Original subtitles

Tuesday, 9:47 a.m.

Hi, baby. I'm... baby, you have to listen to me carefully.

I'm on a plane that's been hijacked. I'm on the plane.

I'm calling from the plane. I want to tell you I love you.

Please tell my children that I love them very much.

And I'm so sorry, babe.

I don't know what to say. There's three guys.

They've hijacked the plane. I'm trying to be calm.

We're turned around, and I've heard that there's planes

that have been flown into the world trade center.

I hope to be able to see your face again, baby.

I love you. Good-bye.

End of message.

Well, the images I saw, the ones that, of course, uh,

affected me the most were the people diving off the building when it was burning.

You know, to be put in such a desperate state, you know,

that shows you the human...

human reaction to those kinds of conditions.

It just, uh, it was just revolting.

And disgusting that we allowed it, basically, to happen.

ThinThread was a program

that absolutely would have prevented 9/11.

So it all went back to that program.

The first time I met Bill Binney,

the first thing he said to me was,

"I need you to know that I would never deliberately commit suicide."

I thought he was being a little overdramatic,

but then when I started to hear what he had to say

about what our country was doing,

I worried about his safety, too.

I wanted to make one thing very certain up front,

that she could be assured that I would not commit suicide.

And I said that because I knew that some of the people involved in this activity,

and it was a pretty serious thing

because people at the highest levels

of the US government were involved

in all departments.

So I wanted to make sure that she understood

that if something happened to me, it was not suicide

because I was going to be in this with her

and everybody else to the end.

I grew up in central western Pennsylvania.

It was basically the wilderness of Pennsylvania.

It was out in the hills, in the Appalachia hills

just before the mountains.

This is my father's grave.

At one time, I asked him, I said, uh,

"in your service in war time, how many men did you kill?"

And he said, "maybe one."

So, but that's all he said.

He didn't really talk about it.

I think that was pretty typical of World War II veterans.

They did what they had to do and they didn't really

think that that was anything different than anyone else did in World War II.

Nobody thought they were heroes. They just thought they did their duty.

One of the worst things about war is those who start it

force others to be as vicious as they are to stop it.

That's what I'm sure he was feeling.

He just didn't want to talk about having to do

what he had to do when in the war.

One of the main reasons we get into war is by mistakes by leadership

because they're informed incorrectly

or they're uniformed about events, you know, totally.

So the principle I thought that I was working under

was to make sure that the information

that I could arrive at through the work I did

was informative to leadership so they didn't make mistakes.

Watch out, watch out.

Recapping now for you once again if you've just joined us,

an explosion underground in the garage section

of the World Trade Center in New York City

has killed now three people according to the New York Port Authority

and has injured at least 150 people.

Hundreds of people as we speak are still being evacuated

from the twin story towers of the structure

at the southern tip of Manhattan Island.

There was no hope whatsoever.

Trapped on 94 floors all the way up.

The walls just blew in and everything went blank after that.

Walls collapsed, furniture's down.

It's a mess downstairs.

The first attack on the Trade Centers in '93,

that was the first thing that woke everybody up

in terms of the threat of terrorism.

They took truck bombs into the basement

and tried to place the bombs next to the supports for the building

and set them off to try to shake the building

or break the supports so the building would fall.

Technically, that didn't function that way,

but it killed a lot of people,

and CIA really did start pounding the drums

across the entire intelligence community to go after the terrorism,

and that's what we did at NSA.

We were looking also at the same time

with the explosion in the digital age,

ballooning of communications in the internet and the phones,

and the terrorists were burying themselves inside

that ballooning communications network worldwide.

So we had to face that as a new technical problem

that we had to solve to figure out

who was communicating with whom and who were the terrorists

and try to extract that information out of that flow of information.

When I first took on the duty of technical director

of the world geopolitical and military analysis and reporting shop,

I pulled together analysts from all over the agency around the world...

And I said, "okay, we're going to start this little project here,

and here's what we're gonna do.

We're going to graph relationships

of everybody in the phone network in the world.

All we have to do is map all the relationships

between two and a half billion phones."

And somebody in the back says, "you can't do that."

And I said, "why not?"

And they said, "because all the combinations are infinite."

So I said, "okay, think of it this way.

At any instantaneous point in time

the numbers of atoms in the universe are finite.

The relationship to 2.5 billion phones

is simply a smaller subset of that.

That's nothing.

Now you can get your mind around it.

That was the whole idea of the ThinThread program.

Of all the analysts I ever knew

in my entire career, spanning from US air force

through my career at NSA,

Bill was by far the most astute,

the most capable, accurate.

He used all the tools necessary,

not only to find the story,

but to complete it.

Where we weren't able to actually read messages,

they were encrypted,

Bill found ways to leverage indicators

that I'd never seen anyone else do.

Bill was Mr. Externals at the NSA.

He was the person in the agency

who knew the most about what we now call

in the digital age, metadata.

If something was encrypted and you couldn't decrypt it,

you could still get information out of the external signals,

and that was Bill's forte.

Bill would find ways around the encryption

to see things that people didn't even look for...

Able to work almost any problem,

almost any challenge,

and draw upon mathematics

and other inherent analytical skills

and the willingness to be open-minded,

not close-minded.

He had a good B.S. detector

and he could tell when somebody was blowing smoke.

And he was the agency's

best traffic analyst, bar none.

In '65, I was coming out of school

and the Vietnam war was cranking up.

That was when we still had the draft,

and so people were getting drafted out of my community.

And since in the country,

you're usually pretty good with a rifle,

and if you go into the army, you end up in the front lines

doing sweeps through the jungles in Vietnam

and shooting people...

I simply did not want to participate in any of that at all.

I even stopped hunting because I didn't like to kill animals.

So why would I want to get involved with Vietnam?

I had the option to volunteer for the army to go to Europe.

They gave me a whole battery of tests,

and they said, "gee, you have a fairly high aptitude for analysis."

So they said, "hey, you're a prime candidate

for the army's security agency,"

and they sent me to Turkey.

We had just come through the Cuban Missile Crisis in 1962,

which was probably the closest time

we ever came to nuclear war,

and that kind of sensitized everybody to the real threat.

Turkey was looking at the southern part of the Soviet Union.

They had a lot of areas where they developed military technology.

I was working a rather large military data system,

a Soviet data system, that had not been solved at that point.

So they would look at things like manual Morse

or speech broadcasts or satellite communication.

Those kinds of things produced information

which they called traffic.

Some of it you would record, some of it you would either get directly translated

or converted into character streams.

Some of it would be continuously encrypted data,

which you wouldn't be able to read,

but you could look at durations and things like that.

It was probably a 20th of the Russian military,

so I was looking at a small segment of it

trying to develop the techniques that were necessary

to see into what they were doing.

It was me sitting at the desk, looking at this abstract data,

struggling to understand what it really meant.

As I was going through there, it became clear to me

that there were mathematical patterns involved.

I started to see systems and relationships,

so I started building things.

It was allowing me then to begin to understand and interpret

the things that we were getting in terms of communications...

And relating that then to real world events

and what was really happening in the field.

When I got to that point, the whole thing started to fall in place

and so then you can reconstruct the entire system,

their entire system of command and control

and everything they were doing.

Everything is human behavior.

Human behavior is extremely patterned.

How do people operate? How do they interact?

And how does that manifest or appear

in this abstract set of communications you're looking at?

It's a matter of getting to a point where you can understand the patterns

and interpret them properly.

When you sit there and look at a problem like that

and you see... you can see the pattern and the break,

that's exhilarating when you do.

It really is.

Which is basically what we call metadata analysis now.

Nobody seemed to be doing that.

So that ended up NSA requesting me to be reassigned

to NSA headquarter.

I got a letter from what we termed "God" in February of '67.

Metadata is actually just data about the data.

And as a traffic analyst,

you don't care about what's in the conversation.

You care about who's talking to whom,

how often are they talking, when are they talking,

what's their pattern of talking,

who are they talking down to,

who are they talking up to,

what's the subordination,

the command and control structure of the organization?

In May of '68, I got transferred

into the current reporting shop.

In the early summer, the Russians started to train

or having exercise along the border of Czechoslovakia.

I started looking at the communications

they were using to move around.

And that's where I started to pick out

some very small number of unique things

that was different from a normal training program.

So I started to capture that.

And I said, "it's obvious that they're going to invade."

It was only two days later that they actually invaded.

That was the first occurrence that I had

direct personal experience with a real world activity

that let me see what was different between that and the training.

That gave me the idea of what were the real warning indicators of impending action.

The real warning indicators, there were only five.

Five, that's all. Just five warning indicators.

I mean, they had maybe a hundred warning indicators on the list,

but none of them meant anything.

The five that were real weren't even on the list.

Okay, so I had my five

and the rest of them were meaningless to me.

So I simply watched for those five.

And it was just case after case over time, over decades,

you could see every time they were going to do something.

One of these or more of these,

one or two of these would show up at least.

When the Yom Kippur War came along in '73,

again, those warning indicators started showing up.

And then again in '79 with Afghanistan.

That was about a week before the invasion.

It was very close to the Christmas holiday.

So at that time, I said, "well, they're gonna

go in at close to midnight on Christmas Eve

because that's when everybody's gonna be asleep in the free world."

And that's exactly what they did.

I was off by an hour.

I couldn't understand why a lot of people couldn't see what I could see.

I didn't look at that understanding

as anything more than just a greater understanding

of people and how they operate.

I sat on the terrorism desk

in the Pentagon in 1993.

I was on that desk when they attempted to drop

the World Trade Center towers the first time with truck bombs.

That's when I first learned about Osama bin Laden.

I distinctly remember issuing reports

in which we said this is a serious matter.

This is not a one-time event.

There is a larger movement here,

and they're planning on doing more.

Al-Qaeda, the base,

and Osama bin Laden as their leader

weren't just issuing pronouncements,

they were actually calling for action.

The senior military officer, the J2,

reports to the joint chiefs of staff,

comes down to our alert center

in the National Military Joint Intelligence Center,

and he says, "yeah, I'm reading your reports.

Who cares about some raghead spouting off fatwas

in the desert underneath a fig tree?

Who cares?

Not a threat."

The National Security Agency traditionally has been

one of the most productive of US intelligence agencies.

We relied very heavily on it ever since World War II at least.

When I got the National Security Agency account,

I had heard that they were in pretty good shape.

I went for intensive rounds of briefings,

and after about three months,

I was shocked at the poor condition of the agency.

We were very ill-prepared for the digital age.

But National Security Agency management continued to think

that they were the best in the world

and had the best technology

because this had always been their reputation

and they were convinced

that there were no fundamental problems.

I was convinced that there were fundamental problems,

both technical and especially cultural,

and I became very worried very quickly.

See, they were drowning in garbage.

The processes that were in place before ThinThread came along

were wild carded garbage collectors,

and they were dumping all the stuff

into a huge storage system

and the analysts were floundering trying to find out,

well, you know, "what's important to me?"

ThinThread was our way of trying to address the digital age.

It was a question of being able to manage the volume of data.

The terrorists are using all the cell phones,

all the land line phones, all of the internet connections to communicate.

So the new Cold War, if you will,

was the worldwide explosion of the digital age.

The NSA took the different perspective that says,

"let's just collect everything

and then we'll have our analysts figure it out later."

The problem is, once they collected everything,

your analysts couldn't figure it out.

There was too much. They couldn't get through it every day.

They'd make their database pulls and they've

got hundreds of thousands of returns.

Well, you could not get through it.

You're burying your analysts with meaningless data.

That's when I started looking at what we had to do

to solve the problem of the volume of data

and how could we isolate things that were important for our analysts to look at.

And the key was metadata.

Just using the metadata in terms of relationships across the network.

So you don't have to look internally at the content.

You look more at the relationships

of the people communicating

or organizations communicating in the world.

So that's a way you can see into large amounts of data fairly quickly,

effectively, and pull out data that's relevant

to targets you want to analyze.

My biggest fear throughout all of my work at NSA

was the fear of human inaccuracy and human failure

in terms of interpreting things properly

or in terms of recognizing things.

I felt that that process needed to be fully automated

as much as possible so that the information coming in

was untouched by human hands.

Oops.

Many people called me things like "the great Satan,"

or, you know, "the evil one," I think,

and many other names, you know.

So I had whole groups of people that didn't want to talk to me

or be involved with me

because they felt if they did, they would lose their jobs.

I was sitting at my desk with pencil and paper

trying to build a structure of the Soviet military

command and control system, of all the forces.

In order to do it, I needed a little help from the cryptologists.

So I went to them to get the bits that they had in possession.

And they came back with a miniscule answer,

and I said, "if you don't want to work this problem,

just give me the bits and I'll do it."

I couldn't wait for them.

For a nuclear intercontinental ballistic missile to be fired,

it takes it no more than 30 minutes

to go from one continent to the other.

That means whether or not you wish to fire back

before the weapons start landing,

you have 30 minutes to make that decision.

It was at a point in time where the civilization

could wipe themselves out in an hour

by nuclear exchange. For the longest time

in my career there at NSA,

I was trying to work... this is probably on the order of 13 years,

I was trying to work with the computer people

because I wanted them to do certain programs

that would assist me in what I was attempting to do.

Every time I would ask them to do something for me,

they would say, "okay, please send us a memo

saying you had this requirement request."

So I'd send them memo after memo.

That gave them a requirement.

That requirement gave them the justification

to go to congress or others internally in NSA and ask for money.

So that helped them build their empire, you know?

But in the meantime, they would never solve my problem.

They would just take the money and keep asking me for another requirement.

At one time in 1983,

miraculously this PC came up on my desk.

I had a friend who was probably

the best C language programmer at NSA.

His name was Bill, too, and I said, "Bill,

what do you think we could do with this PC?

If I could get an algorithm together,

do you think you could program that into this

two disk drive, 64k ram pc?

He says, "yeah, I think I could do that."

I said, "great."

So we would type in the query from unresolved issues out of the database...

And up would pop the answer.

Where the mainframe's eight acres of computers

could not figure out what this data was,

I would take it and put it in my little two disk drive pc

and get the answer right away.

So I had to tell the computer people,

and I said, "you know, I have this little program down here.

You might want to see this."

And I had this printout from the database,

and I was going through, "see, it didn't resolve this here. Watch this here."

And I'd type it in to my little PC,

and up would come the answer.

There's another one, and I'd type it in there, and up would pop the answer.

They started sweating, I could see.

And I said, "don't you think that's neat?"

He said, "yeah, don't you think that should be in the mainframe?"

I said, "sure. I think it should be in the mainframe.

Why don't you send me a memo and I'll give it to you?"

I thought that was great.

In this process, of course, I learned, very simply put,

if you want something done, you have to just go do it.

You never ask for permission, only ask for forgiveness, if you have to.

In the end, sometimes you don't have to.

Once you solve the problem, it's no longer a problem.

My philosophy was very simple.

If somebody says something is impossible,

that's something you have to do.

It's like open-ended thinking. You don't bound your thinking.

You let it float anywhere it wants to go.

So that gives you the idea of how you can be creative in any environment.

And so that's what I was after,

being creative anywhere I went.

And that's what the SARC, the Sigint Automation Research Center,

was all about.

The SARC was this Skunkworks organization,

about a dozen people.

Bill Binney was of the co-founders of that.

Ed Loomis was the integrator.

Kirk Weibe at that time was like a business manager and analyst.

And they had other support people as well. And they had a few contractors.

The SARC prided themselves on having a very small team

that was very creative and flexible.

They didn't want a lot of bureaucracy.

They had so low a profile that when I asked for a briefing from them,

the management didn't even know about them.

They had to scramble to find out who they were.

They were kind of hiding until they had their system perfected.

It was an extraordinary organization.

And Bill, he was the algorithm brain.

He was the crypto-mathematician behind the algorithms

of what became known as ThinThread.

People are telling me, "you gotta fear Bill."

And the reason you have to fear Bill

is he's like this super crypto-mathematician.

And if there's anything wrong, he'll shred you.

So I end up calling him up, we made arrangements to meet,

and I end up showing up in his office one morning.

I spent a couple of hours with him, and it was immediately clear

within the first 15 minutes that we were going to hit it off really, really well.

Because the entire first part of our conversation

was all mathematically based.

The thing that Bill loves most is math.

As soon as anybody raises anything mathematical,

his eyes literally light up and he perks up

and he'll start talking about it, you know?

He just loves math.

And it's wonderful that he found his profession,

you know, the thing that he was really meant to do.

The beauty of math is that it's an attempt

to structure common sense

and consistencies in the universe.

That's all it is.

There is no chaos.

There is nothing that's infinite.

Everything is structured.

And it's just a matter of finding that structure.

I basically envisioned the graph

as a very large globe of the earth, see?

With a total of about 7 billion nodes in it, or dots.

Each representing an individual.

And having all the connections of all those individuals

with all the other connections they have anywhere in the world.

Think of yourselves as drifting into the globe.

So you're going inside the globe

and you can go to any point in the globe

and look at any given point to see the relationships

they have with all the other dots in the globe.

When you go into the globe, you pick out a dot,

and you can pull it out of that relationship graph,

and with it will come all of the nodes

that it has a relationship with.

When you pull that node out and its community, if you will,

the social network it had,

you pulled all of the backup data

that created those relationships to begin with

so that you can now timeline those relationships

and see the interactions over a period of time

whatever you wanted to find.

So that's how I envisioned the graph.

My interest was in making Bill Binney happy.

Bill Binney is a traffic analyst.

So my focus was really on creating the metadata for him

that he could do his traffic analytic thing.

You can extract an awful lot about a person's life

just by looking at metadata.

And today, when you take all the metadata

that's being popped around the world

by GPS, by your cell phones,

by your transponders

for your E-Zpass on the highways,

by your credit card when you insert it into an ATM machine

or when you insert it into a cash register,

a check-out line,

all that data gets all brought together

and it all traces back to you, the individual.

Doesn't matter what you said in an e-mail

or what website you went to or what you downloaded.

What matters is what pattern your digital behavior is taking

and what fingerprints does that leave

of what you've been up to.

I first started to build on my portion of what I considered the ThinThread program,

which was the backend for analysis of the data once it was acquired.

Ed was doing the front end, doing the sessionizer

getting the data together

and assembling it for me to look at.

Ed's sessionizer was our starting point

to access and scan the internet for important information.

I was convinced that the internet was the new way of communications

for terrorists and criminals.

And we knew at that point that some of them

had been moving that way.

The internet exploded.

The NSA was actually playing catch-up.

They were so way behind the time.

NSA management seriously considered, no kidding,

as to whether or not they would actually give people e-mail technology.

And the reason is that we don't know what they're going to send to each other.

And there were no secrets worth knowing on the internet.

It's all open. It's not secret.

There's no effort. We want real secrets.

We don't... internet?

You can't have a secret internet.

In about 1996, I believe it was,

we discovered Osama bin Laden's Inmarsat phone number,

and so we used that to follow his movements

and his intentions from '96 to '98.

Until someone compromised the fact

that we had his specific phone number, and he gave the number.

And so after that, we lost that as a source of intelligence.

More than 80 dead, more than 1,700 injured

in two bomb blasts today, which exploded just minutes

and 450 miles apart.

Those explosions set to go off at the US embassies

in Dar es Salaam, Tanzania, and Nairobi, Kenya,

were clearly a part of someone's war

against the United States, although it is still unclear

at this hour who our enemy is.

There was smoke and debris everywhere.

We immediately left the office because we didn't know what had happened.

And there were people laying about,

and everybody that we came to was just covered in blood.

We're joined now from the White House

by US national security advisor Sandy Burger.

Was this a failure of US intelligence?

Well, the investigation is going on.

I don't believe that's the case, and I think it would be

inappropriate for me to comment on

what we know, what we don't know.

In August of 1998,

we successfully got our sessionizer running at the fiber-optic rates.

Ed gave a copy of that sessionizer

to our friends out at Bad Aibling,

one of our echelon sites for collecting data in southern Germany.

We just wanted to say, "here, why don't you try this

and see what you think and give us some feedback."

And he said, "gee, this sessionizer looks really great and works really great."

He says, you know, "I think I'll put it on the whole site."

They put it on their system Friday night.

And when they did that,

it started producing so much information

and feeding it back to NSA

that by Saturday morning it was about to crash

the entire storage system inside NSA.

It was too much data coming in. They couldn't handle it.

We were told it was bringing the storage system to its knees.

And we were to turn it off.

So one of the programmers had to come in over the weekend

and turn off the sessionizer.

This particular test clearly showed

that there was so much more data there

that was much more valuable as a source of information

than anybody realized at NSA.

This was kind of a wake-up call

that I thought made it clear to them

that they needed to open up their thinking

to a much larger scale than they were thinking at.

I had heard about Bill Binney, but hadn't actually met him.

I called him up. I said, "Bill,

can I come down and see you

and what your organization are doing?"

He said, "sure. Come on down."

So I go down, I join Bill Binney

in the Sigint Automation Research Center.

And I look around, and as I meet these guys and talk to them

and interview them, find out who's doing what,

and begin to see the results of their efforts,

I become literally excited.

In terms of facing the challenges of big data,

digital communications, and the internet,

I saw the answer right before me.

This little research organization

had the keys to NSA's future.

My concern immediately was

does anybody else know that?

Do they even recognize...

does anybody in any position of influence or power

even recognize what they have?

I was aware of the technology

that went into ThinThread from 1997.

I got a formal briefing from Bill and Ed and their team,

and I thought it was pretty impressive

compared to what else I had heard at NSA.

I thought it looked very promising.

I started giving them money.

And I would get Congress to agree

to make sure that they were funded to develop their systems...

Because the agency was admitting they had to update

to the digital age from the analog age.

They had not started it yet.

And even under the analog systems,

there is this deluge of data coming in.

And once you get the digital data,

you get these high capacity pipelines,

it was going to be far worse.

So it was going to be essential that there was some way

to pick out the targets out of this.

Some way far better than the keyword system

that NSA was using at that point.

Shortly after the Bad Aibling test,

we had the whole system running,

and then from there on, we just took in more and more data

of large numbers of phones and large numbers of computers.

We started consolidating that into my big graph,

which we called the big ass graph.

We started pulling in such a massive amount of data

and the system was working so well,

we could see no limit to it...

Building relationships between billions of people.

I mean, we're talking trillions of transactions.

It was pretty clear that we were building

the most powerful analysis tool that had been developed in history

to monitor, basically, the entire world

in near real time.

This is monitoring every member of the population automatically.

You invade everybody's privacy.

That's not compatible with a democracy.

That simply is not compatible.

That's like the stasi on super steroids

or the KGB on super steroids, you know?

This is things that they couldn't even dream of.

That's where I, in particular, said we have to do something

to protect the individuals here,

especially those who are innocent

of any wrong-doing whatsoever.

There's no reason to even have this data about them.

General Hayden came on as a new director of NSA.

By that time, the congressional intelligence staff

had become very disillusioned

with the NSA's ability to manage modernization.

He was somewhat forthcoming in a sense

that he was willing to make changes,

although not necessarily in the areas

that congress was focused on,

and he was a very fast talker.

When he would brief the members, he would talk very rapidly.

He would use all kinds of similes and analogies

and everything that left you scratching your head while he plowed ahead,

and he would get by with a lot that way.

I think the members,

this being a difficult enough topic to understand to begin with,

he left the members in the dust.

When Michael Hayden came in as director of NSA,

one of the first things he said was

the NSA was going deaf and blind.

So he created the NSA transformation office, the NTO.

A friend of mine became the chief of the NTO,

and he realized what we were doing with ThinThread,

and he thought that was great.

So he sent his deputy down, and he said,

"what can you guys do with $1.2 billion?"

Well, to be offered all of a sudden 1.2 billion was flabbergasting, you know?

It was kind of hard to deal with.

We had been operating with just a couple million dollars.

So we asked them to give us a little bit of time

to try to figure out what we might be able to do with that.

So when he came back in,

we said we could upgrade the entire world...

everything we have in the world,

and everything back here in the United States,

but in the process, we could only spend

maybe a little over $300 million to do it.

So he said, "that's really great."

A few days later, he comes back down.

He says, "you guys did such a great job

with 1.2 billion offer.

Now what can you do with 1.4 billion?"

It was very shortly after that that my friend in the NTO was replaced.

He was basically fired.

And Hayden proposed a program called Trailblazer.

And that was a multi-billion dollar program

that was going to be basically outsourced

and run by contractors and developed by contractors.

When the NSA started on Trailblazer,

I asked them if they would send down the design paper

and let me read it.

After I read it, I called NSA

and I asked if I could speak to the Trailblazer management.

And I said, "I want to make sure I understand this paper."

I said, "from what I understand,

to modernize digitally, what you want to do

is start with your analog system

and evolve that into a digital system

by putting out one part at time

and upgrading and adding more parts."

And they said, "yes, that's it exactly."

And I said, "it will never work.

It's a non-starter. It will never work."

Why would we start from outmoded analog technology

and try to evolve that into digital technology? It just... it made no sense.

You had to go with the new modern digital technology from the outset.

One of the reasons that I supported ThinThread

was that I thought it was far more innovative

and much more simple and elegant in design

than what they were considering,

therefore probably posing fewer problems.

From our tests with our big ass graph,

it was pretty clear to us that we had to protect

the privacy rights of individuals.

So we devised a way to build in protections

and make it possible to encrypt all their attributes

so you couldn't tell who they were

and you couldn't analyze what they were doing

because you couldn't identify them.

And in cases where they weren't even in zones of suspicion,

which we would define within the graph,

their data wouldn't even be taken in,

so that no one would have an opportunity to abuse their privacy.

This is also metadata coming in.

The FISA filtering tool that we had

filtered out any American citizen

or party that was in America

unless it was a specific target.

But even then, you would have to get a warrant

to reveal the identity.

By 2000, we had it all connected

and everything together and running from one end to the other

in 24 hours a day

in three different sites focused on Asia.

The only sites that agreed

to cooperate with us

were those sites that were known as research sites.

You know, I never called it a test site, but...

I mean, they were existing sites, I mean.

And they were producing intelligence from that

for a long period of time.

We went to NSA and asked for a briefing.

The SARC team came in and to my great surprise,

the entire Trailblazer team came to the briefing uninvited.

Bill and Ed started the briefing,

and what was astounding was that it went all the way

from the point of collection all the way through reporting,

and including automated reporting.

So I turned to the Trailblazer people,

"are you going to incorporate this into Trailblazer?

Wouldn't this help you?"

And they were pretty much noncommittal and stony-faced.

And it turns out that SARC got in a lot of trouble for this

after I and other congressional staff left.

General Hayden called in the entire SARC team,

lined them up in his office, dressed them down,

berated them without letting them have a chance to speak.

And then issued a dir-gram

telling the entire NSA

that once a corporate decision had been made,

that they were not to undermine it with Congress.

In other words, they were to cover up any adverse information.

It was, to me, an absolute clarion call

to keep Congress in the dark.

I thought it was very brazen.

General Hayden basically said

ThinThread was no contribution to the Trailblazer program.

Trailblazer is the program, and we're not gonna have any other program.

ThinThread, in my view, once we had it all together,

solved all the problems of volume, velocity, and variety.

We had solved all those problems

that they were claiming were problems.

But they rejected it as even a risk mitigation process.

I mean, that's the kind of experience I had

with generals back in 1968 when just prior

to the Tet Offensive we were predicting

that offensive two months in advance.

Some people came to me and showed me some reports

on activity in Vietnam,

and they said that they were preparing a major offensive

that would go from the delta all the way up north Vietnam.

It was all reported to all the major commanders in the field

as well as the pentagon and everybody,

but nobody in a position of authority paid attention to it.

General Westmoreland was again saying to the people in the United States

that we were winning.

We're watching the situation carefully.

And I'm confident that any further initiatives can be blunted.

I basically looked at that as

a philosophy of the arrogance of power.

They felt they had so much power that they were invincible.

I had friends who were stationed in Vietnam.

One colonel that I knew had a battalion

stationed on the road from Saigon to Cambodia,

which was a major point of attack.

He, of course, had the reporting, too.

So he had 50% of his men in the trenches all the time,

24 hours a day...

Until the attack.

He was ready, and he lost one man on the first day,

and he stopped them cold.

The problem was that all the forces on either side of him

had fallen back because they couldn't hold their position on the line

because they weren't ready.

They didn't listen to the intelligence.

So he had to eventually withdraw.

The point was if they were ready,

that offensive would have gotten absolutely nowhere,

but they didn't pay attention.

This was a deceitful

and treacherous act by the enemy.

We thought they were all idiots.

I thought every general in the country should be fired after that offensive.

All of them.

They should have been drummed out of the service.

I mean, we got rid of generals and all that during World War II

for very similar kinds of negligence,

so I don't understand why we didn't get rid of them then.

And in terms of Americans, I think there were a little over 2,000 lives lost

in that offensive.

That doesn't count how many were wounded.

That's why I looked at it as they were treating us

as simply meat on the line.

You know, throw more meat on the line.

We begin our special coverage tonight

with the latest on the terror attack

against the US guided-missile destroyer Cole.

The navy says it has recovered the bodies of seven sailors

and is searching the mangled interior of the ship for ten more.

About 35 other sailors were injured.

In dreary weather fitting the occasion,

the first bodies of sailors killed aboard the Cole

arrived in Germany this afternoon.

The blast which took their lives

is estimated to have been caused

by several hundred pounds of explosives

in a new and deadly terrorist tactic.

Immediately after the attack

there was nothing left of the small local harbor craft

which the navy says had carried the explosives.

Despite instability in the immediate region,

the strategic port of Aden has been used as a refueling stop for two years

and the Cole was on a heightened state of alert.

But there was no defense it seems

against suicide bombers in a small boat.

We decided to continue developing ThinThread

to try to insure that we had something that would functionally work

by the end of 2000

so that we could address the terrorism problem

in the real world, not in the wish world

of Trailblazer, for example.

That's when I went to our terrorism analysis center

and said, "okay, what sites do you have that will...

that produce any meaningful information for you

to analyze the terrorism targets around the world?"

And so they gave me a list of 18 sites,

and so I took that as the target set to go against terrorism.

And then I came in November of 2000

and made a proposal that we do a deployment of ThinThread

to those 18 sites starting in January of 2001.

But it came back rejected.

So I don't know exactly who rejected it.

I assumed it was the chief of SID, Maureen Baginski.

She was the third person in rank in the agency.

There was the director, the deputy director, then her.

General Hayden hired Bill Black

to come in and take over the deputy directorship.

Once Bill Black was brought in from SAIC,

where he had been the vice president,

the following spring, he brought in

another senior vice president from SAIC,

and that person was Sam Visner.

And who gets the contract to develop Trailblazer?

SAIC.

What's wrong with this picture?

SAIC was a private company.

It is made up basically of retired NSA individuals.

People from NSA retire and they'll go into a company like SAIC,

and then they'll use their contacts back at NSA

to get contracts back to SAIC.

So when Sam Visner came in to take over

the NSA transformation office,

the first thing he did was to call me to his office.

So I went up to his office to talk to him,

and when I went in there, the senior developer

for the Trailblazer program was there.

And the first thing Sam told me,

he said, "I do not want you briefing ThinThread to anybody again."

And the reason he said that, of course,

was because we had already set up an appointment

to brief Charlie Allen.

He was the senior collection officer

for the intelligence community,

so he had a lot of influence.

So I asked him if he wanted me to cancel that appointment,

and he said, "no, no, don't do that.

You go ahead and finish that briefing."

I said, "okay, I'll do that."

When Charlie Allen got in there,

the first thing he sat down and said to us,

even before we started to show him anything about ThinThread,

he said, "I hear you have a cheap Trailblazer here,

and I'm here to see it."

Well, I mean, I had been invited.

Maureen Baginski had come up

'cause Maureen knew Charlie Allen.

They were friends for a long...

they had known each other for a number of years, I think.

And she was there, and she said...

at that point, she said, "yes, Charlie,

but we still need the money."

That was the end of our briefing on ThinThread.

On August 20th of 2001,

Maureen baginski called me and asked to speak with me and Bill Binney

at 4:30 in the afternoon that day.

And she said to meet her in her office,

so Bill and I went down to her office.

And it was in this meeting with her that she advised us

that she was terminating ThinThread.

And she said that her decision was made because

she had to decide whether or not

she wanted to make six people unhappy

or almost 500 unhappy.

So she chose to make the six people unhappy.

So her decision was what I basically call

happiness management.

They are letting us stay up. I'm not sure why.

Newark is allowing us to stay up.

We're... only the news helicopters are allowed up of the entire country.

We're, like, five people in the air, that's it.

Oh, my god, what's going on?

It's like the side of the building just blew out, Chad, on the other side.

The side we can't see.

Look at it.

Oh, my God!

The whole tower collapsed.

The whole building is gone. The whole building.

Oh!

Daniela, did you hear me? The entire building collapsed.

It is gone.

The second tower is gone.

Oh, those poor people. Oh, God.

On 9/11, I had taken my father-in-law

to the eye doctor for an examination,

so I was sitting in the waiting room watching television

while he was getting examined.

And that's when I saw the first plane

hit the tower in New York.

And immediately that said to me

that we as an agency had failed to give warning

of this terrorist attack.

When my father-in-law was done with his examination,

I took him home and then tried to get back into NSA

to see what I could do

to maybe help in getting information

about who did this and all.

And they wouldn't let me back in simply because

director Hayden had sent everybody home

that wasn't necessary for the job,

and so the guards were keeping everybody out,

so we couldn't get back in that day.

So the day after 9/11,

I came in dressed like I was going to sweep the floor,

so the guards let me in.

Otherwise they would have kept me out then, too,

because that day,

General Hayden had ordered everybody out of the building, too.

So when I got in there, I went up to the SARC

And while in there trying to look at the material on my computer,

my contracting...

president of the contracting group

that I had working on ThinThread

came over to me and said that he'd just been

in a contractor meeting with Sam Visner.

And in that meeting, Sam Visner had told him that...

he said to him, "do not embarrass large companies.

You do your part, you'll get your share.

There's plenty for everybody."

And he also said that Sam had said that,

"we could milk this cow for 15 years."

We began to go out to the workforce

and I would accompany Maureen Baginski.

The workforce took this extraordinarily hard.

People would ask her, "what are you gonna do?"

I still remember her saying,

"9/11 is a gift to NSA.

We're gonna get all the money we need and then some."

Direct quote.

"9/11 was a gift to NSA."

All she saw was dollars coming from congress,

and we would get to spend freely.

It was not what we were supposed to be about.

It wasn't about making money,

it was about protecting the country

and the free world. It wasn't... you know,

wasn't how big of an empire we could build

or how much money we could make, you know?

It was just sickening.

In late September or early October,

we in the SARC were watching as all this equipment started coming in

and piling up in the hallways and outside the SARC

and inside the SARC in the walkways

between the partitions for the desks.

We knew something was going on because of all this equipment coming in.

And then later on in early October,

it started to disappear.

That meant they were assembling the hardware

somewhere down the corridor

a little bit down from where we were.

And then by mid-October, the second week in October,

some of the people who used to do the...

who I had under contract to do the software for us in SARC,

came to me and said that,

"you know what they're doing down there?

They're taking in data on every US citizen in the country."

To do the spying on US citizens,

they had to use the only software that managed large-scale operations,

which was the software that I put together,

which was the backend of ThinThread,

that part that does the graphing

and the management of information

indexing to the graph.

And one of the first things they did

to make that process work efficiently, of course, according to them,

was to remove of all the encryption algorithms we put into it...

To protect the privacy rights and the identity

of US citizens or any other citizens in that database.

Well, I mean, after I was told

that they were pulling in all this data on US citizens

and basically graphing the relationships

of every US citizen that uses a telephone or anything electronic,

well I... it was clear to me that I had to leave the agency.

I couldn't stay there, so I had to get out as fast as I could.

Bill, Ed, and kirk became so disgusted

with all the resistance they were getting,

and with some of the personal attacks on them, I think also,

that they decided to take an offer to retire

at the end of October,

and they told me that they planned to do that.

Our last breath at NSA was October 31st, 2001.

Halloween day. Kind of fitting.

After 9/11, and after we left,

Tom took over running ThinThread.

And he said, "well, let's take the software in ThinThread

and run it against the database

that we've accumulated inside NSA

to find out if there's any information on the 9/11 attack

that we should have known about, but didn't.

We let it run... it was somewhere around 24 to 36 hours.

I remember my program manager coming back

with the initial results

and they were just devastating.

We discovered critical intelligence,

Al-Qaeda and associate movement intelligence

that had never been discovered by NSA.

They didn't even know they had it in their databases.

But also the specific details

of numbers and movement

and times and locations... travel.

And you're seeing the dispersal patterns

that occurred after it happened.

You're also seeing that there was parts of the plot

in which they weren't successful.

There was multiple planes that had been targeted.

And for a number of reasons,

those particular planes were not actually hijacked.

You can imagine the horror of realizing

what had not been discovered

and the confirmation of what was known

sitting in these databases.

What if this was discovered prior to 9/11?

We had the information in the databases.

NSA's response?

Completely shut the program down.

9/11 would have been avoided.

We would have caught the people

who were communicating from San Diego

with people abroad, with known terrorists abroad.

There was no way we wouldn't have caught that.

After Ed and kirk and I left NSA,

we formed a little company, a consulting company

we called Entity Mapping, LLC.

We decide to try to bring the concepts of ThinThread

to other agencies and government

and bring it to the US government through another door.

We were requested by

the Senate Intelligence Committee staffer

to go to the National Reconnaissance Office.

They manage all the satellites and reconnaissance kinds of missions.

We were working there on a program,

but NSA found out about it.

They got the whole program terminated.

Then we went to customs and border protection.

They got us fired from that one.

Bill and Kirk were just infuriated by every time

we went to get a contract with a promising customer,

at the last minute, it would be shut off.

Happened with INSCOM, it happened with the pentagon,

it happened with NRO, it happened with CIA.

We are stopped at each one of those,

even though were are successful.

- Okay, I think we got it. - Yeah.

Including one

where we were given a set of data

and asked to analyze it

even though NSA had already analyzed it.

We found extremely valuable material in that data

where NSA had not.

We wrote up a report, sent it up the chain.

A senior executive in that agency said,

"my god, these guys are going to embarrass Michael Hayden

and the people at the fort.

We have to stop this program."

And it stopped immediately. We were told to stop work.

Don't do anything else.

And they proceeded to dismantle the computers

that we were working with

and the data that we had discovered

and all of the intelligence

was wiped off the disks.

NSA attempted to keep the principles of ThinThread

being applied anywhere in the US government

so that no other agency in the US government

could take advantage of it either.

So in essence, they wanted to keep the US population

and the free world's population vulnerable,

which is what they did.

And that's why we filed the DoD IG complaint

in September of 2002 against NSA

for corruption, fraud, waste, and abuse.

Kirk drafted up a very strong fraud complaint

and we said, "let's go with it."

And Diane was also a party to this as well.

She agreed to go along.

So there was Kirk, Bill, Diane, and myself,

and then Kirk was actually floating this by Tom Drake.

I was not aware that he was,

but he was sharing it with Tom Drake.

And Tom, since he was still working at NSA,

said that he would support us from the inside.

I had access to critical information

that clearly NSA was hiding.

Because of all these investigations that were ongoing,

NSA said, "let's find where the skeletons are,

capture it all, and then bury the report as deep as possible."

This was actually something that Bill Black,

the deputy director, was doing.

And the order went out,

"no one will ever see that report.

We can't show anybody

what ThinThread discovered that we hadn't discovered.

We can't show ThinThread had confirmed

what was clearly a failure."

So I was contacted officially

and arrangements were made for me to meet with the investigators.

And so the DoD IG's office took that investigation on

with about 12 investigators,

and took them about two and half years to do it.

They finally published their report

in, I think it was, January of 2005.

And at the time, about 98% of it was redacted

because it was so damning to NSA.

Many of the redactions are on paragraphs

that are marked unclassified.

So it was clear to me

that the dirty laundry was blackwashed.

This was our justice department covering up for those crimes.

In 2005,

Trailblazer was declared an abject failure,

having wasted billions of dollars.

A lot of millionaires were made during that time,

but not much intelligence got done.

It was the largest failure in NSA history.

People don't realize that.

And yet, general Hayden,

who presided over the largest failure in NSA history,

was promoted to a higher position of authority,

made deputy director of National Intelligence,

and then promoted again

and made director of Central Intelligence.

It is odd when mankind promotes failure.

Something is wrong.

On the 26th of July of 2007,

the FBI decided to raid four people.

That was Kirk Weibe, Ed Loomis,

Diane Roark, and myself.

So they fabricated evidence

to get a warrant signed by a judge

to go in and raid our houses and take our material.

Specifically, they were after all the material

that was related to ThinThread.

My son answered the door and let them in,

and they came in with their guns drawn

and pushed him out of the way at gunpoint,

and came into the bedroom where my wife was getting dressed

and pointed guns at her, and then came into the bathroom

where I was taking a shower and pointed a gun at me.

I didn't hear them at first, apparently.

By the time I heard them,

it sounded like they were going to break down the door.

They were rattling it madly and banging it back and forth.

When the raid occurred here,

it was such a shock to my inner core.

I couldn't believe it was happening.

I went to call my attorney first.

She told me to ask for the affidavit

that justified the raid.

I asked them, they said it was classified and I couldn't see it.

As soon as I got off the phone, I realized they had gone to the office

and picked up all the most relevant information,

all of the stuff about NSA that was on the shelves,

which was, I think, a first indication

that they had entered the home previously.

When all that equipment was taken from my house,

my computers, my data, my newspapers,

my memorandums that I had shared with friends of mine,

I just could not believe

that I was perceived as an enemy

by my own country.

I think... I keep thinking if this happened to us,

and we were educated people,

sophisticated in government ways,

who could get some help,

what happens to the man on the street?

What happens to the 18-year-old kid?

When I found out that all the things that we were doing

as a result of that 9/11,

and I heard vice president Cheney's statement about,

"we have to go to the dark side,"

I've looked at it as a total destruction of our society,

and we lost all moral standing in the world

as a result of going to the dark side.

It was just sad.