Network News

X My Profile
View More Activity
Posted at 8:00 AM ET, 12/ 5/2007

The CIA's Biggest Bloopers

By Michael Dobbs

Iranian clerics watch a missile test near Qom, November 2006.

So the CIA got it wrong on Iran's nuclear program in the last National Intelligence Estimate, back in 2005. But does that mean they have got it right this time? Not necessarily. The history of the CIA is littered with spectacular intelligence mistakes. Sometimes, the correction of one error can lead to a new error, as analysts atone for past mistakes by moving too far in the opposite direction.

In the spirit of caution and skepticism, here is the official Fact Checker list of the CIA'S Biggest Bloopers, over six decades of intelligence-gathering. I have compiled it with the assistance of researchers at the indispensable National Security Archive, a non-profit group that has published more than half a million government documents. A disclaimer: the Agency has had some successes too, but I will let their public relations operation draw up that particular list.

Note: National intelligence estimates are issued on behalf of the entire intelligence community, not just one agency. Up until 2005, the CIA director was also director of national intelligence.

Getting it Wrong


Soviet control over Eastern Europe. In an NIE released in January, the CIA said that Moscow would remain in full control of Eastern Europe through 1960 at least. Five months later, there were riots in Poland, followed by a revolution in Hungary in October/November that had to be put down by a Soviet invasion.


The "Missile Gap." NIE 11-5-58 predicted that the Soviet Union would have 500 intercontinental missiles "sometime in 1961, or at the latest in 1962." After the U.S. launched a spy satellite called Corona in 1960, the estimate was downgraded to 10-25 ICBMs. The Soviets actually had four ICBMS in 1961. To be fair to the CIA, this is an example of where they get it wrong, but they also thought up the technological solution, in the form of Corona.


The Bay of Pigs. The CIA knew that a planned invasion of Cuba by 1,500 Cuban exiles had little chance of success without the participation of U.S. forces, but failed to inform President Kennedy.


The Cuban Missile Crisis. On September 19, the CIA told Kennedy that the establishment of a Soviet missile force on Cuban soil was "incompatible with Soviet policy as we presently estimate it." A month later, an Air Force U-2 took photographs of Soviet missile sites. This is another case where the CIA got it wrong, and then partially rectified the mistake. (The U-2 was a CIA program.) They still missed a hundred or so battlefield nuclear weapons on Cuba, and underestimated the number of Soviet troops on the island by a factor of three.


The Soviet ICBM buildup. The CIA missed the Soviet missile buildup, partly in response to the humiliation of the Cuban missile crisis. A subsequent CIA director, Robert Gates, later wrote that the Agency "did not foresee this massive Soviet effort to match and then surpass the United States in strategic missile numbers and capabilities -- and did not understand Soviet intentions." This seems to be a case where the Agency swung from one extreme to another. Having overestimated the Soviet missile buildup in the Fifties, they underestimated it in the Sixties.


The Iranian revolution. In August 1978, CIA issued an NIE that said Iran "is not in a revolutionary or even a prerevolutionary situation." The Shah fled Iran six months later.


Two blunders on Iraq. On July 31, The CIA dismissed the likelihood of an Iraqi invasion of Kuwait. Saddam Hussein invaded two days later. The CIA also significantly underestimated the scale of the Iraqi nuclear weapons program.


The Indian bomb. The CIA failed to predict the testing of an Indian nuclear bomb in May 1998. The chairman of the Senate intelligence committee, Richard Shelby, bemoaned "a colossal failure of our nation's intelligence gathering." The CIA was better prepared for the first Pakistan nuclear test a few days later.


Iranian missiles. A September 1999 intelligence forecast said that Iran could test an intercontinental ballistic missile capable of hitting U.S. territory "in the next few years." Eight years later, Iran has made little progress toward acquiring an ICBM. In a January 2002 article for the Post, I argued that the upgrading of the Iranian and North Korean ballistic missile threat came at least partly in response to political pressure from the missile defense lobby.


Iraqi weapons of Mass Destruction. The CIA, in NIE 2002-16HC, said that Iraq had "continued its weapons of mass destruction program," and could build a nuclear bomb "within several months to a year" if it obtained the necessary fissile material. Evidence for such a program was never found and it subsequently turned out that a key CIA source, a defector codenamed Curveball, had lied extensively. As with the October 1962 NIE issued just prior to the Cuban missile crisis, the 2002 NIE illustrates the corrosive power of conventional wisdom. Since Iraq previously had a WMD program, the operating assumption was that it still had one.


Iranian nuclear weapons. An NIE predicted "with high confidence that Iran currently is determined to develop nuclear weapons." In December 2007, a new NIE judged "with high confidence that in fall 2003, Teheran halted its nuclear weapons program."

My thanks to Malcolm Byrne, John Prados, and Jeff Richelson of the National Security Archive and John Pike of for helping me compile this list. I am responsible for any errors. Please use the comments section to nominate your own favorite CIA blooper. I will select the best (i.e. worst) ones, and add them to my list.

By Michael Dobbs  | December 5, 2007; 8:00 AM ET
Categories:  Gov Watch  
Save & Share:  Send E-mail   Facebook   Twitter   Digg   Yahoo Buzz   StumbleUpon   Technorati   Google Buzz   Previous: 'Bomb, Bomb, Bomb; Bomb, Bomb, Iran'
Next: Clinton vs Obama: The 'Fact Check' War


Over the years the CIA has been without enough agents of our own and has relied on what they wanted to hear from just about anyone. Very few people with a little money and determination can do great damage to our country as evidenced by 9/11. Many still thinks JE Hoover had more personal knowledge of enemies of this country than the entire CIA. This just released report has been known for years and many believe it was withheld for the sole purpose of getting this country into another war.

Posted by: lylepink | December 5, 2007 9:50 AM | Report abuse

What is the point of this list? Is Mr Dobbs going to give a pass to candidates who continue to fearmonger over an Iranian nuclear weapons threat, and he just wanted to give us a heads up? I'm going to see if I can google up Mr Dobbs' similarly skeptical reporting of the 2005 NIE on Iran. Wish me luck.

Posted by: david | December 5, 2007 10:05 AM | Report abuse

Like the first poster said , the CIA relied on some wacko named curveball who played Bush and the CIA like a couple first graders - I think right now Bush and Cheney will listen to ANY one who will tell them what they want to hear and dismiss those that - - are not loyal Bushies or willing to tow the party line

Friedman has a excellent article in the NY times that pretty much sums up how ignorant and stupid we have become since 9/11 - very worth checking it out

Posted by: Michael I | December 5, 2007 10:07 AM | Report abuse

...and where are the Pinnocchios or the check mark? Is Mr Dobbs blazing some new mission where he just generally comments on the reliability of government agencies? Perhaps he will consider Pentagon claims over the course of the Iraq War. Or the executive's claims over same.

Posted by: david | December 5, 2007 10:09 AM | Report abuse

60 years and only 11 times where they were wrong? I'd say this article just proved how good the CIA really is, I mean hindsight is 20/20 and we have the benefit of looking back at the mistakes they made but they have the hard job of trying to predict what's going to happen before it happens generally that's what the estimates do. Getting only 11 wrong out of the god knows how many times they had to do an estimate shows that we're not perfect we make mistakes we're allowed to do that, but it also shows that they know how to do what they were chartered to do...but that's just my opinion

Posted by: Bryan | December 5, 2007 10:25 AM | Report abuse

If decades of Post articles were put under the same litmus test, how many errors in judgement and analysis would a researcher find?

Posted by: Teresa | December 5, 2007 10:46 AM | Report abuse

Check out Countdown last night for a real good laugh about Lou Dobbs and how Keith did such a great job on him.

Posted by: lylepink | December 5, 2007 10:47 AM | Report abuse

Bush/Cheney knew of the latest NIE as they continued to use fear to govern. They were also aware their information re:Iraq was wrong.Unfortunately fear works as the fact checker himself is in fear of awarding his pinocchios to Bush/Cheney for these warmongering statements. This site used to be informative and entertaining but it is now a go along to get along offering.The author needs to move to the style section and discuss hemlines. His input would be more useful in that venue.

Posted by: Mark | December 5, 2007 10:53 AM | Report abuse

Amen, Brian. That's a pretty paltry list of mistakes, especially knowing that the CIA is involved in hundreds of clandestine missions every year.

Posted by: Anonymous | December 5, 2007 11:03 AM | Report abuse

Why do you think the CIA got so many things wrong? Is it because of political pressure on the CIA, some institutional problem (like maybe having so many spy agencies simply creates more opinions and uncertainty), or is it simply hard to do good intelligence? I would really like to know.

Posted by: Jon Webb | December 5, 2007 11:09 AM | Report abuse

By charting the intelligence community's mistakes, Dobbs is focusing on the wrong angle. The important thing is that Bush--who gets regular briefings on intelligence about Iran, and undoubtedly has known for a long time what was in the NIE--has been misleading the U.S. public about the nuclear threat supposedly posed by Iran, in an attempt to justify another war.

Posted by: Patrick J. Kiger | December 5, 2007 11:12 AM | Report abuse

How about the collapse of the Soviet Union?

Posted by: Matt | December 5, 2007 11:40 AM | Report abuse

The Indian nuclear tests being a surprise has never made sense. I knew that India was going to test. How? I read the papers. The BJP publicly promised to test when they got into power, and they followed through. Those of us working on nuclear testing issues all knew it was coming, so why didn't the CIA? Something's fishy.

Posted by: Tracy | December 5, 2007 11:53 AM | Report abuse

How is 9/11 not on this list?

Posted by: Bob | December 5, 2007 11:53 AM | Report abuse

Perhaps not huge, but it makes you wonder:

"We believe the Soviets are planning for a manned Mars landing mission some time after the year 2000," the CIA analysis stated.

See "Soviet Options for a Manned Mars Landing Mission," CIA Directorate of Intelligence, December 1989, released in "sanitized" form in 1999, here:

Posted by: TexLex | December 5, 2007 1:32 PM | Report abuse

Agree with Teresa. How many blunders and downright lies has the Post published over the years? Add to that, the questionable motivation of these sources: Malcolm Byrne, John Prados, Jeff Richelson, and John Pike, at least a few of whom have knowingly compromised national security by disclosing classified information.

Posted by: Anonymous | December 5, 2007 1:35 PM | Report abuse

Bob, 9/11 is not on the list because they got it right. Bush was warned. He dropped the ball. And who, but the Bush administration, were touting Curveball?

For the WaPo to print this list when Bush has been deceiving us about what he knows vs. what he has been saying about Iran, is just an apology for Bush.

Fact: Bush lies about what the intelligence community is saying about Iran, just as he lied to us about what they were saying about Iraq before the war (aluminum tubes, uranium from Africa).

WaPo response: Hey, look at all the mistakes the CIA has made.

I dare the editors to put up a list of Bush's biggest "bloopers".

Posted by: epistemology | December 5, 2007 2:00 PM | Report abuse

I think you left out the fact that there were CIA agents like Tyler Drumheller, who knew that Curveball was a con man. He tried to warn Tenet and Colin Powell not to use Curveball's info at the UN in 2002.

Convenient ignorance of the truth is the hallmark of the Bush admin's foreign policy.

And it still happens today, to our shame.

Posted by: SPENCER | December 5, 2007 2:10 PM | Report abuse

Oh, come on, why stop at only 60 years? Intelligence failures are an old, and honored, American tradition, how about these:

• Korea, 1950--missing, or disregarding, evidence of Chinese troop buildups along the Korean border, and Chinese statements of their intention to prevent the American occupation of North Korea.

• Belgium, 1944--missing the buildup of German troops and armor in the Ardennes in advance of the Battle of the Bulge.

• Pacific Ocean, 1941--we are caught completely by surprise by Japanese aggression from Hawaii through the Philippines east into Malaysia.

• Sharpsburg, Maryland, 1862--President Lincoln is unable to get General McClellan to pursue Lee's devastated forces after the Battle of Sharpsburg. Federal intelligence reports dramatically overestimated the size of the Lee's force and number of casualties experienced.

• Washington, 1814--the White House and Treasury building are burned after British troops capture the city. Intelligence sources failed to anticipate their attack as retaliation for the American burning of Toronto.

• New York, 1776--lacking intelligence on British intentions, General Washington divides his troops between Manhattan and Long Island in the face of superior British forces.

• Pennsylvania, 1755--British General Edward Braddock (and militia Lt. Col. George Washington) is defeated by an inferior sized force of French troops and Native American allies. Gen. Braddock was unaware the French had left Fort Duquesne to launch a pre-emptive attack on his superior force.

Posted by: 20th Street and Pennsylvania Ave., NW | December 5, 2007 2:22 PM | Report abuse

Do you believe the NIE report to be right or wrong regarding Iran's Nuclear program?


Posted by: jeffboste | December 5, 2007 2:23 PM | Report abuse

"How about the collapse of the Soviet Union?"

Ditto. The cash and manpower thrown at the FSU was gi-normous, and they STILL blew it.

Posted by: Amos 'n' Andy | December 5, 2007 2:58 PM | Report abuse

how does CIA involvement with cocaine smuggling never make these lists. This is all very well documented.
or instead of listing the hundreds of available reports and stories, just review

Americans need to understand that the CIA protects its own interests at the expense of the American population. Some quick reading will make this plainly obvious

Posted by: citizen | December 5, 2007 4:04 PM | Report abuse

DR Sidney Gottleib and the Special Operations Division at Fort Detrick and MKULTRA and the human experiments using 7120 enlisted soldiers and airmen as test subjects in experiments that violated the Nuremberg Codes of 1947, I think that was a big blooper and DOD and the VA refuse to help these veterans or their widows, these men and their families are still paying for the CIA's run amok test program.

Posted by: Mikey30919 | December 5, 2007 4:27 PM | Report abuse

The CIA's involvement with cocaine smuggling is not "very well documented." In fact, the very doj report cited by the individual as support for that contention states the opposite. The executive summary to that report (hyperlink above) concludes that the allegations of such activity were exaggerations or unsupported by the evidence:

"[T]he allegations contained in the original Mercury News articles were exaggerations of the actual facts."

"We also found that the claims that Blandon and Meneses were responsible for introducing crack cocaine into South Central Los Angeles and spreading the crack epidemic throughout the country were unsupported . . . . [W]e did not find that their activities were responsible for the crack cocaine epidemic in South Central Los Angeles, much less the rise of crack throughout the nation, or that they were a significant source of support for the Contras."

Posted by: Response to Citizen | December 5, 2007 4:48 PM | Report abuse

Since 2002 the public statements of Vladimir Putin on Iraq and Iran's WMD programs have proven more accurate than the CIA's assessments.

Posted by: Chris Baker | December 5, 2007 4:54 PM | Report abuse

In the recent released documents "THE FAMILY JEWELS" they meaning the CIA admit to drug testing on military personel.
Volunteer ya right!

4) CIA Science and Technology Directorate Chief Carl Duckett "thinks the Director would be ill-advised to say he is acquainted with this program" (Sidney Gottlieb's drug experiments) (p. 213)

(pg 413)

Posted by: Class of 72 | December 5, 2007 8:26 PM | Report abuse

Funny how you can run an article on how off the CIA has been when they say that Iran is not a threat when all the Neocons are chomping at the bit to bomb Iran, but where was the same article before we blundered into Iraq on CIA info.

Posted by: mdyoung | December 5, 2007 9:11 PM | Report abuse

I really do not know if this is showing how bad the CIA is or how good it is? Out of 60 years they made about 11 mistakes, isn't that good? They should not be making mistakes at all but, they are people like us, they can make mistakes, we all do. The CIA will trust anyone because we are short of agents and they trusted some guy named Curveball, and he lied, big time. I think, instead of having all the join the army and navy and stuff, they should have join the CIA advertisements, there has to be at least some people who still has dreams of being a secret agent. All the CIA has to do now is learn from their mistakes, which I think they already are doing, and make the future better for all of us.

Posted by: somemongolianguy | December 5, 2007 10:12 PM | Report abuse

It's not even fair to call an estimate which later turns out to be flawed a "mistake." Only when such an estimate is conveyed with firm confidence is the term mistake fair.

DC Warped Groupthink (DCWG) holds that the NIE is politically motivated when it supports military action, and noble and unbiased when it opposes the use of force.

Posted by: The Angry One | December 6, 2007 12:04 AM | Report abuse

Gee Mr. Dobbs, ever consider that the CIA is ORDERED to tell lies by certain administrations? Take for instance the WMD's and Iraq. If you read between the lines, or listen to people in a position to know, there was tremendous opposition to that fraud from inside and outside the agency. Colin Powells U.N. presentation of U.S. intel was comically fraudulent on it's face.
The 2005 NIE also said what the President WANTED it to say. Todays NIE, more then likely, reflects the shifting power in Washington, and the intel communities unwillingness to provide false intel so Bush, the Neo-Cons, and you freaking insane supporters of them, can take us to war in Iran.

Posted by: pclement1 | December 6, 2007 3:45 AM | Report abuse

somebody at the CIA just decided to tell the truth - NIE Report - and the Bushies can't stand it. Seemingly, the Republican Senators think that the drums for war is their chant for re-election.

Posted by: dinnylou | December 7, 2007 7:17 AM | Report abuse

Jon Webb asked why the CIA gets things wrong.

One reason is deliberate disinformation by the enemy. During the Cold War, virtually every single one of our agents operating in East Germany had been identified and turned by the East German service, the STASI, and were used by them to feed the CIA disinformation. We only found out about how successful the STASI had been at this, after German reunification and the end of the Cold War, and the CIA and the STASI finally got to compare notes.

And that is the big fear that British MI-6 now has about the current NIE: They believe that the Iranians found out that they were being bugged by the CIA and were feeding us false disinformation. The East German example should give us pause: When evaluating a potential adversary, if something sounds too good to be true, it usually is.

Posted by: sinz52 | December 10, 2007 10:47 AM | Report abuse

The reason US intelligence on Iran's nuclear ambitions is so bad is that a bunch of traitors to the USA, outed the agent in charge of US counter proliferation office for the Middle East along with all her assets and her cover company. And obviously any agent or asset that had contact with either her or the cover company also had their cover blown by the same heinous act.

We all know who is to blame.

Posted by: walker1 | December 11, 2007 6:17 PM | Report abuse

Pretty egregious to exclude from the list the Intel Community's failure to predict the scope and pace of the collapse of the Soviet Empire.

Also, there is the matter of hyping the Soviet military threat from the Intel Community for decades, until we got to see those military systems first hand in the years after the fall of the Berlin Wall ...

Posted by: HillRat | December 12, 2007 4:13 PM | Report abuse

The comments to this entry are closed.

RSS Feed
Subscribe to The Post

© 2011 The Washington Post Company