
and private satellites circling the earth, tracking the movement
of all manner of human and natural events that matter to nations
and independent actors. We have billions of communications
a day on social media, email, shared documents, digital video
and much more. There are thousands of producers
of both real information and blatant propaganda and misinformation, and the roles of those actors
are almost never clear. A teenager in his parent's home
can track a billionaire's jet, while nuclear states
can act like Internet trolls. And North Korea finances itself with ransomware attacks
against private corporations. There is an explosion
of private surveillance from people tracking their own heartbeats, package
delivery and who's at the front door to companies with online cookies and retail loyalty cards,
tracking our reading and spending to firms monitoring society for the next big trend
and for all kinds of Real-Life Spy Companies for hire, private
and public both hackers and muscle not least, there's an enormous entertainment culture
around espionage. Another kind of abundance
where intelligence analysis and successful
violent action are completely intertwined Think of James Bond, the show
24 Homeland Movie Zero Dark 30. Just to name a few. So what does this mean for our country's
original surveillance industry? Now with some 18 agencies
strong with a very large budget. What are they supposed to do in this
supposed new paradise of surveillance? Well, the first thing they need to do
is to dispel the myths about the purpose they serve. Get to the realities and move on to how all of them can best
serve America's national interest. And this is the purpose of Amy Zagat's
new book. She wants to move past the lazy concepts
of American intelligence and spying that are shaped by books, movies and TV. In spies, lies and algorithms. She lays out the history
and purpose of U.S. intelligence, its increasingly important
role, its successes and failures and how it might cope
with today's challenges. We couldn't be meeting in a more
appropriate time when Russia and NATO's seemed headed for a fateful conflict
marked by photos of Russian soldiers on TikTok, of Ukrainian troops showing off
shoulder fired missiles on Twitter, Russian videos of tanks advancing
or maybe retreating And the US president openly discussing
what's in his intelligence brief. It seems like we're in new territory
and intelligence is at the center of it. So I guess I'll begin
right at the beginning of things. Amy Zegart,
welcome to the Commonwealth Club. Take us into the Ukraine. How would you describe what's going on
there from an intelligence point of view? Well, Quentin, let me just start by saying
thank you for taking the time to have this conversation. I want to thank everybody
joining the Commonwealth Club. I just wish we were doing this in person,
but I understand COVID restrictions make that hard,
but next time we'll do it in person. It's a great question to start off. You know, intelligence is everywhere
in the headlines. Right now. And perhaps nowhere cent more centrally
than in the current crisis with Ukraine. I think we're seeing something really new and we are seeing it
play out over the past several days where the Biden administration,
as you allude to, is really sharing intelligence
in a novel way. One colleague of mine described it
today as deterrence by disclosure. So in some ways, you can see an effort
to combat information warfare and Russia's false flag operations
by revealing Russian plans. Right. So among the disclosures
have been intelligence that Vladimir Putin was in the process
of planning a fake video with a pretext for invasion
of so-called Ukrainian atrocities with actors and corpses and things
that seem like they would just come right out of the movies. I think this is new territory
it's a new strategy. And I think it's it seems to be working. It's hard to tell right now
because there are still, of course, troops massing along the border. But it appears that it has surprised
Putin and his lieutenants. It appears
you hear more talk of diplomacy. And so at the very least, I think what we
have is an effort to inoculate the public to be prepared for these false narratives
so they don't take hold. So I'm encouraged that this is a new trend and a new strategy
by the Biden administration. It's very interesting because in some ways their objectives are objectives
of perception. They don't want to occupy Ukraine per se. They wish to hold a kind of dominance
over it that is perceptual. They don't want Ukraine joining Naito. They don't want a sense of another country
on their border, but they don't really want to occupy
the capital of Kiev, as best we can tell. Correct? I think we don't know. I think Putin may not know
his own intentions. I think for sure. One of the major strategic objectives
is to divide the Naito Alliance. And ironically,
I think so far that's proved the result has been the opposite
of what Putin intended. But, you know, I do think it's important to bear
in mind history, which is that, you know, active measures or these sort of Russian
false pretexts for invasions are nothing new. Right. The Russians
the Soviets did this in Hungary. In 1956. They did it in Czechoslovakia in 1968. But they're doing it at a speed and scale
that is really unprecedented. And I think that's why it's
so much more difficult to combat today. Right. And so who's spearheading
this rather unusual effort of full disclosure and transparency
on the U.S. side or what seems like full disclosure
and transparency? You know, the public reporting really
doesn't give us a clue to that. What we do, what we can see from public
reporting is that it's not just one type of intelligence or not
just intelligence about troop movements. It really is about,
I think, revealing information in my in my opinion, it does two things. Number one, inoculates
the public against these false narratives. So you see it coming right before the
Russians start concocting false stories. But number two, it's creating friction
within the Russian regime. Where is this intelligence coming from? How are we getting so much of it? How are we getting it so fast? And creating that friction
is part of putting Putin on his back heel. And that gives us more of an advantage
in this confrontation. Right. It's a little bit like the way
they shut down an Internet troll, we are being told by Russia,
and they are preventing them from putting out
more misinformation, perhaps yes. But I think, you know, the
I was just in a call with Stanford colleagues earlier today are experts
in Ukraine and and Naito and Russia. And they pointed out
something really important, which is one of the crucial things
that we can't lose sight of right now is we have to provide
diplomatic exit ramps for Putin. It's not enough just to have a showdown. It's not enough
just to reveal intelligence. We have to figure out a way to provide FACE-SAVING exits
to de-escalate this crisis. And so that diplomatic component is a key is a key
part of what we need to do moving ahead. That's very interesting. Historically, they said
that was the strongest thing the U.N. could do. The U.N. never got credit
for giving people an avenue out of words. You know,
you would just take it to the U.N. and discuss it
and put some blue helmet troops in there. And maybe that's a U.N. function or we need some new function to give people a face
saving way to de-escalate. Well, I think, Quentin, it's
so interesting you bring that up because the the U.N. dilemma is an intelligence dilemma
to which is that they don't get credit for averting bad things. You hear a lot about failures
when things go wrong. But a lot of intelligence successes
come from avoiding bad outcomes in the first place. And we usually don't hear about those. The dog that doesn't bark So our leaders are seemingly somewhat
prepared for this conflict. They may be moving
ahead with a certain level of uncertainty because everything's so novel,
but they're trying interesting things and they appear to be thinking ahead to
what happens next. How's the general public, do you think? The general public understands
that this is a different kind of conflict and understands what's at stake
and can pull together, as it were? You know, I wish I had really good data
to share with you to answer that question. Public opinion data going back
several years about what the public is understanding with respect to Russia. I just don't have that data. But I will say this, I think when we think
about all threats cyber related, we are nowhere near as prepared
as a nation that we as we need to be. And I think part of that is the sense
that cyber is distant, it's sterile. It may not affect me in the here and now. We can think about physical or kinetic conflict
we can imagine what that looks like. But it's harder to imagine
what a cyber bad event would look like. And I think that's part of the challenge of public education
about the threat landscape today. Well, I think that takes us somewhat
into your book as well, because this is very much, while highly
readable, volume and educational volume, it could be used to teach a course
to be a textbook. It could be used to educate responsible
citizens. I'd like to ask you what your goal was,
who your ideal audience is here, and how you began to write this
and where it took you. Well, Quentin, it warms my heart that
you think that it could be a textbook, because that's and it could be a general
audience book because those were my goals. So the book actually started
in the classroom. I was teaching a class at the time at UCLA
where I was on the faculty, and I was pulling my students
about where they got information about intelligence,
what they knew about intelligence and I found that most of them
didn't know anything about intelligence and what they learned
they learned from the movies. And I found statistically
significant correlations between those people who viewed a lot of spy
themed entertainment and their attitudes towards
intelligence issues of the day, like waterboarding
or NSA warrantless surveillance. And so what I really wanted to do
was write an intelligence one on one, a book that could reveal the facts
of the intelligence business for not only students, but their parents
and policymakers and the general public. And then what happened
was the world changed. And in the course, I said one of the
benefits of taking so long to write a book is, you know, things change,
and you end up changing the book. And so I started off wanting to write
intelligence one on one, and I ended up also writing a book that I hope
will convey intelligence to point zero. And what I mean by
that is how emerging technologies are fundamentally challenging every aspect
of the intelligence enterprise For many of the reasons of abundance
that you mentioned at the outset, a really wonderful description
of the abundance of organizations and data and complexity
that our intelligence agencies face today. So there's a big technology
component to the book, and I hope that will be of use not only to students
and not only to the general public, but also to policy makers who are in this
policy and intelligence base You spent ten years
you obviously are extremely well-connected inside the intelligence community
over those ten years. Did you witness a kind of growing
education, a growing awareness of the threat landscape
and the changes that have taken place? I think yes. And the threat landscape
changed dramatically. And just to give one example, you know, we talked a little bit
about the cyber threat. Well, the cyber threat
looks a lot different today than it did even ten years ago. So if we go
if we win back the clock and go to 2007, the word cyber wasn't mentioned
once in the intelligence annual threat assessment
that's publicly available and delivered to Congress every year. It's a really important document,
really important testimony. Cyber wasn't on the radar screen at all. That's 27 not that long ago. Fast forward five years
and the cyber threat was really imagined to be a cyber pearl harbor physical
effects of our critical infrastructure. And fast forward to 2017. And the cyber threat
is not just hacking machines it's hacking
minds it's Russian election interference. It's polarization of American society. It's information warfare. So I think there's an acute appreciation
in the intelligence community about the complicated threat landscape
and the velocity of the threat landscape, which is much different than it was
even ten years ago or 15 years ago. And this is an interesting dimension of your book that there is all this change at the same time You are an historian. You deal in history and you go deep
in the history of American intelligence. As we were talking beforehand,
you told me something reminded me of something
I really had passed over, that George Washington had not just a great belief in espionage and surveillance,
but an enormous budget, a budget proportionally larger
than the intelligence budget today. Talk a little bit about the growth
of intelligence and what it's meant. Do you see it as linear
and continuous over time, or have there been shifts that have changed
what it does inside American policy? So I confess that I became obsessed
with early American history and the role that intelligence played. So I spent quite some time delving into
George Washington and how he thought about and used intelligence. And what you find, if you look at the
broad brush of history, intelligence started off as playing a central role
in the founding of this country. Not just George Washington,
but Benjamin Franklin was very adept at information warfare. He was a printer in his day,
and he actually cranked out literal fake news from a printing press
and his parents in his Paris basement to sway opinion in favor of the rebels
and against the redcoats. So you see this early era
where intelligence plays a pivotal role in the revolution. And then what you find
is this period of stagnation that intelligence, much like our military,
was demobilized in times of peace. And we really didn't
see a significant peacetime intelligence capability until relatively recently
after World War Two with as we were talking before,
the National Security Act of 1947. So the CIA has only been around
since 1947. And so compared to other countries
like Russia and China, the United States is a relative newcomer
to the world of espionage. And so we've seen sort
of a dramatic increase in intelligence that corresponds
with America's role in the world. Another fascinating element of the birth
and the growth of the CIA is the kind of melding of intelligence gathering and analysis to advise policymakers and what should we call it, kinetic efforts. Street guys assassinations, but the projection of violence
by intelligence agencies. Do you think these are two things that had a place together or did they
somehow get lined up together? How did that come to be? Because today there is a military function
against intelligence as well. So this is a great question. You know, the Central Intelligence Agency
really has several different cultures and different missions
within one organization. So there's the collection of human
intelligence, right? Human sources. There's the all source analysis,
which you alluded to. And then there's the covert action piece, which includes,
you know, creating physical effects. So if you think of intelligence
is collecting for insight, you know, covert action
is about changing something in the world. And that's very different than the rest
of the mission of the agency. And so those missions have always lived in
some some degree of tension. But what we've seen is in the last
20 years with the global war on terror, that melding of intelligence
and supporting the warfighter on the battlefield
in counterterrorism operations, that's gotten much more important
and much more tightly integrated. And so what I argue in the book
and what I argue in a piece I wrote in Politico was that there's good news
and bad news of the past 20 years. The good news is intelligence and kinetic
or warfighting operations are much more tightly connected. The bad news is
they're much more tightly connected. Right. Why is it the same? Because the more time
the Central Intelligence Agency is spending hunting,
which is what militaries do, the less time it can spend gathering,
which is what intelligence agencies do. In a world where you can't tell those
two things apart, where the CIA stops and where the military begins is a world
where the CIA is not putting enough effort
into its primary mission, which is preventing strategic surprise,
like a Pearl Harbor, like a nine 11. That's the unique mission of the CIA. And so there's a cost and opportunity cost to this integration
of military operations. Yes. And one can one can have a bias towards action,
as they say, when you have a hammer in your hand,
everything looks like a nail. This is not particularly good
for contemplation. But on a personal level,
and I assume on an institutional level, because there's a lot of pressure to go
more than sometimes at the expense of the pressure to know Absolutely. And, you know, I've said this before. One of the shortly after nine 11,
a former senior intelligence official said something to me
that I always remembered. He said he worried that
by the time we mastered the Al Qaida problem,
would al Qaeda be the problem? And he was right. So we've hardwired our agencies
to do a very good job at counterterrorism. We saw this with the takedown of the ISIS
leader just a couple of weeks ago. But the threat landscape has changed
dramatically. We talked about cyber threats. We I'm sure we'll get to China. We've talked a little bit about Russia. It's a different threat environment today. And we need organizational change
in order to meet that threat environment. Let's speak about the organizations. Why do we have 18 different
intelligence agencies? It sounds like a big number
It's a very large number. And part of it is the political dynamic. So I'll be
I'll be a little bit more less flip and a little bit more serious to begin
with. And then I'll give you my flip answer, that intelligence agencies
are specialized. Some collect human intelligence. Some like the NSA, focus
on signals intelligence. And there's some benefit to having
specialization in intelligence missions. There's also a cost,
which is coordination. The somewhat flip answer
is a political answer, which is whenever there is an intelligence
failure, the natural inclination is to create a new agency
because creating new agencies look like you're doing something big,
even if it's not the best solution. So, you know, you see, after intelligence
failures, we get yet another new agency. And if you think that nine 11,
the key problem was coordination, which is what I found in my research,
creating more intelligence agencies to coordinate
may not be the best organizational reform. I give you the Patriot Act,
the greatest restructuring of government since the 1947 Intelligence Act
with probably even less debate about it less debate
and then in terms of restructuring. So that was really a lot
of new authorities. But in terms of restructuring
the Intelligence and Terrorism Reform Prevention Act of 2004,
creating the Director of National Intelligence
was a massive change as well. And that's evolved over time
and it's gotten better. But there's still serious
coordination problems that the Director of National Intelligence has for
many of the same reasons the CIA director had in the half century before that,
which is that if you're running, if you want to coordinate across 17
other agencies, you need two real levers of power
in Washington. Power
over the money and power over people. And Congress watered down
those authorities when it passed that legislation,
just like it did in the 1940s. Hobbling the ability of the DNI,
like the CIA director, to really knock
bureaucratic heads together. And let's talk about something that I bet will send
you spinning around the room these spy entertainment complex you have congressmen
who watched Jack Bauer and fully expect
we will extract the terrorist's confession under pain of torture
before the commercial break. So I have to acknowledge that I love spy
themed entertainment as much as the next person. So I don't want to be
too much of a Debbie Downer. I've I've watched by theme shows
just like everybody else. But my research found that spy
themed entertainment isn't just entertainment,
that it has real consequences for public opinion on the one hand
and policymaking on the other. I talked a little bit about my
my polls of my UCLA students while I did national polls about what Americans
thought about intelligence agencies, too. And what I found was at the height
of the Snowden controversy in 2013, when former NSA contractor Edward Snowden
revealed highly classified programs that many found disturbing conducted
by the National Security Agency at the height of that press coverage most Americans in my national survey
had no idea what the National Security Agency
even did for a living, so they didn't understand
the main mission. 75% of Americans
thought the NSA interrogated detainees. They don't. And so the upshot of that is that American the American public is forming opinions
on the basis of of either no information or misinformation
coming from the movies. On the policy side, I found a number
of examples of policymakers using Jack Bauer plot lines from 24 to reason
through real intelligence policy. You can't make this stuff up. The Senate Select
Committee on Intelligence actually asked in a confirmation
hearing of a CIA director about fictional ticking time bomb
scenarios from the entertainment business, which experts say are completely
unrealistic and have never happened. I found Supreme Court justices
using Jack Bauer two reasons through cases
that might come before the court. I found cadets at West Point
and the dean at West Point so concerned about the show. 24 and how it depicted torture
and how it always worked that he visited the scene
the the set of the show. 24 in Los Angeles to request that they made episode shows
where torture didn't work. And In A Truth is stranger
than fiction moment. He came wearing his uniform. The crew thought he
was an actor, not a general So this is you know, entertainment is
is not just entertainment. And you're right.
I could go spinning for a long time. Zero Dark 30 CIA director
had to write a memo to his own workforce dispelling the fiction
that Zero Dark 30 was perpetrating. So when the when the head of the agency has to debunk a movie
about the agency to the agency. Entertainment is not just entertainment. Well,
would you qualify it as misinformation? I think it could qualify as misinformation in the case of Zero Dark 30,
I think in particular, because the movie portrays itself as true now many spy themed entertainment movies. James Bond. No one's claiming that James Bond is
realistic, but the team behind Zero Dark but the team behind Zero Dark 34 minutes I hate to burst your bubble,
but God is not real. But the team behind Zero Dark 30,
if you remember when that movie came out, the opening
frame says based on firsthand accounts of actual events, not inspired by based
on firsthand accounts of actual events. And the writer and director marketed
that film as a first draft of history and a reported film. And it was not It was a fictional film. And so I think there's more responsibility
that Hollywood could take for making clear when we're talking about fact
and when we're talking about fiction. Well, there are politicians misinformed or congressmen playing to the home
audience however you want to take it. And then there is the body
of the working government, their staffs, for example,
the career professionals at these agencies they know the difference. And surely I think it's hard to know. You know, government employees
watch the same entertainment that we all watch And so I don't think
I would not assume that. And especially,
you know, what I find in my book is that it's one thing if there's a lot of spy themed entertainment
and there are also sources of facts, but spy themed entertainment is largely
all we have. Spy themed entertainment has become
adult education. So, you know, for example,
I looked at A.P. exams in U.S. government. Right. And U.S. history. I looked at A.P. exams over ten years and almost no
questions were asked about intelligence. I looked at syllabi and courses offered by the top
25 universities ranked by U.S. News and World Report. Almost none of them are very few of them,
not almost not very few of them offer courses on intelligence. I found more of the top 25. Quentin offered courses on the history
of rock and roll than U.S. intelligence,
which means I can joke to my students. They have a better chance of learning about you to the band
rather than you to the spy plane. So we have an education crisis,
I think, in this country that we don't have many sources
of real information about how these agencies operate, and they're pretty
secretive about what they do. And this is this is a real problem. And perhaps the transparency the U.S. is showing in its intelligence
findings now is a source of power and hope because there is too much information
for the agencies. There's also too much information
for the general public. And in some ways, this is why the misinformation of various bad actors or the less noxious misinformation of a spy themed show is so dangerous
because it creates a sense of certainty
or a desire for certainty and action. I think Russia enjoys misinformation because it sets the stage
for authoritarianism. People are tired of having
to think through all these things and having all this uncertainty
in their lives. So they choose what seems like strength
at the expense of freedom, potentially I don't know how you solve that problem
other than make people better at understanding uncertainty,
working through things, seeking authoritative facts, Do you have any clue for our audience? Yeah. I mean, I think we're living in a time
where there's so much misinformation and it's hard to get authoritative
expertize on a whole host of topics. And I think that's particularly vaccine
for intelligence agencies. I think that what we what we tend to see
is a public that swings between thinking these agencies are incompetent
and they're omnipotent, right? So they can see anything, do anything,
go anywhere, you know, track your phone calls
with your grandmother. On the one hand, and then how how could
they ever do anything right on the other? Right. And so and I think it's really important
for secret agencies in a democracy in particular
to be well understood, because if American voters
don't know the truth about what these agencies do and don't care
Members of Congress can't ensure that their weaknesses are fixed
and their excesses are prevented. So whether you care
more about civil liberties or intelligence efficacy, both of them actually are improved by having better
information, public information about the nature of this business
and what our agencies can and cannot do. You're an intelligence expert,
dangerously close to being optimistic. I I catastrophize knowledge,
but I catastrophize for a living. So I'm probably not
that optimistic on a day to day basis. Let's move to China What is the relationship with that like? How do in the intelligence we've spoken
we've spoken of Russia. Let's speak of China a little bit. So, you know, it's interesting,
the FBI director just in the past couple of weeks had a major speech
where he really singled out China and he talked about China
being the greatest threat to our freedom, to our intellectual property
of any country in the world. And he spoke at length
about the strategies that China is using, including, you know, buying and spying
and coercing Chinese nationals studying in the United States. So it was it was quite a forceful
statement about the China challenge, not just from an intelligence perspective,
but from a geopolitical perspective. And he said something else that I think
is important, which is that on in general, the FBI is opening a new China
related counterintelligence investigation every 12 hours
This is an extraordinarily active foreign intelligence service
that is going full bore to steal our intellectual property,
to recruit human assets, to gain decision advantage,
which is what intelligence is about. It is the most active foreign adversary
from an intelligence perspective, confronting the United States That's astonishing. Has there ever been a threat like that
before? You know, I don't know that there's data
about the FBI director doesn't usually talk about how many investigations
are ongoing problem. And it's a lot of probably a lot of
those are little pings. It's not like they're planning somebody at the NSA every 12 hours there. No, but what we can't
what is unprecedented and I think what's fair to say is
unprecedented is the scale of China's
intelligence activities. So if you think just about one
cyber attack in 2015 China hacked the Office of Personnel Management
it sounds like a very boring organization what they did you know a lot of times that's where your security clearances
are processed so 22 million records of people with the highest level
of clearances and others in the country. That's an intelligence bonanza for China
and it's an intelligence nightmare for the United States and that's
just one of China's massive cyber attacks. You have the goods on everyone, but if you track careers and processes, you also basically build a Facebook
of the intelligence community, who knows who and how they related
and who knew what and how power exists and what their medical problems might be
if they have financial problems, who their foreign friends are, so you can
get to them and possibly recruit them You hear about these things. You have friends in the business. Are there countermeasures? Are there things they can do,
any remediation? There are. And I should hasten to add that,
you know, I'm an outsider to this world. So I, I often say I'm like an anthropologist
who goes to the distant reaches of Washington, D.C., to observe this rare and secret
people called intelligence officers. So I don't know all the things that our government is doing, but yes,
I my understanding is they're very aware of the counterintelligence challenge
and taking steps to try to combat it. But it is something that is into, again, in terms of speed and scale,
I think really different than the counterintelligence
challenges of the past. You think about how easy it is to download
thousands of classified documents and compare that to the Cold War,
when it took years for traders to pull documents and take them out
in their pants and in garbage bags. One bag at a time
and put them under a bridge. Right. Which is what Robert Hanssen did
when he betrayed the FBI. So it's a big challenge now. And I know the government is certainly
aware of it and working hard to combat it. Now, I know you're just a simple country Stanford professor, but in the movie Rocky
they say you hang up with smart people, you get smart friends,
you hang out with stupid people, you get stupid friends,
you hang out with spy people. So you got a lot of spy friends. What are they like? So one of the I'm
so glad you asked that question, because one of the things
that I'm really proud of about this book is that I really worked hard
to try to have their voices come through. So I'm very critical
of the intelligence community. I've written a lot of critical things
about their failures, but I also wanted the reader to know what it's like
to work inside the intelligence community. So I interviewed a number of people
and I asked them questions like, When did you tell your kids
what you did for a living? And what did they do when they found out? What were your ethical challenges? What were your best and worst days
in your career? At the end of that safety? Internet go when you tell their kids,
and what's that like? So it depends. They have discretion about
when to tell their children. The reactions really vary. I was really struck
by one intelligence officer who spent two decades in the clandestine service,
so he worked in some dangerous places recruiting human assets, and he lived in
some of these places with his family. And he recounted to me a moment
where there was a credible death threat not just against him,
but against his children. And he had to tell his kids
not only what he did, but how to stay safe on the streets,
how to look for whether something's the same or whether it's changed,
whether there's something suspicious. And so I interviewed him
over a course of hours, and in between the two different times, and in between,
he decided to ask one of his sons what his son remembered
about this experience. And he related to me. Yeah. His son remembered that when he heard
what his dad did for a living, he thought it was pretty cool.
So that was one reaction. Other reactions have been very,
you know, kids have been very upset. I guess what really struck me in
these conversations was how much the intelligence officers
I spoke to from a number of different agencies
how much they really think about ethics, how much they lose sleep over the ethical quandaries and decisions
they have to make. This is not a mindless
Do whatever it takes. Go over the line community of people. They are very much whether to keep
a collection stream running or reveal the information to help a policymaker
and remove that ability to collect. That's an ethical decision. Right. How are we going to do? How are we going to weigh those tradeoffs? How do we think about whether
to put people in harm's way and what the value might be
if they're risking their lives? They think about these things every day. And so I hope those those stories and that really reality of their lives
really comes through. Well, I think you're making a really
interesting point, which is fundamentally your account and missile silos or you're wondering
who's up and who's down at the Politburo, but you're collecting these
through human agents or you're reading encrypted conversations
that you've decrypted. And you you know,
you're seeing people's lives and careers. You cannot do this kind of analysis without an emotional dimension
and a certain level. You know, a good one would have
a level of empathy for the enemy and an emotional capability
and an imagination to understand these motivations. That's got to be an enormous burden over time. These people do they compartmentalize, do
how do they? How do the best ones cope with this? You know, I think there are
a range of tools and tactics. I know one of the folks I interviewed
talked about he began mindfulness training inside the agency
to deal with the stress of the job. You know, one of the you know, you
mentioned you're dealing with the enemy. And how do you feel responsibility
for people that you're recruiting? I talked to one analyst
who was one of the crucial analysts in the Cold War really focused on Poland,
which was really would be ground zero if the Soviets and the Americans
ever really came to war. And he had the source and the source
feared that he was going to be blown. And he was an incredibly valuable source. And he was exfiltrated from Europe
and brought into the United States And what they did was he he had such great
insight into the thinking of, you know, Soviet bloc officials, military officials
that they kept him involved And so this analyst, whose name is Eric
Pappas, he's since retired when he describes
when he met the source at the safe house for the first time in person and how
he says, oh, my God, it's this man. And he came with his family
and they called him George. And he said, you know, I never briefed
my wife and I never briefed my kids, but they read the news.
They knew this guy had defected. And why was I hanging out with a 50 year
old Polish military officer? And I was a 30 something year old analyst. And George would come for dinner
and George would go on boats. And nobody talked about George. Right. So they knew his family knew,
even though he didn't tell them anything, how important this man was and how he gave
everything to help the United States. So a great analyst has an emotional capability bears
an emotional burden. But increasingly, we're dealing with an enormous amount
of digital information, an enormous amount of data
and there is a lot of pressure to use AI, be that in drones,
be that in going through collection. What do you see as the role of a computer
driven intelligence service? How will that seem to these analysts? How can we have the best possible outcome using A.I.? So I think A.I. is going to be essential for analysis
moving forward. And I was part of a CSIS task
force of the Center for Strategic and International Studies did a deep dove
with our task force looking at A.I. and Intelligence There are certain things that machines
do much better and faster than humans, and there are certain things that humans
do much better and faster than machines. And so the two have to go hand in hand. So AI is so good at pattern recognition. If the data that you have from the past
looks like the present and the future,
I can really save a lot of time. So identifying surface to air missile
sites over a large swath of territory in China In fact, the National Geospatial
Intelligence Agency commissioned a team, a university team,
to create an algorithm to do that. And the algorithm
had the same level of accuracy as the human team,
but did the job 80 times faster. So I can really free up time for analysts
to focus on the things that humans do better than machines,
which is bringing creativity to problems, trying to divine the intentions
of adversaries, things where the past may not be and big data
may not give you a good guide to the future
if you're in or your sample size is one. What's in the mind of Kim Jong un or
what's in the mind of Vladimir Putin? So I think A.I. is essential, and the intelligence
community needs to move faster to adopt these technological tools and is an argument out in the Valley it's not A.I.,
it's IEA intelligence augmentation. These turn the machines over to things. Machines are good at but ultimately the imaginative, emotional taste, if you will, capability rests
more on the human side. And the two together are stronger
than either on their own I have a question from the audience. It says, How concerned
are you by the use of foreign like intelligence domestically Perhaps that means the sort of things the NSA is assigned
to do to foreign countries, turning those capabilities
on the domestic population. Well, I think we always have
to be vigilant about oversight. So our intel of the 18 agencies
we've talked about, there's only one that has clear
marching orders to collect intelligence domestically,
and that's the FBI. And the FBI operates within the Justice
Department for a reason. Right. More oversight,
more statutory requirements, more oversight by the judiciary,
more oversight by the Congress. The other intelligence agencies
are supposed to and by law, should train their collection
abroad. Now, there are some gaps, right. And there are some weaknesses
to doing that. And we saw some of those weaknesses
before nine 11 when terrorists communicate between the United States
and foreign countries. Right. That was a gap in intelligence. So it's an important issue
how to make sure that our intelligence
agencies are effective, but also that we're being vigilant
about protecting civil liberties at a time when information is moving
seamlessly across borders. That's a challenging task. Another question, how closely to social media companies
work with intelligence agencies it depends. So it depends on the time. It depends on the topic. It depends on the company,
depends on where they're headquartered. Depends on where they're headquartered. It depends on what
the leaders of those companies believe. So some companies are much more
forthcoming about partnering with the U.S. government than others. I would say as a general matter,
the trust between social media companies in particular
and the U.S. government was very bad
after Edward Snowden. So in sort of 20, 13, 2014,
there was tremendous distrust between those companies
and the government. I think it's gotten much better
since then. And one of the reasons it's gotten a lot
better is the China threat. The China threat isn't just to American
national security, it's to American economic prosperity. And it also affects the intellectual
property of American companies. And so I think you see, you know, the government has worked hard at this to try to build bridges
and better work with the private sector. But I often say that, you know,
the part of the challenge is language. So Defense Department officials
love to use the D word destroy, defeat, degrade. Right. And I think folks in the valley
I don't know if you think this, Quentin, but I think folks in the Valley like to
use see words create, collaborate, change. And so there's a
there's an important coming together culturally that needs to happen,
not just in terms of incentives, but in terms of how,
as I call them, suits and hoodies. Think about the world. And I think we're making progress
on that front. I again, I do not speak for Google, but I do think we are in a period where previously comfortably siloed definitions don't work very well when someone at a university in Beijing is doing a scam involving the Russian mob that is getting intelligence
information via a server in Nigeria. And the target is the U.S.. Is that a crime? Is that an act of war? Is that intelligence gathering? Is that all of the above? You know, where do we bucket that one? And there's a lot of that
in the digital world. That's a somewhat extreme example. But probably happened somewhere this week. Right. And so we need a sense of definitions, guardrails, as behaviors
that are appropriate for a world where we're still finding our way through
how some of this works. I think one of the key definitional issues for Google
and other companies is what's a weapon? Because
from my perspective, if we're in a world where information is so critical and it is
data is essential for national power, national prosperity, tech companies
aren't just victims, they're vectors. They can be weapons. They're being weaponized
by our adversaries. And so I think it's really important for
tech company leaders to acknowledge that that they are in the crosshairs,
whether they like it or not. And they have to think
about those responsibilities, whether they like it or not. They are policymakers
that have power akin to many governments. And that's a new role
for the U.S. government. To have to work with the private sector. And it's a new role for the private sector to have to assume or think
about responsibilities to the nation. Before we started,
we talked about the lessons of history And you and I both know
that multinational oil companies and multinational communications
companies worked closely with those. Those aren't necessarily new problems
or new opportunities. They the players are new, but the problems are timeworn. Shall we say. I don't really know how to take the everything is a weapon thing
because everything's always been a weapon. I don't think so. I everything is always going for energy. I don't think you know, I don't think
everything has always been a weapon. I think if you think about
what are attacks that could cripple our financial sector,
that could create widespread physical damage,
that could bring our country to its knees, those attacks are through networks and those networks are owned
and operated by private sector companies. I think that's fundamentally different. No doubt they take extremely seriously. Yeah, and that is documented. I don't think they need the government
to be brought in for that. But let's set that aside another question. What can be done about political blowback
with media and politics? Is outing agents and secrets. What happens when there is a change
in political parties and policy that says that? Is that a problem? I think I think there have been increasing concerns
because of the last administration. And I really think it's important
for intelligence to be nonpartisan, to be professional. If people think that intelligence has a dog in the fight of a
of a particular political point of view. That's bad for any future president. And what we saw, I think, in the Trump
administration was a real politicization of intelligence. You know,
he called intelligence officials Nazis. He thought he said he believe
Vladimir Putin more than his own agencies. He accused intelligence agencies
of spying on him illegally. Which has turned out
not to be the case, of course. And so that was damaging. No question that was damaging. And I think what we're seeing
is more of a return to how things used to be with intelligence
You see, you know, you mentioned earlier more transparency
from our intelligence community and a return to that apolitical professional function just as the military
is apolitical and professional Funny to say I was a journalist for many decades, but I'm getting weary of cynicism
about all these things. I think for the most part, these are good
actors trying to do their best, particularly career professionals in these in these agencies. Another question, to what degree is Huawei
an extension of the Chinese government? I'm not qualified to touch that. How about you? Well, my own view is when you have a state
authoritarian model, every Chinese company is to some extent
an extension of the Chinese government. There is no separate private sector
like there is in the United States. And I think industries in those critical
oil sectors, critical technologies are far more tied to the Chinese Communist
Party than we might think they are. And sometimes those ties are opaque. Right. So it's. Are there party officials
that have a board seats or are there what are the requirements? And, of course, by Chinese law, companies have to turn over
information to the government if it's requested. So even if they're not populated
by government officials and even if they're not working
at the behest of the government, there's tremendous coercive power
that the Chinese Communist Party has,
and everyone knows that coercive power. So they're not free in the way
that companies in this country are free to do what they want.
It's fundamentally different there. How is jurisdiction determined
for international acts of espionage? Directed at the U.S.? Well, international acts of espionage are actually not
against international law. So it's one of those. It's not illegal. It's never been against
international law to spy. So all countries, spy allies spy on each other, enemies
spy on each other, and they're supposed to obey their own
country's laws. So American intelligence officials
have to follow American law, but they are free to violate foreign law
where they're operating. And our allies and adversaries
do the same. So it's not. It may come as a surprise,
but there's no real international law against espionage. We're heading towards the top of the hour. And I don't want you
to talk a little bit about the challenges and the ways forward. And one thing you talk about a lot
is the five more years. Describe that for the audience,
because I think it's a really good jumping off place for talking about
the problems that and how to come out So I love your your opening
where you talked about the abundance. And that is certainly a similar way to how I think
about the challenges in intelligence. So as we think about this moment
of converging technologies, we think about the Internet, social media,
we think about satellites. You mentioned the thousands
of satellites in space. We think about air and quantum computing that we have this convergence
of a number of emerging technologies. And together they're creating these
five wars, the challenge for intelligence agencies. The first more is more threats,
more threats that can reach across cyberspace and threaten us
from the comfort of their own laptop. We've never had that before. You know, in physical space,
we have two oceans that protect us from bad neighborhoods. But there are no good neighborhoods
in cyberspace. So we have more threats. The second more is more speed. These threats are moving faster,
and intelligence has to move at the speed of relevance for policymakers
and so that's accelerating. Think about the crisis in Ukraine
playing out right now, how fast that intelligence
needs to move some more speed. The third, more and Clinton, you've talked
about this so well, more data, right? We're drowning in data. And how do we make sense of
and get insight from data today? The fourth more. More customers. So people at Google need intelligence
from the intelligence community about the cyber threat
landscape, for example, and vice versa. I'm sure Google engineers have a lot
they could share with the U.S. government,
but voters need intelligence today about foreign election interference. So intelligence agencies have to produce for people
who don't have security clearances. And that's a big change from the past where they've only produced
for the classified world less lest our viewers think
that just pertains to tech companies. You mentioned financial services,
power companies, pretty much anyone who is on the Internet
which would be anyone. So think Colonial Pipeline
or any kind of critical infrastructure power, water, transportation,
financial services, is given
cyber threats to critical infrastructure. Those leaders need intelligence to. So that's so intelligence
agencies have to do the unnatural act which is producing
for an unclassified customer. So that's the fourth
more is more customers and then the fifth. And perhaps
one of the more interesting things that I that I learned about in the course
of writing the book, more competitors because of new technologies. Spying isn't just for governments anymore. And so I have a chapter on citizen
detective nuclear sleuths and what everyday people
and organizations are doing to track nuclear threats around the world
using commercially available imagery and Open-Source
or publicly available intelligence. And they're making some remarkable
discoveries. I mean, their risks to this world, too,
and I get into those. But the upshot for the U.S. intelligence agencies is governments
don't dominate the collection and analysis of intelligence like
they have been like they did in the past. This is a much more democratizing
world of espionage. Last night on Twitter, I saw a map of the flight path of a drone over the Ukraine or maybe I did. You know,
I saw what was purported to be right. Let's not be too certain. That wasn't misinformation. Small snapshot of what this world is like they understand the problems
you're outlining. These aren't stupid people.
These are intelligence people. And they know these five more,
as I'm sure they agree with you to questions, how well-prepared are they? They see the challenges. Do they understand actions? And lastly, does history provide us with any guidance? Two big questions. So I think you have 3 minutes
the world in 3 minutes. Thanks for that. I'll do my best. So, you know,
I think a number of people, senior people in the intelligence community are very well aware of these challenges
and working hard to deal with them. The CIA director, Director Burns, has announced
new initiatives to focus on technology, to focus on China, to bring better talent
in the door faster, for example. The director of national intelligence issued similar calls,
including calls for more transparency. I think they're quite well aware
it's a hard challenge. The most powerful interest group
in Washington is the status quo. It's hard to get organizations to change
even when you have things in the private sector,
like the ability to hire and fire, the ability
to measure success and failure. These things are much more challenging in intelligence than they are
in any other kind of business. And so there's a real challenge
to adopting technology, in particular inside
the intelligence community. You know, in the Cold War, the government invented technology
and then it became commercialized. Think about GPS satellites invented by the government,
then commercialized the Internet, invented by a government agency,
the ARPANET, and then commercialized. But now it's the opposite. Innovations are invented in the private sector, and the government
has to figure out ways to bring them in. And that's really hard
for a whole host of reasons. So I think there's an awareness
of the need to change. But getting from here to
there is exceptionally difficult. Your second question, does history
provide a guide? I think it does. I think it suggests to us
that the cost of adaptation failure can be extraordinarily high. The lesson from Pearl Harbor
was that intelligence agencies collected the signals
but didn't put them together in time. We had information,
but we didn't use it well enough and incoherently enough
to see the signals through the noise. And then we see nine 11, right? Nine 11. Similar coordination problems
that we had in my research. 23 opportunities that the CIA and the FBI
had to penetrate that plot. And they missed every single one. And they adapted after failure. But they didn't adapt fast
enough before failure. And there were lots of indicators
in the decade before nine 11. That smart people saw this threat coming. Smart people at the top of the intelligence community
tried to get change and failed. And so I think this is another adapt
or fail moment with this convergence of technology
and the rise of China. And I think if the intelligence community
can learn the lesson from history is we have to get better at adapting before
disaster strikes rather than reforming after disaster strikes. And in that vein, you had a wonderful quote
in this book of someone pointing out that when you spend $1,000,000,000
finding out a secret, by definition,
that secret is worth $1,000,000,000. That's one of my favorite quotes
in the book that someone said to me, it's absolute genius, because that's not just a government thing. Any bureaucracy tends towards
that kind of observation. This is what I took to said it's valuable. And with that in mind,
you have a very interesting if mind melting idea. Let's create a 19th intelligence agency. You wouldn't think that was the plan,
but your point is a 19th agency
devoted to Open-Source information and in a sense, transparency to take from
what the what the U.S. appears to be doing in Ukraine
against Russia and times to just find open source stuff. Because perhaps
these institutional cultures of secrecy aren't capable
of dealing with public information the way they should Say
a little more about that. Yeah. So, you know, secret agencies
are always going to favor secrets. That's what they're designed to do. It's what they value. It's the business they're in. Now, there's nothing I know of any damage
here, and they're bad at sharing. There's nothing wrong with that. You want them to be focused on secrets
but there's a reason that it's an unnatural act
to embrace Open-Source Intelligence. It's not the priority of these agencies. And the analogy I give is, you know,
air power didn't get the resources and attention that it needed until
the Air Force was separated from the Army. So they used to be Air
Force used to be part of the Army. And there's a reason
it was separated after World War Two. I think open is is the same thing. I think Open-Source Intelligence,
what we can glean from this abundance of capabilities
that you've alluded to is the name of the game
for insight in the future. And it's not just about stuff. It's about an ecosystem of organizations. That are developing tradecraft,
that are developing norms, that are developing
training, better or worse. And so our intelligence community,
I think, needs a dedicated open source agency,
not only to bring stuff in, but to engage with this ecosystem of players so that it
can shape the dynamics to its advantage. We don't want more misinformation,
more deception to creep into our intelligence processes
through open source. We want to be able to raise the game
of Open-Source collectors and analysts so that they can be a positive force
for intelligence in the world and not part of what adversaries do. Well, you've just outlined an overlap of rationality, optimism and plausibility. So I'm going to leave the conversation
right there. What a great note to end on. I could go on much, much longer. It's a fascinating talk. Thank you so much, Amy Zegart. That brings us to the end of today's
program. On a positive note, which is great to know and I do highly personal plug
recommend the book. I'm sorry
we can't have a book signing tonight. But your next book, I'd like to encourage
everyone to purchase the book Spies, Lies
and Algorithms Wherever Books Are Sold. And with that, tonight's meeting
of the Commonwealth Club is adjourned. I'm Quentin Hardy. Thank you very much.
