Benlog

crypto and public policy

If you know that I know that Chalabi knows….

Filed under: Security & Crypto June 2, 2004 @ 12:12 pm

CNN reports that Ahmed Chalabi told the Iranians that the US had broken its codes. Read the article more carefully, and you get the feeling that either Iranian intelligence needs an upgrade, or CNN (well, Associated Press) does:

Chalabi told the Baghdad chief of the Iranian spy service that the United States was reading its communications and […] the Iranian spy described the conversation in a message to Tehran, which was intercepted by U.S. intelligence.

As John Stewart would say “so the Iranian spy hears that the US broke his codes, and he immediately reports this news to headquarters… using those codes!

Iranians in Tehran then sent a bogus message to Baghdad purportedly disclosing the location of an important weapons site, in an apparent attempt to test whether what they were hearing from Chalabi was true.

The idea was that if the United States was able to intercept such transmissions, Americans would react by going to the weapons site. They intercepted the message […] but did not take the bait by going to the weapons site.

Umm, so after using the codes to discuss how they suspected the codes were compromised, the Iranians try to bait the US to check if the codes were really compromised? But if the codes were compromised, the US would know that Iran knows, thanks to the Iranian spy’s first mistake. The US would know it’s a bait. And Iranian intelligence surely understands this.

But of course, the real kicker is this:

Chalabi reportedly told the Iranian he had gotten the information from an American who had been drunk.

Note to US intelligence: make sure your agents have some kind of alcohol tolerance before giving them security clearance.

2 Comments

  1. Ravi:

    What confuses me is why the Iranians were using breakable codes in the first place… I mean RSA and friends are old hat at this point, aren’t they? Maybe I’m making too much out of the language choice of “broken the code” vs. “stolen the key”. Though if the situation is really “stolen the key” it points to an even bigger compromose (since, to steal keys we’d probably have to have an agent or agent(s) inside Iranian intelligence, while codebreaking could theoretically happen without penetration).

  2. Ben Adida:

    It’s very likely that the breach was one of social engineering / down-and-dirty espionage. A mathematical compromise is, as you mention, unlikely given today’s technology. Though one should also take into account human factors in technology selection: many commercial cryptographic algorithms originated in the US, and an Iranian intelligence agency might not trust American technology, thereby resorting to some home-cooked scheme with potential weaknesses.

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.