Soviets Burned By CIA Hackers?

Godlike Productions ^ | 3/28/04 | Godlike Productions

[freedom44]

The author of a new book detailing a plan to use a Trojan horse embedded in stolen software to wage economic war against the Soviet Union fired back Thursday at charges the book´s revelations are "rubbish."

Thomas C. Reed, a former secretary of the Air Force and special assistant to President Reagan, detailed the stunning story in At the Abyss: An Insider´s History of the Cold War.

According to Reed, the Reagan administration faced a choice in 1981 when it "gained access to a KGB agent in their technical intelligence directorate" and discovered that Soviet theft of American technology had been "massive."

"In essence, the Pentagon had been in an arms race with itself," Reed said in a phone interview.

Rather than arrest everyone they could to try to close the operation down and halt further espionage, CIA director William Casey and National Security Council staffer Gus Weiss cooked up a better plan: They turned into hackers.

"(Soviet agents) stole stuff, and we knew what they were going to steal," Reed said. "Every microchip they stole would run fine for 10 million cycles, and then it would go into some other mode. It wouldn´t break down, it would start delivering false signals and go to a different logic."

The most spectacular result of this hacking, according to Reed, was a massive explosion during the summer of 1982 in the controversial pipeline delivering Siberian natural gas to Western Europe.

Soviet spies stole software needed to operate the pipeline, not knowing that "it had a few lines of software added that constituted a Trojan horse," said Reed. "They checked it out, it looked fine, and ran just fine for a few months. But the Trojan horse was programmed to let it run for four or five months and then the pumps and compressors are told, ´Today is the day we are going to run a pressure test at some significantly increased pressure.´"

He continued: "We expected that the pipeline would spring leaks all the way from Siberia to Germany, but that wasn´t what happened. Instead the welds all blew apart. It was a huge explosion. The Air Force thought it was a 3-kiloton blast."

Former KGB agent Vasily Pchelintsev, who was reportedly head of the KGB office in the area of the 1982 blast, told the English-language Moscow Times in a recent interview that Reed´s account was inaccurate. "What the Americans have written is rubbish," the former agent said.

Pchelintsev said the only explosion that occurred in Siberia that year came in April, not during the summer, and was near the city of Tobolsk in the Tyumen region. A government investigation blamed the explosion -- which was not disclosed in public until after Reed´s book -- on construction violations, Pchelintsev said.

The former KGB agent added that no one was killed in the explosion, the damage was repaired within one day and the pipeline in question supplied gas locally, to the city of Chelyabinsk, not to Western Europe along the Urengoi-Uzhgorod pipeline.

"I have the greatest respect for Russian old-timers trying to piece together the shards of history," Reed responded. "I do not know Vasily Pchelintsev, and his use of the word ´rubbish´ is a little strong, but if he really was there 25 years ago, in Tyumen, he may have access to some pieces of the story.

"On the other hand, the KGB is hardly a repository of factual reporting, and the findings of any ´government commission´ from the Soviet era should be discarded prima-facie. Protection of ´state secrets´ was their mission, not truth or accuracy."

Reed acknowledged one mistake. Another former KGB agent pointed out that the technology-stealing "Directorate T" was set up in 1918, not 1970, a point Reed conceded. Lenin, he said, talked of the need to pursue Western technology "with both hands."

Reed said the details of his account had been thoroughly vetted by the CIA and approved for publication. He said several former top Reagan officials had confirmed the reliability of Weiss, his source on the story. Weiss died in November 2003.

"Weiss was awarded the Intelligence Medal of Merit by the U.S. government, the French Legion of Honor by that nation´s government," Reed said. "Weiss published the Farewell Dossier story within channels, specifically the CIA´s Studies in Intelligence in 1996. In 2000, an expanded version was published in the Journal of U.S. Intelligence Studies. During the years that followed there was every opportunity for the intelligence community to take issue with his account. To my knowledge, no one did."

Reed only learned of the pipeline explosion in recent years. At the time, he was one of many White House officials scrambling to figure out what had caused the massive explosion.

"Gus Weiss came down the hall to tell me and others, ´Don’t worry about it,´" Reed recalled. "I asked why. He said, ´Some things in the White House you don´t ask why.´"

[marron]

I obviously don't know the particulars, but this sounds somewhat bogus. Software designed to run a pipeline still needs to be programmed by the users, with actual logic and actual wiring addresses. You wouldn't be able to put such a routine in it as described in the article, because it wouldn't correspond to anything in the actual program.

What you possibly could do would be to design an abrupt failure into it. But any system would be designed to respond to a sudden failure of the software, or the chip, because things like that happen in the best of cases. So in this case the Russians would be right, even if the chip was doctored, there should have been no pipeline failure at all. If there was, there was a design error or a construction error, or more likely both. Chip failure alone wouldn't do it. The chip fails, you get an alarm, you change out the card or the program itself, and thats it.

This sounds bogus.

[jungleboy]

marron - you say "Chip failure alone wouldn't do it"

But, it wasn't chip "failure"... it was chip "design".

[NewRomeTacitus]

Not bogus. This was related in the book "Reagan's War" by Peter Schweizer. Not only did they pull this off, the Gipper destabilized the price of natural gas and made the Soviets expend vast sums of money to maintain Russia's energy needs. When he suckered them into the missile shield race, bam! Game over.

[Western Phil]

This issue has come up before. It is possible that the Soviets had defective software supplied by us, but I suspect there are other reasons for their pipeline failures.

The Siberian pipeline was to be built by American technology with Catapiller supplied equipment. The Soviets then invaded Afganistan and in addition to not participating in the Moscow Olympics, Carter nixed the pipeline work. They had to figure out how to build it themselves.

Catapiller had started to build a gigantic building north of Interstate 74 near Morton, Illinois. The iron was up, but the roof and walls never appeared. After twenty or so years they took the columns, beams and girders down and made engines with the steel.

The Soviets were quite capable of building junk pipelines without the aid of bad software.

[marron]

The person who actually did the control program for the pipeline could easily plant a few lines in the program to sabotage it. That person would probably have been a Russian, unless the programming were subcontracted to a foreign firm.

That would be possible, exactly as described.

But such software would only be "stolen" if they didn't pay the subcontractor.

But the kind of raw, unprogrammed software used to control a pipeline does nothing until someone adds the actual wiring addresses, and the actual control logic. Someone ahead of time couldn't program in a routine to do a pressure test, as described in the article, in advance of the actual program, as he couldn't know the actual addresses which would ultimately be used.

The sabotage has to be done by the software user. Or, as I suggested, you could design in a chip failure.

But in any case, pipeline design assumes chip failure, it assumes failure of the control system, because things like that happen all the time, anyway. So I could write a routine to sabotage a pipeline, but the damage would be limited if the mechanical design was done properly. In this case, it appears that the mechanical design was not properly done, or the construction was poorly done, or more likely both.

[NewRomeTacitus]

Soviet managers had difficulty in translating laboratory results to products, quality control was poor, and plants were badly organized. Cost accounting, even in the defense sector, was hopelessly inadequate. In computers and microelectronics, the Soviets trailed Western standards by more than a decade…formidable apparatus was set up for scientific espionage; the scale of this structure testified to its importance…proposed using the Farewell material to feed or play back the products sought by Line X, but these would come from our own sources and would have been ''improved," that is, designed so that on arrival in the Soviet Union they would appear genuine but would later fail…American industry helped in the preparation of items to be "marketed" to Line X. Contrived computer chips found their way into Soviet military equipment, flawed turbines were installed on a gas pipeline, and defective plans disrupted the output of chemical plants and a tractor factory.

http://www.cia.gov/csi/studies/96unclass/farewell.htm

And then there’s the Bill Safire article on it:

http://www.taipeitimes.com/News/edit/archives/2004/02/04/2003097438

[marron]

Thanks for the links. I was quibbling over details when the details were written by journalists who didn't understand quite what they were saying. In the Safire article the description is better, implying a trojan horse that would scramble data in a random fashion. That would be quite practical with no need to know the details of a future program.

And the other article describing as you mention the sale of defective hardware sounds quite believable.

I repeat for the benefit of anyone who worries about pipelines that any design assumes the failure of your control system and is designed to withstand such things. Because, as I say, these things do fail, you have to design expecting such a failure. But, as you pointed out, these were the Soviets, where people pretended to work for their masters who pretended to pay them and incompetence was institutionalized.